GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 1 Introduction 1.1 * * * * * FEATURES Four 16-Bit CMOS ADC Input Ports Programmable Closed Loop VGA Control With 6-Bit Outputs for Each ADC Input Port Provide Received Total Wide Band Power (RTWP) Measurement for the Composite Power Across Carriers With Programmable Time Window for Measurement 8 UMTS Digital Down Converter (DDC) Channels or 16 CDMA or 16 TD-SCDMA DDC Channels With Programmable 18 Bit Filter Coefficients Each DDC channel includes - Real or Complex DDC Inputs - 115 dB SFDR NCO - UMTS Mode Rx Filtering: 6 Stage CIC (m=1 or 2), Up to 40 Tap CFIR, Up to 64 Tap PFIR - CDMA Mode Rx Filtering: 6 Stage CIC (m=1 or 2), Up to 64 Tap CFIR, Up to 64 Tap PFIR - Power Measurements - Final AGC * * * 1.5V Digital Core Supply, 3.3V Digital I/O Supply 305 Ball Plastic BGA (19 mm x 19 mm) With 1.0 mm Pitch Power Dissipation: 2.5W 1.2 * * * * * * * * * * APPLICATIONS Wireless Base Station Receiver Multi-Carrier Digital Receiver UMTS (4 Carriers-1 Sector With Diversity) CDMA (8 Carriers-1 Sector With Diversity) TD-SCDMA (16 Carriers-1 Sector Without Diversity, 8 Carriers-1-Sector With Diversity) Digital Radio Receivers Wide Band Receivers Software Radios Wireless Local Loop Intelligent Antenna Systems Contents 1 2 3 4 5 Introduction ............................................... 1 1.1 FEATURES ........................................... 1 1.2 APPLICATIONS ...................................... 1 5.2 General Description ..................................... 2 RECEIVE DIGITAL SIGNAL PROCESSING ......... 2 3.1 Receive Input Interface ............................... 3 3.2 DDC Organization ................................... 15 6 GC5018 GENERAL CONTROL ....................... 44 4.1 Microprocessor Interface Control Data, Address, and Strobes ......................................... 44 4.2 Synchronization Signals ............................. 46 4.3 Interrupt Handling ................................... 48 4.4 GC5018 Programming .............................. 48 GC5018 PINS........................................... 122 5.1 7 Microprocessor Signals ............................ 126 ...................................... 5.4 Factory Test and No Connect Signals ............. 5.5 Power and Ground Signals ........................ 5.6 Digital Supply Monitoring .......................... 5.7 JTAG ............................................... SPECIFICATIONS ..................................... 6.1 ABSOLUTE MAXIMUM RATINGS................. 6.2 RECOMMENDED OPERATING CONDITIONS ... 6.3 THERMAL CHARACTERISTICS .................. 6.4 DC CHARACTERISTICS .......................... 6.5 AC TIMING CHARACATERISTICS ................ Revision History ...................................... 5.3 JTAG Signals 127 127 127 128 128 128 128 129 129 129 130 131 Digital Receive Section Signals .................... 122 Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of Texas Instruments semiconductor products and disclaimers thereto appears at the end of this document. PRODUCTION DATA information is current as of publication date. Products conform to specifications per the terms of the Texas Instruments standard warranty. Production processing does not necessarily include testing of all parameters. Copyright (c) 2005-2005, Texas Instruments Incorporated GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 2 General Description The GC5018 is a multi-channel communications signal processor that provides digital downconversion optimized for cellular base transceiver systems. The device supports UMTS, CDMA-1X and TD-SCDMA air interface cellular standards. The chip provides up to 8 UMTS digital downconverter channels (DDC), 16 CDMA DDCs or 16 TD-SCDMA DDCs. The DDC channels are independent and operate simultaneously. The GC5018 has four 16-bit inputs. Each DDC channel can be programmed to accept data from any one (or two for complex input mode) of the four input ports. dvga_b dvga_d dvga_a dvga_c I DDC0 6 16 6 6 rxin_a adcclk_a 16 Digital receive data ports 6 2 CDMA2000-1X, 2 TD-SCDMA or 1 UMTS I DDC1 rxin_b adcclk_b 2 CDMA2000-1X, 2 TD-SCDMA or 1 UMTS rxin_c adcclk_c DDC2 rxin_d adcclk_d 2 CDMA2000-1X, 2 TD-SCDMA or 1 UMTS Receive Input Interface Power Measurements and Wideband ACG tdo Q sync I DDC3 JTAG Q sync I 16 16 trst_n tck tdi tms Q sync 2 CDMA2000-1X, 2 TD-SCDMA or 1 UMTS Q sync I DDC4 2 CDMA2000-1X, 2 TD-SCDMA or 1 UMTS Q 32 Output Format Parallel or Serial rxout_X_X 8 sync rx_sync_out_X rxclk_out I DDC5 4 rx_sync a-d interrupt reset_n Control and Sync rx_sync_out 2 CDMA2000-1X, 2 TD-SCDMA or 1 UMTS Q sync I DDC6 2 CDMA2000-1X, 2 TD-SCDMA or 1 UMTS 16 6 d(15:0) a(5:0) rd_n wr_n ce_n Q sync I DDC7 rxclk 2 CDMA2000-1X, 2 TD-SCDMA or 1 UMTS Q sync Figure 2-1. Functional Block Diagram 3 RECEIVE DIGITAL SIGNAL PROCESSING The down conversion section of the GC5018 consists of the receive input interface, the rx_distribution bus, and 8 digital downconverter blocks. The purpose of the receive input interface is to accept signal data from four 16 bit input ports, measure the input signal power, control the digital VGA and to distribute the data to the DDC blocks. The input interface also has a user-controlled test generator and noise source. The rx_distribution bus distributes the four channels of signal data to each of the 8 DDC blocks. 2 General Description GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Each DDC block selects one of the four channels (or 2 for complex input data) from the rx_distribution bus and then performs downconversion tuning, programmable delay, channel filtering with decimation, power measurement, fixed gain adjust and/or automatic gain control. Each DDC block can support 1 UMTS channel, 2 CDMA channels or 2 TD-SCDMA channels. An optional mode permits stacking two DDC blocks in UMTS mode to provide double-length final pulse shaping filtering. Tuned, filtered, and decimated signal data is output in bit serial or parallel format. 3.1 Receive Input Interface dvga_a 6 18 rxin _a 16 test & noise signal generator 16 FIFO 16 dual real or single complex Power Meter to testbus dual real or single complex AGC 1 to 64 sample delay line delay_a rxin _b 16 FIFO 16 dvga_b 6 1 to 64 sample delay line dvga_c 6 delay_b 16 test & noise signal generator 18 rx_distribution bus to DDC channels 18 rxin _c 16 test & noise signal generator 16 test bus select and decimation FIFO 16 dual real or single complex Power Meter testbus sources dual real or single complex AGC 1 to 64 sample delay line delay_c rxin _d 16 dvga_d test & noise signal generator 16 FIFO 16 18 1 to 64 sample delay line 6 delay_d Figure 3-1. Receive Data Input Interface The GC5018's receive input data interface accepts data from two sources: * Signal data presented at the four 16-bit digital data input ports. * A LFSR test signal generator allows the GC5018 to be tested using a known repetitive data sequence. Signal data can be provided in binary or 2's complement form. The location of the ADC's MSB can be programmed to allow for additional AGC headroom if desired. For example, a 14-bit ADC may be connected with the MSBs aligned, or shifted down to allow the AGC additional gain range before clipping the signal. Signal data can be accepted at rates up to rxclk in UMTS mode for either 8 normal channels or 4 double length final pulse shaping filter channels. In CDMA mode the maximum input rate is rxclk for real inputs, or rxclk/2 for complex inputs. For maximum filter performance, higher clock rates generally allow longer filters. Complex signal data is input with I data driving one input port and Q data driving another. This means that there are only two signal data ports available when using complex input mode. The mapping of I and Q data onto the four input ports is programmable. RECEIVE DIGITAL SIGNAL PROCESSING 3 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Signal input data is clocked into 8-stage FIFOs using a matching external clock signal adcclk_a/b/c/d. Signal data is clocked out of the FIFO from a gated rxclk (the GC5018 receive section clock). The FIFO allows arbitrary phase relationship between adcclk_a/b/c/d and rxclk. The frequency relationship is mandated by the programmed configuration. The test and noise generator can supply test sequences or add noise to the input signal data. The test sequences, when combined with the checksum generators, are useful for initial board debug or power-on self-test. For applications that require receiver desensitization, the noise generator can add noise to input data streams. Many internal chip signals can be routed to the testbus for evaluation and debug purposes. When the testbus is enabled, the rxin_c and rxin_d ports are driven as digital outputs. Each of the four outputs to the DDC channels includes a 1 to 64 sample delay line. PROGRAMMING VARIABLE DESCRIPTION ssel_ddc(2:0) Selects the sync source for the DDC data input mux and mixer. This sets the sync source for DDC input clock generation and synchronization for all DDC channels. offset_bin_X Selects offset binary input when set, 2's complement input when cleared. X={a,b,c,d} msb_pos_X(2:0) Identifies the connection location of the ADC's MSB. Programmed values of {0..7} corresponds to msb at {rxin_x_15.. rxin_x_8}. X={a,b,c,d} 3.1.1 Receive FIFO The receive FIFO consists of an 8 stage memory and 2 counters generating the input write pointer and output read pointer. When the FIFO receives a sync signal, the input and output pointers are initialized with a write to read pointer offset of four samples. Input samples from rxin_X (writes) are clocked with the adcclk_X input clock rising edges, and the input pointer advances on each clock rising edge. Output samples (reads) and the output pointer are clocked with the rxclk input signal rising edges, divided by the programmed sample rate loaded into the rate_sel(1:0) control register. PROGRAMMING VARIABLE DESCRIPTION adc_fifo_bypass When set, bypasses the input FIFOs and input data is latched directly using the rxclk. When cleared, input data is latched using the adcclk_a/b/c/d inputs. ssel_adc_fifo(2:0) Selects the sync source for the FIFO state machines. This sync signal initializes the FIFO input and output pointers. rate_sel(1:0) This selects the FIFO input and output rate; {rxclk, rxclk/2, rxclk/4 or rxclk/8 }. For example, with rxclk at 153.6MHz, set rate_sel to 0, 1, 2 or 3 respectively for adcclk_a/b/c/d 153.6, 76.8, 38.4 or 19.2MHz. adc_fifo_strap_ab When set, the rxin_a and rxin_b FIFO input and output pointers are synchronized to support complex input signals. adc_fifo_strap_cd When set, the rxin_c and rxin_d FIFO input and output pointers are synchronized to support complex input signals. 4 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 3.1.2 Receive Input Power Meters from rxin_a FIFO output pmeter_iq0 from rxin_b FIFO output pmeter_iq1 from rxin_c FIFO output pmeter_iq2 from rxin_d FIFO output pmeter_iq3 I power meter 0 power meter 0 results power meter 1 power meter 1 results power meter 2 power meter 2 results power meter 3 power meter 3 results Q I Q I Q I Q Figure 3-2. Receive Input Power Meters Four Receive Input RMS power meters are provided. For real inputs, the four power meters can be used to measure the RMS power of the combined carriers in each of the four input signals (the Q input is held at zero). For complex inputs, two power meters can be use to measure the combined complex power and two can be disabled. RECEIVE DIGITAL SIGNAL PROCESSING 5 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 32 16 I 33 58-bit Integrator 58-bit Register 32 16 Q transfer clear 9-bit sync delay counter sync RMS power 21-bit 21-bit interval counter integration counter 9 21 21 delay (in 8 sample interval (in 8 sample integration (in 8 sample increments) increments) increments) Figure 3-3. Detailed Functionality of Receive Input Power Meter sync delay integration time integration time integration time interval time sync event integration start integration start integration start Figure 3-4. Receive Input Power Meter Timing Power is calculated by squaring each 16 bit I (I and Q for complex inputs) sample, summing, and then integrating the summed-squared results into a 58 bit accumulator over a programmable integration period. The integration period is programmed into the 21 bit counter, in 8 sample increments. The power read is: power = [ (I2) x (Nx8 + 1) ] for real inputs where N is the integration count. power = [ (I2 + Q2)x (Nx8 + 1) ] for complex inputs where N is the integration count. A programmable 21 bit interval counter sets the power measurement interval (how often power will be measured) in 8 sample increments. A measurement integration period is started at the beginning of each interval period. The process begins with a sync event starting the 9 bit delay counter. After (8xsync_delay + 2) samples, the integration interval is started. Integration continues until the integration count is met, at which point the 58 bit integrator results are transferred to the read only register and an interrupt is generated. A new measurement period will start at the end of the interval period. NOTE Each of the four composite RMS power meter blocks has its own delay sync, interval, and integration period counters, as well as separate sync source registers. The 21-bit counters in 8 sample increments allow up to 104.8mS interval times at 160MHz clock. 6 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 PROGRAMMING VARIABLE DESCRIPTION recv_pmeterX (57:0) 58 bit power measurement result. X= {0,1,2,3}. recv_pmeterX_sqr_sum(20:0) 21 bit integration (square and sum) period. X= {0,1,2,3}. recv_pmeterX_sync_delay(8:0) Power meter delay sync period. X= {0,1,2,3}. recv_pmeterX_strt_intrvl(20:0) 21 bit measurement interval. X= {0,1,2,3}. The strt_intrvl value must be greater than the sqr_sum value. ssel_recv_pmeter_X(2:0) Sync source. X= {0,1,2,3}. pmeterX_iq Selects complex power measurement input mode when set. X= {0,1,2,3}. recv_pmeterX_ena Enables power meter when set. X= {0,1,2,3}. 3.1.3 Receive Input AGC (RAGC) Input signals from the ADCs can be used to create a front end composite AGC loop when combined with a digitally controlled variable gain amplifier (DVGA) connected before the ADCs. The AGC system operates by integrating the square of the ADC samples over a programmable interval and applying a table driven error signal to a loop integrator based on the squared integration output. The error table maps the signal power to a user programmed error value. The loop integrator output is used to drive map tables to control the DVGA output pins and a gain adjustment multiplier. Fast updates can be enabled if desired, to cause the loop integrator to quickly adjust to interfering signals. The ADC input signals can also be passed through a high pass filter to remove DC offset before squaring the input. The programmable error table, integrator mapping tables, and clip thresholds, when combined with the user programmable interval timers provide a highly flexible AGC function. integrate and dump signal power measurement enable corner Samples 16 from ADC FIFO Filter Highpass Filter 55 acc_shift 5 31 X2 7 7 limit {127..0} + 0 1 shift limit & limit acc_offset - 6 {127..0} 128w x 8b ram Map 8 Error Table Map Table update err_shift 5 sd_thresh signal detect mode controls Signal Level Detect error shift loop accumulator no_signal freeze control register bit freeze from sync source clear control register bit clear sync source 16 64w x 22b RAM Map 6 MSBs DVGA Table 6 Map Table Gain Map Gain Table 16 Map Table to DVGA pins 32 Clip Detect Mag clip_error 16 delay adjust 16 5 Delay to DDC channels clip_hi_thresh 16 clip_low_thresh clip detect controls sync update sync delay update interval Figure 3-5. Receive Input AGC RECEIVE DIGITAL SIGNAL PROCESSING 7 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 The AGC measurement interval timer is a 24-bit timer initialized by a sync after a programmable 8-bit delay. During the integration interval, the squared input signal is shifted by the programmed value and accumulated. At the end of the interval time, an update pulse is generated, and the selected 7 bits of the 55-bit accumulated power is upper limit checked and transferred to the power holding register. A programmable offset is applied, and the following limit check produces a 7 bit address value for the error map table RAM. The user programmable error map table and following gain shift setting are used to determine the loop error signal to be added to the 32-bit AGC loop accumulator. The error value is only added to the loop accumulator once per update. The loop accumulator upper 6 MSBs are used as the address for the programmable DVGA map table and gain map table. The gain map table address can be delayed from 0 to 31 clock cycles to align DVGA changes to signal level changes at the output of the AGC. The AGC includes four sources for freezing the loop and holding the loop accumulator constant. A general sync source can be used to directly control the freeze; when the selected sync source is high, the AGC will be held, and when low, the AGC will operate. A control register bit freezes the AGC in the same fashion; when the bit is set, the AGC is held, and when cleared, the AGC will operate. A signal level detector is provided that can be used to automatically freeze the AGC loop in the event of input signal loss. A programmable signal detection threshold value, number of samples below the signal detection threshold, and window timer are used to determine when no signal is present. Finally, a programmable number of AGC updates after sync can be programmed, and the AGC will he held until the next sync event. Freeze holds the loop accumulator constant, the integrate and dump accumulator constant and the interval timer constant. When freeze is released, the interval timer will resume counting. A sync event will always reinitialize the integrate and dump interval timer, and terminate the pending update to the loop accumulator from the current integrate and dump measurement interval. For example, if a sync event occurs during an integrate and dump interval, that interval will be terminated without updating the loop, and the integrate and dump accumulator will be cleared. After the programmed sync delay, a new interval will start. The AGC includes a dual threshold clip detect function, using two programmable 16-bit thresholds and programmable counters. The clip detector will cause immediate loop accumulator updates while the clip event is active. The 16-bit clip error value is aligned at the MSBs of the loop accumulator. Clip events are qualified when a programmed number of samples are above the high clip threshold during the programmable clip window time. For example, a clip event can be defined as 8 samples above the clip high threshold in a 256 sample window; the clip high threshold, the number of samples above the high clip threshold and the sample window time are programmable. Once the clip event has occurred, the clip duration is controlled by the clip low threshold value, clip low samples value and clip low timer. The clip event is cleared when the number of samples below the low clip threshold exceeds the programmed value within the clip low timer window. The clip low threshold, number of clip low samples and the clip low window timer are programmable. The AGC blocks can be paired together, rxin_a with rxin_b, and rxin_c with rxin_d, to produce a complex input AGC mode. The clip detector output from the rxin_b/d AGCs is logically OR'ed with the rxin_a/c clip detect outputs. The squared input function before the integrate and dump and signal level detector is replaced with a I2 + Q2 power calculation. The accumulator MSBs from the rxin_a/c AGCs are connected to the rxin_c/d DVGA map table and gain map table inputs. This arrangement allows the AGCs to operate in a direct conversion receiver system by controlling the I2 + Q2 complex signal level. The highpass filter is a 32 bit accumulator followed by an adjustable shift to control the corner frequency, a subtractor to remove the accumulated offset and a final limiter to produce a 16 bit result. The highpass filter function is enabled by setting hp_ena; clearing hp_ena holds the accumulator reset. 8 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 32 Samples from ADC FIFO - 16 hp_corner 3 17 shift & limit + 16 hp_ena 17 16 limit Samples to X 2 block Figure 3-6. High Pass Filter in Receive Input AGC PROGRAMMING VARIABLE DESCRIPTION ragc_bypass_X Bypasses the entire receive AGC circuit when set. X = {0,1,2,3} hp_ena_X Enables high pass filter when set hp_corner_X(2:0) Adjusts the corner frequency of the high pass filter integ_interval_X(23:0) Integrate and dump signal power measurement interval in samples. acc_shift_X(4:0) Shift down amount following the integrate and dump accumulator. acc_offset_X(5:0) Offset value applied to the shifted integrate and dump output. ragc_sync_delay_X(7:0) AGC sync delay interval, from 1 to 256 samples. ssel_ragc_interval_X(2:0) Sync source selection for the interval timer. ssel_ragc_freeze_X(2:0) Sync source selection for AGC freeze ssel_ragc_clear_X(2:0) Sync source selection for the AGC loop accumulator clear ragc_freeze_X Register bit to freeze the AGC when set ragc_clear_X Register bit to clear the AGC accumulator when set ragc_update_X(7:0) Sets the number of updates per sync event, after which no further updates will occur until the next sync event. Program to 0x00 to continually update. sd_ena_X Enables freezing the AGC with the signal detector when set sd_thresh_X(15:0) Signal detection threshold for AGC channel X. This 16 bit word is lined up with bits 23 down to 8 of the square output. The smallest signal level is that can be programmed is therefore 16 LSBs on the ADC input, and the largest is 4095 LSBs at the ADC input. sd_samples_X(15:0) The number of samples below the signal detect threshold within the signal detect sample timer window required to freeze on the AGC. sd _timer_X(15:0) Window timer to qualify signal detection. clip_hi_thresh_X(15:0) Clip detector high threshold clip_lo_thresh_X(15:0) Clip detector low threshold clip_hi_samples_X(7:0) A clip event is detected when this number of samples above the clip high threshold within the clip high sample timer window exceeds this value. clip_lo_samples_X(7:0) A clip event ends when this number of samples below the clip low threshold within the clip low sample timer window exceeds this value. clip_hi_timer_X(15:0) Window timer to qualify clip events. clip_lo_timer_X(15:0) Window timer to determine when the clip event ends. clip_error_X(15:0) Error signal applied to the AGC accumulator when a clip event is active. This data is MSB aligned, and therefore can cause immediate changes to the accumulator. ragc_error_map_X 128w x 8b memory holding the log to error look up table. dvga_map_X 64w x 6b memory holding the accumulator to DVGA look up table gain_map_X 64w x 16b memory holding the accumulator to GAIN look up table (256 decibels is unity gain). delay_adj_X(4:0) Delay between DVGA output updates and gain map updates to compensate for ADC pipeline delays, etc. err_shift_X(4:0) Error map table output shift up before adding to loop accumulator RECEIVE DIGITAL SIGNAL PROCESSING 9 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 PROGRAMMING VARIABLE DESCRIPTION complex_01 Enables complex AGC mode on inputs rxin_a and rxin_b when set complex_23 Enables complex AGC mode on inputs rxin_c and rxin_d when set ragc_accum_X(31:0) 32-bit read only register holding the current contents of the loop accumulator. tristate(10:7) 3-state controls for the dvga_d/c/b/a output pins; pins are in tristate when the 3-state bits are set. ragc_mpu_ram_read When set, the receive AGC map rams are readable via the MPU control interface. The GC5018 signal path is not operational when this bit is set, it is intended for debug purposes only. 3.1.4 Test and Noise Signal Generator The test and noise generator can generate test signals to replace the rxin_a/b/c/d inputs as a tool for debug, evaluation and self test. Checksum generators included in the individual DDC channels at the outputs can be used in conjunction with the noise generator and the internal sync timer block to create the built in self test function. The test and noise signal source included in this block is a 23-bit linear feedback shift register (LFSR) with a fixed polynomial and fixed initialization state. A sync input is required to initialize the LFSR, and the sync source is connected to the ddc_counter output signal. sync LFSR lfsr(22:0) adcclk_X initialized on sync event - each of the four generators has a different seed 22 5 0 Figure 3-7. Noise Signal Generator Receive Input Port 10 LFSR Seed Value, MSB to LSB rxin_a 100 0000 0000 0000 0001 0000 (0x400010) rxin_b 010 0110 1110 0110 1100 1110 (0x26E6CE) rxin_c 110 1110 1010 0010 1001 1000 (0x6EA298) rxin_d 000 1011 0001 1110 1011 0111 (0x0B1EB7) RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 The 23-bit LFSR output signal if used to create a 16-bit "dout(15:0)" test signal using XOR combinations of the LFSR bits. lfsr(22) lfsr(20) lfsr(19) lfsr(22) lfsr(16) dout(15) lfsr(18) lfsr(22) lfsr(15) dout(14) lfsr(17) lfsr(22) lfsr(14) dout(13) lfsr(16) lfsr(22) lfsr(13) dout(12) lfsr(15) lfsr(22) lfsr(12) dout(11) lfsr(14) lfsr(22) lfsr(11) dout(10) lfsr(13) lfsr(12) lfsr(11) lfsr(10) dout(9) dout(8) dout(7) dout(6) dout(5) dout(4) dout(3) dout(2) dout(1) dout(0) Figure 3-8. Mapping of LFSR Values to Output Bits To enable the test signal generator, the slf_tst_ena control bit is set. The rxin_a/b/c/d signals will be then replaced by the four generator output streams. To use this test signal generator as a signal source for self test, the user must also set the adc_fifo_bypass control bit. Setting the adc_fifo_bypass control bit causes the adcclk_a/b/c/d input clocks to be internally replaced with rxclk/N, where N is as programmed with the rate_sel(1:0) control bits to {1,2,4 or 8}. The test signal generators can also output a programmable constant value. All four test signal generators output the same programmable constant value. RECEIVE DIGITAL SIGNAL PROCESSING 11 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 16 rxin_a 16 Test and Noise Generator sync 16 data to FIFO for rxin_a 16 rxin_b 16 Test and Noise Generator 16 data to FIFO for rxin_b 16 rxin_c 16 Test and Noise Generator 16 data to FIFO for rxin_c 16 rxin_d 16 Test and Noise Generator rduz_sens_ena 16 data to FIFO for rxin_d slf_tst_ena Figure 3-9. Block diagram of Noise Generator Input Options The LFSR circuits can also be used to add noise to the rxin_a/b/c/d input signals by setting the rduz_sens_ena control register bit. The magnitude of the noise added can be adjusted by programming the nz_pwr_mask(15:0) control register. In the figure below, X = {a,b,c or d}. 16 16 rxin _X(15:0) lfsr (15:0) nz_pwr_mask(15:0) to FIFO for rxin _X 16 16 16 16 XORs 16 ANDs lfsr(17) lfsr(16) rduz_sens_ena Figure 3-10. Detail Circuit for Adding Noise Generator Signal to rxin Signal 12 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 PROGRAMMING VARIABLE DESCRIPTION slf_tst_ena When set, the test signal generators replace the rxin_a/b/c/d input signals with internally generated psuedo random sequences. The fifo_bypass bit must be set when this bit is set. rduz_sens_ena Enables the LFSR, adding noise to the ADC input data when set. nz_pwr_mask(15:0) Selects the power of the noise added to the ADC input data. adc_fifo_bypass When set, the FIFO is essentially bypassed, and the adcclk_a/b/c/d clock input ports are ignored. ddc_counter(31:0) 32 bit general purpose counter interval ddc_counter_width(7:0) 8 bit general purpose counter timeout width pulse ssel_ddc_counter(2:0) Sync source selection for the general purpose counter self_test_constant(17:0) 18-bit self test constant value applied to all 4 rxin_a/b/c/d inputs when self_test_const_ena is set. self_test_const_ena Enables the self test constant value for rxin_a/b/c/d 3.1.5 Sample Delay Lines The four sample delay line blocks each consist of a 64 register memory and a state machine. The state machine uses a counter to control the write (input) pointer, and the programmed read offset register data to create the read (output) pointer. Programming larger read offset register values increases the effective delay at a resolution equal to the sample rate. The read offset registers, delay_line_X, are double buffered. Writes to these registers may occur anytime, but the actual values used by the circuit will not be updated until a delay line sync event occurs. PROGRAMMING VARIABLE DESCRIPTION delay_line_X(5:0) Read offset into the 64 element memory for each delay line. X= {0,1,2,3}. ssel_delay_line_X(2:0) Selects the sync source used to update the double buffered delay line register. 3.1.6 Test Bus When the test bus is enabled, the rxin_c(15:0) and rxin_d(15:0) ports become outputs, and the dvga_c and dvga_d pins are combined with these pins to allow 36 bit wide signals from the DDC channels and the receive input interface to be multiplexed to this test output port. Many of these sources can be decimated to reduce the output sample rates. RECEIVE DIGITAL SIGNAL PROCESSING 13 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 DDC0 MUX ddc_tst_sel(5:0) zeros pfiroutput cfiroutput tadjchannel A tadjchannel B ncosin ncocos cicoutput mixer i * cos & i * sin mixer q * cos & q * sin ddc mux channel A ddc mux channel B MUX DECIMATE (35:20) tst_select(3:0) tst_decim17 (19:18) tst_decim_delay (17:2) (1:0) tst_clk tst_aflag tst_sync rxin_d(15:0) dvga_c(3:2) rxin_c(15:0) dvga_c(5:4) dvga_c(1) dvga_d(5) dvga_c(0) DDC1 sync DDC2 DDC3 DDC4 DDC5 DDC6 DDC7 Receive Interface rxin_a & rxin_b FIFO outputs Figure 3-11. Test Bus Output Circuit Showing Options for Selecting Signal PROGRAMMING VARIABLE DESCRIPTION ssel_tst_decim(2:0) Selects the sync source for the testbus decimator tst_decim_delay(3:0) Sets the testbus decimator delay from sync tst_decim17 When set the decimation factor of the test bus output block is 17X. When cleared, the decimation factor is 1X (no decimation). tst_on Enables the test bus; rxin_c(15:0) and rxin_d(15:0) are changed from inputs to outputs, dvga_c(5:0) and dvga_d(5) are used as part of the test bus. tst_select(3:0) Selects the source block for the testbus output; DDC0-7 or Receive Interface. ddc_tst_sel(5:0) Selects the signal to be output from the DDC block tst_rate_sel(4:0) Sets the testbus output clock tst_clk period to (tst_rate_sel + 1) rxclk cycles. 14 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 3.2 DDC Organization 18 18 18 18 4 to 2 (complex) or 4 to 1 (real) switch CDMA DDC A Output Interface or 1 UMTS DDC 4 to 2 (complex) or CDMA DDC B 4 to 1 (real) switch DDC0 DDC1 DDC2 DDC3 DDC4 DDC5 DDC6 DDC7 Figure 3-12. DDC Organization for Single Length Filter Mode The GC5018 provides downconversion for up to 8 UMTS receive channels, 16 CDMA2000 receive channels or 16 TD-SCDMA receive channels. Downconversion channels are organized into 8 DDC blocks. Each individual DDC block provides 2 CDMA2000 or 2 TD-SCDMA DDC channels, A and B, or 1 UMTS channel. Both CDMA DDC channels in a DDC block can be independently tuned, though they would likely be used as diversity pairs and tuned to the same frequency. Filter coefficients are shared between the two CDMA DDC channels within a block. Two adjacent DDC blocks (for example, DDC0 and DDC1) can be strapped together to form a single UMTS DDC channel with double-length final pulse shaping filtering. The GC5018 can therefore provide 4 UMTS DDC channels with double-length final PFIR filtering as shown in the following diagram. RECEIVE DIGITAL SIGNAL PROCESSING 15 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4 UMTS DDCs with up to 128 tap PFIR 18 18 18 18 4 to 2 (complex) or CDMA DDC A 4 to 1 (real) switch Output Interface 4 to 2 (complex) or CDMA DDC B 4 to 1 (real) switch DDC0 DDC1 4 to 2 (complex) or CDMA DDC A 4 to 1 (real) switch Output Interface or 1 UMTS DDC 4 to 2 (complex) or CDMA DDC B 4 to 1 (real) switch DDC0 plus DDC1 DDC2 plus DDC3 DDC4 plus DDC5 DDC6 plus DDC7 Figure 3-13. DDC Organization for Double Length Filter Mode PROGRAMMING VARIABLE DESCRIPTION ddc_ena When set, turns on the DDC. cdma_mode When set, puts the DDC block in dual channel CDMA mode. gbl_ddc_write When set, all subsequent programming (writes only) for DDC0 and DDC1 is also written to DDC2/4/6 and DDC3/5/7. 3.2.1 Downconverter Function Blocks from rx_distribution bus 18 18 18 18 Delay Adjust 4 to 2 Select Frequency Phase 32 16 NCO CFIR Filter Dec by 2 Zero Pad Six Stage CIC Filter Dec 4 to 32 PFIR Filter Dec by 1 Checksum Generator AGC RMS Power Measure Figure 3-14. DDC Functional Block Diagram 16 RECEIVE DIGITAL SIGNAL PROCESSING Serial Interface serial I, Q up to 18 (25-bits with AGC disabled) parallel I, Q GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Each GC5018 downconversion block can process two CDMA carriers or a single UMTS carrier. Signal data is selected from one of four ports for real inputs, or two of four ports for complex inputs. Data from the selected port(s) is multiplied with a complex, programmable numerically controlled oscillator (NCO) which tunes the signal of interest to baseband. The delay adjust and zero pad blocks permits adjustment of the delay in the end-to-end channel. Zero padding interpolates the signal to the rxclk rate. Filtering consists of a six stage CIC filter which decimates the tuned data by a factor from 4 to 32, a compensating FIR filter (CFIR) which decimates by a factor of two, followed by a programmable FIR filter (PFIR) which does not decimate. The output interface block can be programmed to decimate by 2 if desired. The RMS power meter measures the power within the channel's bandwidth. The AGC automatically drives the gain and keeps the magnitude of the signal at a user-specified level. This allows fewer bits to represent the signal. The serial output interface formats and rounds the output data. Each of the above blocks is described in greater detail in the following sections. 3.2.2 DDC Mixer from rx_distribution bus 18 18 18 18 4 to 2 Select Demux and Round 18 18 18 18 18 18 IA QA IB QB to channel delay 20 20 cos sin mixer_gain from NCO Figure 3-15. Mixer Functional Block Diagram The receive mixer translates the input (from one of the input signal sources) to baseband where subsequent filtering is performed to isolate the signal of interest. The mixer is a complex multiplier that accepts 18 bit I and 18 bit Q signal data from the receive input interface and 20 bit Sine and Cosine sequences from the NCO. The NCO generates a mixing frequency (sometimes referred to as a local oscillator, or LO) specified by the user so that the desired signal of interest is tuned to 0 Hertz. A DDC channel can support one UMTS signal directly, or two CDMA channels at half the input rate. When in CDMA mode, each channel may be set independently; the path selection and the mixer tuning and phase. The mixer output produces two complex streams; one representing the signal path for the A-side DDC, the other the B-side. Each of these streams drives a channel delay and zero pad block. The maximum input rate for UMTS is rxclk for either real or complex input data. The maximum input rate in CDMA mode with real inputs is rxclk (remix_only is set, see below). The maximum input rate in CDMA mode with complex inputs is rxclk/2 due to sharing of multiplier resources. RECEIVE DIGITAL SIGNAL PROCESSING 17 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 PROGRAMMING VARIABLE DESCRIPTION ddcmux_sel_a(3:0) Programs the I and Q complex input data routing onto two of the four input ports for stream A of CDMA DDC ddcmux_sel_b(3:0) Programs the I and Q complex input data routing onto two of the four input ports for stream B of CDMA DDC remix_only For CDMA mode only, set this bit for real input data at the rxclk rate. For complex inputs in CDMA mode, the maximum input data rate is rxclk/2, and this bit must be cleared. For CDMA mode with real inputs at the rxclk/2 rate or lower, this bit must be cleared zero_qsample When set, the Q samples used by the mixer are always zero. This bit should be set for real only inputs in UMTS mode, or real only inputs in CDMA mode when the input sample rate is rxclk/2 or lower. ch_rate_sel(1:0) Specifies the input channel data rate (rxclk, rxclk/2, rxclk/4, or rxclk/8 MSPS). mixer_gain When asserted, adds 6dB of gain in the mixer. This gain is highly recommended. 3.2.3 DDC Number Controlled Oscillator (NCO) Frequency Sync Frequency Word 32 Reg 32 Reg 32 23 23 Clear Aligned to top 32 bits Zero Phase Sync Phase Offset Sync 5 sin/cos table 20 cos 20 sin Aligned to bottom 5 bits Dither Generator Phase Offset 16 Reg 16 Dither Sync Figure 3-16. Detailed NCO Circuit The NCO is a digital complex oscillator that is used to translate (or downconvert) an input signal of interest to baseband. The block produces programmable complex digital sinusoids by accumulating a frequency word which is programmed by the user. The output of the accumulator is a phase argument that indexes into a sin/cos ROM table which produces the complex sinusoid. A phase offset can be added prior to indexing if desired for channel calibration purposes. This will change the sin/cos phase with respect to other channels' NCOs. A 5-bit dither generator is provided and generates a small level of digital pseudo-noise that is added to the phase argument below the bottom bits and is useful for reducing NCO spurious outputs. This dither generation is enabled by setting the dither_ena bit; the magnitude of the dither can be reduced by setting one or both of the dither_mask bits DITHER PROGRAMMING VARIABLE DESCRIPTION dither_ena When set turns dither on. Clearing turns dither off. dither_mask(1:0) Masks the MSB and MSB-1 dither bits, respectively, when set. The NCO spurious levels are better than -115 dBc. Added phase dither randomizes the periodic nature of the phase accumulation process and reduces low-level spurious energy. For some frequencies (N x Fs/24, where N = {1,2, . . . 23}) dither is ineffective - in these cases an initial phase of 4 reduces NCO spurs. The figures below show the spur level performance of the NCO without dither, with dither, and with a phase offset value. 18 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 a) Worst Case Spectrum Without Dither b) Spectrum With Dither (Tuned to Same Frequency Figure 3-17. NCO SFDR - Without Dither Figure 3-18. NCO SFDR - With Dither Frequency - Fs Frequency - Fs a) Plot Without Dither or Phase Initialization Figure 3-19. NCO Spectra (0 and Fs/4) - Without Dither or Phase Initialization b) Plot With Dither or Phase Initialization Figure 3-20. NCO Spectra (0 and Fs/4) - With Dither or Phase Initialization The tuning frequency is specified as a 32 bit Frequency Word and is programmed as two sequential 16 bit words over the control port. The NCO frequency resolution is Fclk/ 232. As an example, at an input clock rate of 61.44 MHz, the frequency step size would be approximately 14 milli-Hertz. The Frequency Word is determined by the formula: Frequency Word (in decimal)= 232 x Tuning Frequency / Fclk Note that frequency tuning words can be positive or negative valued. Specifying a positive frequency value translates complex negative frequencies upwards towards 0 Hertz. Specifying a negative tuning frequency translates complex positive frequencies downwards towards 0 Hertz. FREQUENCY PROGRAMMING VARIABLE DESCRIPTION phase_add_a(31:0) 32 bit tuning frequency word for the A-side DDC when in CDMA mode. Also for UMTS mode. phase_add_b(31:0) 32 bit tuning frequency word for the B-side DDC when in CDMA mode. Not used in UMTS mode. Each of the 16 CDMA DDC channels can be loaded with unique frequency words. The phase of the NCO's Sin/Cos output can be adjusted relative to the phase of other channel NCOs by specifying a Phase Offset. The Phase Offset is programmed as a 16 bit word, yielding a step size of about 5.5 m. The Phase Offset Word is determined by the formula: Phase Offset Word = 216 x Offset_in_Degrees / 360 or, Phase Offset Word = 216 x Offset_in_Radians / 2 RECEIVE DIGITAL SIGNAL PROCESSING 19 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 PHASE PROGRAMMING VARIABLE DESCRIPTION phase_offset_a(15:0) 16 bit phase offset word for the A-side DDC when in CDMA mode. Also for UMTS mode. phase_offset_b(15:0) 16 bit phase offset word for the B-side DDC when in CDMA mode. Not used in UMTS mode. Each of the 16 CDMA DDC channels can be loaded with unique phase offset words. Various synchronization signals are available which are used to synchronize the NCOs of all channels with respect to each other. Frequency Sync and Phase Offset Sync determine when frequency and phase offset changes occur. For example, generating a Frequency Sync after programming the two frequency words will cause the NCO (or multiple NCOs) to change frequency at that time, rather than after each of the three frequency words is programmed over the control bus. The Zero Phase Sync signal is used to force the sine and cosine oscillators to their zero phase state. Dither Sync can be used to synchronize the dither generators of multiple NCOs. The NCOs used in the transmit section are identical to what is described for the receive section. Note that there is one set of sync's provided for each DDC. When one DDC is used to process two CDMA signals, the syncs are shared between them. SYNC PROGRAMMING VARIABLE DESCRIPTION ssel_nco(2:0) Sync source for NCO accumulator reset ssel_dither(2:0) Sync source for NCO dither reset ssel_freq(2:0) Sync source for NCO frequency register loading ssel_phase(2:0) Sync source for NCO phase register loading 3.2.4 DDC Filtering and Decimation The purpose of the receive filter chain is to isolate the signal of interest (and reject all others) that has been previously translated to baseband via the mixer and NCO. The overall decimation through the chain needs to be considered. The goal, generally, is to output the isolated signal at a rate that is twice (2X) the signal's chip rate. For UMTS this would be 7.68 MSPS and for CDMA the output rate should be 2.4576 MSPS. TD-SCDMA systems require the output rate be the chip rate of 1.28 MSPS. The output interface is programmed to decimate by 2 for the TD-SCDMA case. Receive filtering and decimation is performed in several stages: * Zero padding to interpolate the input sample rate (if needed) up to the rxclk rate * High rate decimation (4 to 32) using a six stage cascade-integrate-comb filter (CIC) * Decimate by two compensation filtering using the programmable compensating FIR filter (CFIR) * Pulse-shape filtering via the programmable FIR filter (PFIR) with no decimation * Output interface, serial or parallel format, with no decimation or decimate by 2 From Mixer Delay Adjust Zero Pad Interp by {1,2,4,8} Six Stage CIC Filter Dec by {4 -32} CFIR Filter Dec by 2 PFIR Filter no decimation Output Interface Dec by {1,2} Figure 3-21. DDC Filtering Functional Block Diagram The table below contains some examples of decimation and sample rates at the output of each block for UMTS, CDMA and TD-SCDMA standards at various supported input samples. For each example, the differential ADC clocks are provided to the GC5018 at the input sample rate and the rxclk is provided at the zero pad output rate. 20 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Table 3-1. Examples of Decimation and Sample Rates (1) Input Sample Rate (MSPS) Zeros Added rxclk(MHz) and Zero Pad Output Rate (MSPS) CIC Decimation CIC Output Rate (MSPS) CFIR Decimation CFIR Output Rate (MSPS) PFIR Decimation PFIR Output Rate (MSPS) Output Decimation UMTS 122.88 0 122.88 8 15.36 2 7.68 1 7.68 1 UMTS 92.16 0 92.16 6 15.36 2 7.68 1 7.68 1 UMTS 76.80 1 153.6 10 15.36 2 7.68 1 7.68 1 UMTS 61.44 1 122.88 8 15.36 2 7.68 1 7.68 1 CDMA 122.88 0 122.88 25 4.9152 2 2.4576 1 2.4576 1 CDMA 78.6432 0 78.6432 16 4.9152 2 2.4576 1 2.4576 1 CDMA 78.6432 1 157.2864 32 4.9152 2 2.4576 1 2.4576 1 CDMA 61.44 1 122.88 25 4.9152 2 2.4576 1 2.4576 1 TD-SCDMA 92.16 0 92.16 18 5.12 2 2.56 1 2.56 2 TD-SCDMA 81.92 0 81.92 16 5.12 2 2.56 1 2.56 2 TD-SCDMA 76.80 0 76.80 15 5.12 2 2.56 1 2.56 2 TD-SCDMA 76.80 1 153.6 30 5.12 2 2.56 1 2.56 2 TD-SCDMA 61.44 1 122.88 24 5.12 2 2.56 1 2.56 2 (1) The DDC output interfaces, both serial and parallel formats, can be programmed to decimate by 2. For the TD-SCDMA examples listed above, the DDC output rate is 1.28Msps (1x chip rate). 3.2.5 DDC Channel Delay Adjust and Zero Insertion interpolation (number of zeros stuffed between samples) input rate I samples from Q Mixer read offset 18 Delay Memory I:8 slots x 18-bits Q:8 slots x 18-bits 18 3 18 18 Zero Pad 18 I 18 Q full rxclk rate samples to CIC Filter 3 sync (offset registers) sync (zero stuff moment) insert offset 3 Figure 3-22. DDC Delay and Zero Insertion Block The Receive Channel Delay Adjust function is used to add programmable delays in the channel downconvert path. Adjusting channel delay can be used to compensate for analog elements external to the GC5018 digital downconversion such as cables, splitters, analog downconverters, filters, etc. The Delay Memory block consists of an 8 register memory and a state machine. The state machine uses a counter to control the write (input) pointer, and the programmed read offset register data to create a read (output) pointer. Programming larger read offset register values increases the effective delay at a resolution equal to the input sample rate. The Zero Pad block is used in conjunction with the Delay Memory for delay adjustments. For example, with input rates of rxclk/8, the Zero Pad block interpolates the input data to rxclk by inserting 7 zeros. The Zero Pad's sync insert offset 3-bit control specifies when the zeros are inserted relative to the Sync signal. This permits a fine adjustment at the rxclk resolution. RECEIVE DIGITAL SIGNAL PROCESSING 21 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 The read offset register, tadf_offset_course_a/b, and the insert offset register, tadj_offset_fine_a/b, are double buffered. Writes to these registers may occur anytime, but the actual values used by the circuit will not be updated until a register sync PROGRAMMING VARIABLE DESCRIPTION tadj_offset_coarse_a(2:0) Read offset into the 8 element memory for the UMTS or CDMA mode A channel DDC. tadj_offset_coarse_b(2:0) Read offset into the 8 element memory for the CDMA mode B channel DDC when in CDMA mode. tadj_offset_fine_a(2:0) Controls the zero pad (or stuff) insert offset (fine adjust) for the UMTS or CDMA mode A channel of the DDC. tadj_offset_fine_b(2:0) Controls the zero pad (or stuff) insert offset (fine adjust) for the CDMA mode B channel of the DDC when in CDMA mode. tadj_interp(2:0) The interpolation value (1, 2, 4, or 8). Same used for both the A and B channels when in CDMA mode. Selects the number of zeros to be inserted. ssel_tadj_fine(2:0) Selects the sync source for the fine time adjust zero stuff moment. Same for A and B channels when in CDMA mode. ssel_tadj_reg(2:0) Selects the sync source used to update the double buffer course and fine delay selection registers. Same for A and B channels when in CDMA mode. 3.2.6 DDC CIC Filter Shift 18 Z -1 N Z -1 Z Z -m1 Z -m 2 -1 Z Z -m 3 -1 Z Z -m 4 -1 Z Z -m 5 -1 54 24 Z -m 6 24 Decimate by 4-32 Sh ift 0-3 1 Round & L imit 18 m1, m2, m3, m4, m5, m6 = 1 or 2 Figure 3-23. DDC CIC Filter Block Diagram The CIC filter provides the first stage of filtering and large-value decimation. The filter consists of six stages and decimates over a range from 4 to 32. I data and Q data are handled separately with two CIC filters. In addition, when in CDMA mode (two CDMA channels processed within a single DDC), another pair of CIC filters handles the B-side channel. The filter response is 6x(Sin(x)/x) in character where the key attribute is that the resulting response nulls reject signal aliases from decimation. A consequence of this desirable behavior is that only a small portion of the passband can be used, less than 25% generally. This means that the CIC decimation value should be chosen so that the signal exiting the CIC filter is oversampled by at least a factor of four. The filter is equivalent to 6 stages of a FIR filter with uniform coefficients (6 combined boxcar filter stages). Each filter would be of length N if m=1, or 2N if m=2. The filter is made up of six banks of 54 bit accumulator sections followed by six banks of 24 bit subtractor sections. Each of the subtractor sections can be independently programmed with a differential delay of either one or two. A shift block follows the last integration stage and can shift the 54 bit accumulated data down by 36-rcic_shift (a programmable factor from 0 to 31 bits). 22 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 The CIC filter exhibits a droop across its frequency response. The following CFIR filter compensates for the CIC droop with a gradually rising frequency response. It is also possible to compensate for CIC droop in the PFIR filter. The gain of the receive CIC filter is: Ncic6 x 2(number of stages where M=2) x 2(-36+RCIC_SHIFT) where RCIC_SHIFT is 0 to 31. There is no rollover protection internal to the CIC or at the final round so the user must guarantee no sample exceeds full scale prior to rounding. For practical purposes this means the CIC gain can only compensate for peak gain less than one or must be less than or equal to one. A fixed gain of +12 dB at the output of the CIC can also be programmed. PROGRAMMING VARIABLE DESCRIPTION cic_decim(4:0) The CIC decimation ratio (4 to 32). The ratio is cic_decim + 1. This ratio applies to both A and B channels of the DDC block in CDMA mode. cic_scale_a(4:0) The shift value for the A channel. A value of 0 is no shift, each increment in value increases the amplitude of the shifter output by a factor of 2. cic_scale_b(4:0) The shift value for the B channel. A value of 0 is no shift, each increment in value increases the amplitude of the shifter output by a factor of 2. cic_gain_ddc When asserted, adds a gain of 12 dB at the CIC output. cic_m2_ena_a(5:0) Sets the differential delay value M for each of the CIC subtractor stages for the UMTS or CDMA mode A channel. cic_m2_ena_b (5:0) Sets the differential delay value M for each of the CIC subtractor stages for the CDMA mode B channel. cic_bypass Bypasses the CIC filter when set, for factory testing. ssel_cic(2:0) Sets syncing (1 of 8 sources) for the CIC decimation moment. 3.2.7 DDC Compensating FIR Filter (CFIR) The receive compensating FIR filter (CFIR) decimates the output of the CIC filter by a fixed factor of two. Filter coefficient size, input data size, and output data size are 18 bits. The CFIR length can be programmed. This permits "turning off" taps and saving power if shorter filters are appropriate (the CFIR power dissipation is proportional to its length). The filter is organized in two partial filter blocks, each containing a data RAM, a coefficient RAM and a dual multiplier, a common state machine and output accumulator. RECEIVE DIGITAL SIGNAL PROCESSING 23 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 MPU control interface read data write data write pointer COEF RAM 32x18 mpu ram_read COEF RAM 32x18 reg MUX complex output samples read pointer complex input samples DATA RAM 64x36 DATA RAM 64x36 read pointer write pointer output sample valid State Machine crastarttap Figure 3-24. DDC CFIR Block Diagram The maximum CFIR filter length is a function of GC5018 clock rate, output sample rate and the number of coefficient memory registers. The maximum number of taps is 64 and the minimum number is 14. Lengths between these limits can be specified in increments of 2. Subject to the above minimum and maximum values, in the general case, the number of taps available is: UMTS Mode: 2 x (rxclk / output sample rate) CDMA Mode if cic_decim is even (decimating by an odd number): 2 x (cic_decim) CDMA Mode if cic_decim is odd (decimating by an even number): 2 x (cic_decim + 1) Example CFIR filter lengths available based on mode and rxclk frequency: Mode rxclk (MHz) CIC DECIMATION cic_decim CFIR MAX LENGTH CFIR MIN LENGTH UMTS 153.60 10 9 40 14 UMTS UMTS 122.88 8 7 32 14 UMTS CDMA 157.2864 32 31 64 14 CDMA2000 CDMA 122.88 25 24 48 14 CDMA2000 CDMA 78.6432 16 15 32 14 CDMA2000 low power configuration CDMA 153.60 30 29 60 14 TD-SCDMA CDMA 81.92 16 15 32 14 TD-SCDMA CDMA 76.80 15 14 28 14 TD-SCDMA low power configuration 24 RECEIVE DIGITAL SIGNAL PROCESSING COMMENTS GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 A single set of programmed tap values are used for both the A-side and B-side DDC channels (two CDMA channels) within a single DDC block when in CDMA mode. After the CFIR filter performs the convolution, gain is applied at full precision, the signal is rounded, and then hard limited. A shifter at the output of the filter then scales the data by either 2e-19 or 2e-18. The gain through the filter is therefore: Sum(CFIR coefficients) x 2 -(18 or 19) Coefficients are organized in two groups of 32 words, each 18 bits wide. For fully utilized filters, the 64 coefficients are loaded 0 through 31 into the first RAM, and 32 through 63 into the second RAM. The 16 bit MSBs and 2 bit LSBs are written into the RAMs using different page register values. Shorter filters require the coefficients be loaded into the 2 rams equally, starting from address 0. For example, a CFIR coefficient set for a symmetric 58 tap TD-SCDMA CFIR is: Taps Coefficient Taps Coefficient 0 = 57 -13 15 = 42 -4975 1 = 56 -20 16 = 41 -4649 2 = 55 14 17 = 40 -232 3 = 54 101 18 = 39 6581 4 = 53 184 19 = 38 11266 5 = 52 133 20 = 37 8917 6 = 51 -147 21 = 36 -1957 7 = 50 -562 22 = 35 -16736 8 = 49 -768 23 = 34 -25469 9 = 48 -364 24 = 33 - 17599 10 = 47 719 25 = 32 11560 11 = 46 1905 26 = 31 56455 12 = 45 2126 27 = 30 102215 13 = 44 567 28 = 29 131071 14 = 43 -2416 The first 29 coefficients are loaded into addresses 0 through 28 in the first coefficient RAM, and the remaining 29 are loaded into addresses 0 through 28 in the second coefficient RAM. Loading the 18 bit coefficients requires 2 writes per coefficient, one for the upper 16 bits and another for the lower 2 bits. To program this coefficient set for the DDC2 CFIR, the following control microprocessor interface sequence would be used. Step Address a[5:0] Data d[15:0] Description 1 0x21 0x0480 Page register for DDC2 CFIR Coefficient RAM 0-31, LSBs. 2 0x00 0x0003 2 lower bits of coefficient 0 3 0x01 0x0000 2 lower bits of coefficient 1 4 0x02 0x0002 2 lower bits of coefficient 2 5 0x03 0x0001 2 lower bits of coefficient 3 6 0x04 0x0000 2 lower bits of coefficient 4 7 0x05 0x0001 2 lower bits of coefficient 5 8 0x06 0x0001 2 lower bits of coefficient 6 9 0x07 0x0002 2 lower bits of coefficient 7 10 0x08 0x0000 2 lower bits of coefficient 8 11 0x09 0x0000 2 lower bits of coefficient 9 12 0x0A 0x0003 2 lower bits of coefficient 10 RECEIVE DIGITAL SIGNAL PROCESSING 25 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Step 26 Address a[5:0] Data d[15:0] Description 13 0x0B 0x0001 2 lower bits of coefficient 11 14 0x0C 0x0002 2 lower bits of coefficient 12 15 0x0D 0x0003 2 lower bits of coefficient 13 16 0x0E 0x0000 2 lower bits of coefficient 14 17 0x0F 0x0001 2 lower bits of coefficient 15 18 0x10 0x0003 2 lower bits of coefficient 16 19 0x11 0x0000 2 lower bits of coefficient 17 20 0x12 0x0001 2 lower bits of coefficient 18 21 0x13 0x0002 2 lower bits of coefficient 19 22 0x14 0x0001 2 lower bits of coefficient 20 23 0x15 0x0003 2 lower bits of coefficient 21 24 0x16 0x0000 2 lower bits of coefficient 22 25 0x17 0x0003 2 lower bits of coefficient 23 26 0x18 0x0001 2 lower bits of coefficient 24 27 0x19 0x0000 2 lower bits of coefficient 25 28 0x1A 0x0003 2 lower bits of coefficient 26 29 0x1B 0x0003 2 lower bits of coefficient 27 30 0x1C 0x0003 2 lower bits of coefficient 28 31 0x1D 0x0000 2 lower bits of unused coefficient RAM location 32 0x1E 0x0000 2 lower bits of unused coefficient RAM location 33 0x1F 0x0000 2 lower bits of unused coefficient RAM location 34 0x21 0x04A0 Page register for DDC2 CFIR Coefficient RAM 32-63, LSBs. 35 0x00 0x0003 2 lower bits of coefficient 29 36 0x01 0x0003 2 lower bits of coefficient 30 37 0x02 0x0003 2 lower bits of coefficient 31 38 0x03 0x0000 2 lower bits of coefficient 32 39 0x04 0x0001 2 lower bits of coefficient 33 40 0x05 0x0003 2 lower bits of coefficient 34 41 0x06 0x0000 2 lower bits of coefficient 35 42 0x07 0x0003 2 lower bits of coefficient 36 43 0x08 0x0001 2 lower bits of coefficient 37 44 0x09 0x0002 2 lower bits of coefficient 38 45 0x0A 0x0001 2 lower bits of coefficient 39 46 0x0B 0x0000 2 lower bits of coefficient 40 47 0x0C 0x0003 2 lower bits of coefficient 41 48 0x0D 0x0001 2 lower bits of coefficient 42 49 0x0E 0x0000 2 lower bits of coefficient 43 50 0x0F 0x0003 2 lower bits of coefficient 44 51 0x10 0x0002 2 lower bits of coefficient 45 52 0x11 0x0001 2 lower bits of coefficient 46 53 0x12 0x0003 2 lower bits of coefficient 47 54 0x13 0x0000 2 lower bits of coefficient 48 55 0x14 0x0000 2 lower bits of coefficient 49 56 0x15 0x0002 2 lower bits of coefficient 50 57 0x16 0x0001 2 lower bits of coefficient 51 58 0x17 0x0001 2 lower bits of coefficient 52 59 0x18 0x0000 2 lower bits of coefficient 53 60 0x19 0x0001 2 lower bits of coefficient 54 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Step Address a[5:0] Data d[15:0] Description 61 0x1A 0x0002 2 lower bits of coefficient 55 62 0x1B 0x0000 2 lower bits of coefficient 56 63 0x1C 0x0003 2 lower bits of coefficient 57 64 0x1D 0x0000 2 lower bits of unused coefficient RAM location 65 0x1E 0x0000 2 lower bits of unused coefficient RAM location 66 0x1F 0x0000 2 lower bits of unused coefficient RAM location 67 0x21 0x04C0 Page register for DDC2 CFIR Coefficient RAM 0-31, MSBs. 68 0x00 0xFFFC Upper 16 bits of coefficient 0 69 0x01 0xFFFB Upper 16 bits of coefficient 1 70 0x02 0x0003 Upper 16 bits of coefficient 2 71 0x03 0x0019 Upper 16 bits of coefficient 3 72 0x04 0x002E Upper 16 bits of coefficient 4 73 0x05 0x0021 Upper 16 bits of coefficient 5 74 0x06 0xFFDB Upper 16 bits of coefficient 6 75 0x07 0xFF73 Upper 16 bits of coefficient 7 76 0x08 0xFF40 Upper 16 bits of coefficient 8 77 0x09 0xFFA5 Upper 16 bits of coefficient 9 78 0x0A 0x00B3 Upper 16 bits of coefficient 10 79 0x0B 0x01DC Upper 16 bits of coefficient 11 80 0x0C 0x0213 Upper 16 bits of coefficient 12 81 0x0D 0x008D Upper 16 bits of coefficient 13 82 0x0E 0xFDA4 Upper 16 bits of coefficient 14 83 0x0F 0xFB24 Upper 16 bits of coefficient 15 84 0x10 0xFB75 Upper 16 bits of coefficient 16 85 0x11 0xFFC6 Upper 16 bits of coefficient 17 86 0x12 0x066D Upper 16 bits of coefficient 18 87 0x13 0x0B00 Upper 16 bits of coefficient 19 88 0x14 0x08B5 Upper 16 bits of coefficient 20 89 0x15 0xFE16 Upper 16 bits of coefficient 21 90 0x16 0xEFA8 Upper 16 bits of coefficient 22 91 0x17 0xE720 Upper 16 bits of coefficient 23 92 0x18 0xEED0 Upper 16 bits of coefficient 24 93 0x19 0x0B4A Upper 16 bits of coefficient 25 94 0x1A 0x3721 Upper 16 bits of coefficient 26 95 0x1B 0x63D1 Upper 16 bits of coefficient 27 96 0x1C 0x7FFF Upper 16 bits of coefficient 28 97 0x1D 0x0000 Upper 16 bits of unused coefficient RAM location 98 0x1E 0x0000 Upper 16 bits of unused coefficient RAM location 99 0x1F 0x0000 Upper 16 bits of unused coefficient RAM location 100 0x21 0x04E0 Page register for DDC2 CFIR Coefficient RAM 32-63, MSBs. 101 0x00 0x7FFF Upper 16 bits of coefficient 29 102 0x01 0x63D1 Upper 16 bits of coefficient 30 103 0x02 0x3721 Upper 16 bits of coefficient 31 104 0x03 0x0B4A Upper 16 bits of coefficient 32 105 0x04 0xEED0 Upper 16 bits of coefficient 33 106 0x05 0xE720 Upper 16 bits of coefficient 34 107 0x06 0xEFA8 Upper 16 bits of coefficient 35 108 0x07 0xFE16 Upper 16 bits of coefficient 36 RECEIVE DIGITAL SIGNAL PROCESSING 27 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Step Address a[5:0] Data d[15:0] Description 109 0x08 0x08B5 Upper 16 bits of coefficient 37 110 0x09 0x0B00 Upper 16 bits of coefficient 38 111 0x0A 0x066D Upper 16 bits of coefficient 39 112 0x0B 0xFFC6 Upper 16 bits of coefficient 40 113 0x0C 0xFB75 Upper 16 bits of coefficient 41 114 0x0D 0xFB24 Upper 16 bits of coefficient 42 115 0x0E 0xFDA4 Upper 16 bits of coefficient 43 116 0x0F 0x008D Upper 16 bits of coefficient 44 117 0x10 0x0213 Upper 16 bits of coefficient 45 118 0x11 0x01DC Upper 16 bits of coefficient 46 119 0x12 0x00B3 Upper 16 bits of coefficient 47 120 0x13 0xFFA5 Upper 16 bits of coefficient 48 121 0x14 0xFF40 Upper 16 bits of coefficient 49 122 0x15 0xFF73 Upper 16 bits of coefficient 50 123 0x16 0xFFDB Upper 16 bits of coefficient 51 124 0x17 0x0021 Upper 16 bits of coefficient 52 125 0x18 0x002E Upper 16 bits of coefficient 53 126 0x19 0x0019 Upper 16 bits of coefficient 54 127 0x1A 0x0003 Upper 16 bits of coefficient 55 128 0x1B 0xFFFB Upper 16 bits of coefficient 56 129 0x1C 0xFFFC Upper 16 bits of coefficient 57 130 0x1D 0x0000 Upper 16 bits of unused coefficient RAM location 131 0x1E 0x0000 Upper 16 bits of unused coefficient RAM location 132 0x1F 0x0000 Upper 16 bits of unused coefficient RAM location 133 0x21 0x0500 Page register for DDC2 control registers 0-31 134 0x00 0x8EE0 DDC2 FIR_MODE register; cdma_mode enabled, 60 tap PFIR, 58 tap CFIR 135 0x01 0x2000 DDC2 PFIR gain = sum(taps)x2^-18 and CFIR gain = sum(taps)x2^-19 PROGRAMMING VARIABLE DESCRIPTION crastarttap_cfir(4:0) Number of DDC CFIR filter taps is 2x(crastarttap + 1) mpu_ram_read What set, the PFIR and CFIR coefficient rams are readable via the MPU control interface. The GC5018 signal path is not operational when this bit is set, it is intended for debug purposes only. cfir_gain 0 = 2e-19, 1 = 2e-18 The CFIR filter's 18 bit coefficients are loaded in two 32 word memories. Note: CFIR filter coefficients are shared between A and B channels of a DDC block in CDMA mode. 3.2.8 DDC Programmable FIR Filter (PFIR) The receive programmable FIR filter (PFIR) provides final pulse shaping of the baseband signal data. It does not perform any decimation. Filter coefficient size, input, and output data size is 18 bits. A special strapped mode can be employed for UMTS where two adjacent DDCs (2k & 2k+1, k=0 to 7) can be combined to yield a filter with twice the number of coefficients. This means the GC5018 can support 4 UMTS DDC channels with double-length filter coefficients (up to 128 taps). The filter is organized in four partial filter blocks, each containing a data RAM, a coefficient RAM and a dual multiplier, a common state machine and output accumulator. 28 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 MPU control interface read data write data write pointer Filter cell 1 cell 2 cell 3 cell 4 COEF RAM 16x18 mpu ram_read reg MUX from adjacent DDC (if double_tap="10") read pointer complex input samples from cfir (or adjacent DDC if double_tap="01") to adjacent DDC (if double_tap="10" ) DATA RAM 32x36 read pointer write pointer crastarttap complex output samples output sample valid State Machine Figure 3-25. DDC PFIR Block Diagram The PFIR length is programmable. This permits turning off taps and saving power if short filters are appropriate. The filter's output data can be shifted over a range of 0 to 7 bits where it is then rounded and hard limited to 18 bits. The shift range results in a gain that ranges from 2e-19 to 2e-12. The gain of the PFIR block is: sum(coefficients) x 2-shift, where shift ranges from 12 to 19. The maximum PFIR filter length is a function of GC5018 clock rate and output sample rate and is limited by the number of coefficient memory registers. The maximum number of taps is 64 and the minimum number is 32 (for both CDMA and UMTS). Lengths between these limits can be specified in increments of 4. For strapped UMTS with double length filters, the range of taps available is 64 to 128 in increments of 8. Subject to the above minimum and maximum values, the number of maximum taps available is: UMTS Mode: 4 x (CIC DECIMATION x 2) Strapped UMTS Mode: 8 x (CIC DECIMATION x 2) CDMA Mode: 2 x (CIC DECIMATION x 2) PFIR coefficients and gain shift values are shared between both A and B CDMA channels in a DDC block. Example PFIR filter lengths available based on mode and rxclk frequency: RECEIVE DIGITAL SIGNAL PROCESSING 29 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Mode rxclk (MHz) CIC DECIMATIO N PFIR MAX LENGTH PFIR MIN LENGTH COMMENTS UMTS 153.60 10 64 32 UMTS, 1 to 6 DDC channels UMTS 122.88 8 64 32 UMTS, 1 to 6 DDC channels UMTS 153.60 10 128 64 Strapped UMTS double length PFIR configuration; 1, 2 or 3 DDC channels. UMTS 122.88 8 128 64 Strapped UMTS double length PFIR configuration; 1, 2 or 3 DDC channels CDMA 157.2864 32 64 32 CDMA2000 CDMA 122.88 25 64 32 CDMA2000 CDMA 78.6432 16 64 32 CDMA2000 low power configuration CDMA 153.60 30 64 32 TD-SCDMA CDMA 81.92 16 64 32 TD-SCDMA CDMA 76.80 15 60 32 TD-SCDMA low power configuration Coefficients are organized in four groups of 16 words, each 18 bits wide. For fully utilized filters, the 64 coefficients are loaded 0 through 31 into the first and second RAMs, and 32 through 63 into the third and fourth RAMs. The 16 bit MSBs and 2 bit LSBs are written into the RAMs using different page register values. Shorter filters require the coefficients be loaded into the 4 rams equally, starting from address 0 and address 16. For example, a CFIR coefficient set for a symmetric 60 tap TD-SCDMA PFIR is: Taps Coefficient Taps 0 = 59 -2 15 = 44 Coefficient 420 1 = 58 1 16 = 43 -331 2 = 57 4 17 = 42 -319 3 = 56 -8 18 = 41 744 4 = 55 -2 19 = 40 -440 5 = 54 21 20 = 39 -1005 6 = 53 -13 21 = 38 2389 7 = 52 -28 22 = 37 514 8 = 51 46 23 = 36 -6182 9 = 50 1 24 = 35 1845 10 = 49 -85 25 = 34 12959 11 = 48 96 26 = 33 -8691 12 = 47 82 27 = 32 -27246 13 = 46 -266 28 = 31 34166 14 = 45 38 29 = 30 131071 The first 15 coefficients are loaded into addresses 0 through 14 in the first coefficient RAM, the second group of 15 are loaded into addresses 16 through 30 corresponding to the second coefficient RAM, the third group of 15 are loaded into the third coefficient ram at addresses 0 through 14, and the fourth group of 15 are loaded into addresses 16 through 30 in the fourth coefficient RAM. Loading the 18 bit coefficients requires 2 writes per coefficient, one for the upper 16 bits and another for the lower 2 bits. To program this coefficient set for the DDC2 PFIR, the following control microprocessor interface sequence would be used. 30 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Step Address a[5:0] Data d[15:0] Description 1 0x21 0x0400 Page register for DDC2 CFIR Coefficient RAMs 0-15 and 16-31, LSBs. 2 0x00 0x0002 2 lower bits of coefficient 0 3 0x01 0x0001 2 lower bits of coefficient 1 4 0x02 0x0000 2 lower bits of coefficient 2 5 0x03 0x0000 2 lower bits of coefficient 3 6 0x04 0x0002 2 lower bits of coefficient 4 7 0x05 0x0001 2 lower bits of coefficient 5 8 0x06 0x0003 2 lower bits of coefficient 6 9 0x07 0x0000 2 lower bits of coefficient 7 10 0x08 0x0002 2 lower bits of coefficient 8 11 0x09 0x0001 2 lower bits of coefficient 9 12 0x0A 0x0003 2 lower bits of coefficient 10 13 0x0B 0x0000 2 lower bits of coefficient 11 14 0x0C 0x0002 2 lower bits of coefficient 12 15 0x0D 0x0002 2 lower bits of coefficient 13 16 0x0E 0x0002 2 lower bits of coefficient 14 17 0x0F 0x0000 2 lower bits of unused coefficient RAM location 18 0x10 0x0000 2 lower bits of coefficient 15 19 0x11 0x0001 2 lower bits of coefficient 16 20 0x12 0x0001 2 lower bits of coefficient 17 21 0x13 0x0000 2 lower bits of coefficient 18 22 0x14 0x0000 2 lower bits of coefficient 19 23 0x15 0x0003 2 lower bits of coefficient 20 24 0x16 0x0001 2 lower bits of coefficient 21 25 0x17 0x0002 2 lower bits of coefficient 22 26 0x18 0x0002 2 lower bits of coefficient 23 27 0x19 0x0001 2 lower bits of coefficient 24 28 0x1A 0x0003 2 lower bits of coefficient 25 29 0x1B 0x0001 2 lower bits of coefficient 26 30 0x1C 0x0002 2 lower bits of coefficient 27 31 0x1D 0x0002 2 lower bits of coefficient 28 32 0x1E 0x0003 2 lower bits of coefficient 29 33 0x1F 0x0000 2 lower bits of unused coefficient RAM location 34 0x21 0x0420 Page register for DDC2 CFIR Coefficient RAMs 32-47 and 48-63, LSBs. 35 0x00 0x0003 2 lower bits of coefficient 30 36 0x01 0x0002 2 lower bits of coefficient 31 37 0x02 0x0002 2 lower bits of coefficient 32 38 0x03 0x0001 2 lower bits of coefficient 33 39 0x04 0x0003 2 lower bits of coefficient 34 40 0x05 0x0001 2 lower bits of coefficient 35 41 0x06 0x0002 2 lower bits of coefficient 36 42 0x07 0x0002 2 lower bits of coefficient 37 43 0x08 0x0001 2 lower bits of coefficient 38 44 0x09 0x0003 2 lower bits of coefficient 39 45 0x0A 0x0000 2 lower bits of coefficient 40 46 0x0B 0x0000 2 lower bits of coefficient 41 47 0x0C 0x0001 2 lower bits of coefficient 42 48 0x0D 0x0001 2 lower bits of coefficient 43 RECEIVE DIGITAL SIGNAL PROCESSING 31 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 32 Step Address a[5:0] Data d[15:0] Description 49 0x0E 0x0000 2 lower bits of coefficient 44 50 0x0F 0x0000 2 lower bits of unused coefficient RAM location 51 0x10 0x0002 2 lower bits of coefficient 45 52 0x11 0x0002 2 lower bits of coefficient 46 53 0x12 0x0002 2 lower bits of coefficient 47 54 0x13 0x0000 2 lower bits of coefficient 48 55 0x14 0x0003 2 lower bits of coefficient 49 56 0x15 0x0001 2 lower bits of coefficient 50 57 0x16 0x0002 2 lower bits of coefficient 51 58 0x17 0x0000 2 lower bits of coefficient 52 59 0x18 0x0003 2 lower bits of coefficient 53 60 0x19 0x0001 2 lower bits of coefficient 54 61 0x1A 0x0002 2 lower bits of coefficient 55 62 0x1B 0x0000 2 lower bits of coefficient 56 63 0x1C 0x0000 2 lower bits of coefficient 57 64 0x1D 0x0001 2 lower bits of coefficient 58 65 0x1E 0x0002 2 lower bits of coefficient 59 66 0x1F 0x0000 2 lower bits of unused coefficient RAM location 67 0x21 0x0440 Page register for DDC2 PFIR Coefficient RAMs 0-15 and 16-31, MSBs. 68 0x00 0xFFFF Upper 16 bits of coefficient 0 69 0x01 0x0000 Upper 16 bits of coefficient 1 70 0x02 0x0001 Upper 16 bits of coefficient 2 71 0x03 0xFFFE Upper 16 bits of coefficient 3 72 0x04 0xFFFF Upper 16 bits of coefficient 4 73 0x05 0x0005 Upper 16 bits of coefficient 5 74 0x06 0xFFFC Upper 16 bits of coefficient 6 75 0x07 0xFFF9 Upper 16 bits of coefficient 7 76 0x08 0x000B Upper 16 bits of coefficient 8 77 0x09 0x0000 Upper 16 bits of coefficient 9 78 0x0A 0xFFEA Upper 16 bits of coefficient 10 79 0x0B 0x0018 Upper 16 bits of coefficient 11 80 0x0C 0x0014 Upper 16 bits of coefficient 12 81 0x0D 0xFFBD Upper 16 bits of coefficient 13 82 0x0E 0x0009 Upper 16 bits of coefficient 14 83 0x0F 0x0000 Upper 16 bits of unused coefficient RAM location 84 0x10 0x0069 Upper 16 bits of coefficient 15 85 0x11 0xFFAD Upper 16 bits of coefficient 16 86 0x12 0x0FFB0 Upper 16 bits of coefficient 17 87 0x13 0x0B0A Upper 16 bits of coefficient 18 88 0x14 0xFF92 Upper 16 bits of coefficient 19 89 0x15 0xFF04 Upper 16 bits of coefficient 20 90 0x16 0x0255 Upper 16 bits of coefficient 21 91 0x17 0x0080 Upper 16 bits of coefficient 22 92 0x18 0xF9F6 Upper 16 bits of coefficient 23 93 0x19 0x01CD Upper 16 bits of coefficient 24 94 0x1A 0x0CA7 Upper 16 bits of coefficient 25 95 0x1B 0xF783 Upper 16 bits of coefficient 26 96 0x1C 0xE564 Upper 16 bits of coefficient 27 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Step Address a[5:0] Data d[15:0] Description 97 0x1D 0x215D Upper 16 bits of coefficient 28 98 0x1E 0x7FFF Upper 16 bits of coefficient 29 99 0x1F 0x0000 Upper 16 bits of unused coefficient RAM location 100 0x21 0x0460 Page register for DDC2 PFIR Coefficient RAMS 32-47 AND 48-63, MSBs. 101 0x00 0x7FFF Upper 16 bits of coefficient 30 102 0x01 0x215D Upper 16 bits of coefficient 31 103 0x02 0xE564 Upper 16 bits of coefficient 32 104 0x03 0xF783 Upper 16 bits of coefficient 33 105 0x04 0x0CA7 Upper 16 bits of coefficient 34 106 0x05 0x01CD Upper 16 bits of coefficient 35 107 0x06 0xF9F6 Upper 16 bits of coefficient 36 108 0x07 0x0080 Upper 16 bits of coefficient 37 109 0x08 0x0255 Upper 16 bits of coefficient 38 110 0x09 0xFF04 Upper 16 bits of coefficient 39 111 0x0A 0xFF92 Upper 16 bits of coefficient 40 112 0x0B 0x00BA Upper 16 bits of coefficient 41 113 0x0C 0xFFB0 Upper 16 bits of coefficient 42 114 0x0D 0xFFAD Upper 16 bits of coefficient 43 115 0x0E 0x0069 Upper 16 bits of coefficient 44 116 0x0F 0x008D Upper 16 bits of unused coefficient RAM location 117 0x10 0x0009 Upper 16 bits of coefficient 45 118 0x11 0xFFBD Upper 16 bits of coefficient 46 119 0x12 0x0014 Upper 16 bits of coefficient 47 120 0x13 0x0018 Upper 16 bits of coefficient 48 121 0x14 0xFFEA Upper 16 bits of coefficient 49 122 0x15 0x0000 Upper 16 bits of coefficient 50 123 0x16 0x000B Upper 16 bits of coefficient 51 124 0x17 0xFFF9 Upper 16 bits of coefficient 52 125 0x18 0xFFFC Upper 16 bits of coefficient 53 126 0x19 0x0005 Upper 16 bits of coefficient 54 127 0x1A 0xFFFF Upper 16 bits of coefficient 55 128 0x1B 0xFFFE Upper 16 bits of coefficient 56 129 0x1C 0x0001 Upper 16 bits of coefficient 57 130 0x1D 0x0000 Upper 16 bits of coefficient 58 131 0x1E 0xFFFF Upper 16 bits of coefficient 59 132 0x1F 0x0000 Upper 16 bits of unused coefficient RAM location 133 0x21 0x0500 Page register for DDC2 control registers 0-31 134 0x00 0x8EE0 DDC2 FIR_MODE register; cdma_mode enabled, 60 tap PFIR, 58 tap CFIR 135 0x01 0x2000 DDC2 PFIR gain = sum(taps)x2^-18 and CFIR gain = sum(taps)x2^-19 RECEIVE DIGITAL SIGNAL PROCESSING 33 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 PROGRAMMING VARIABLE DESCRIPTION crastarttap_pfir(4:0) Number of DDC PFIR filter taps is 4x(crastartap+1) For double length PFIR the number of taps is 8x(crastartap+1) cdma_mode When set, puts the CFIR & PFIR blocks in CDMA mode. mpu_ram_read What set, the PFIR and CFIR coefficient rams are readable via the MPU control interface. The GC5018 signal path is not operational when this bit is set, it is intended for debug purposes only. pfir_gain(2:0) Sets the gain of the PFIR filter. The range is from 2e-19 to 2e-12; "000"= 2e-19 and "111"= 2e-12 double_tap(1:0) When set, puts two adjacent DDC (2k and 2k+1, k=0 to 2) in double length (from 64 to128 tap) UMTS mode. Set to "00" for normal mode. In double tap mode, data out of the last PFIR ram in the main DDC (DDC0, DDC2, DDC4 or DDC6) is sent to the adjacent secondary DDC (DDC1, DDC3, DDC5 or DDC7) PFIR as input thus forming a 128-tap delay line. Data received from the adjacent PFIR summers is added into the Main DDC's PFIR sum to form the final output. When using double tap mode, set double_tap to "10" for the main DDC, and to "01" for the secondary DDC. When in double tap mode, the first half of the coefficients should be loaded into the main DDC (DDC0, DDC2, DDC4 or DDC6), the remaining coefficients are loaded into the secondary DDC (DDC1, DDC3, DDC5 or DDC7). In double tap mode, the main DDC must be turned on (ddc_ena=1), and the secondary DDC must be turned off (ddc_ena=0). The PFIR filter's 18 bit coefficients are loaded in four 16 word memories. Note: PFIR filter coefficients are shared between A and B channels of a DDC block when in CDMA mode. 3.2.9 DDC RMS Power Meter 18 I 36 37 Q 18 55-bit Integrator 36 clear sync RMS power 55-bit Register 8-bit sync delay counter 8 delay (in samples) 18-bit interval counter 8 interval (in 1024 sample increments) transfer 18-bit integration counter interrupt 16 integration (in 4 sample increments) Figure 3-26. DDC RMS Power Meter Block Diagram 34 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 interrupt interrupt interrupt sync delay integration time integration time interval time sync event integration start integration time interval time integration start integration start Figure 3-27. DDC RMS Power Meter Timing Each DDC channel includes an RMS power meter which is used to measure the total power within the channel pass band. The power meter samples the I and Q data stream after the PFIR filter. Both 18 bit I and Q data are squared, summed, and then integrated over a period determined by a programmable counter. The integration time is a 16 bit word which is programmed into the 18 bit counter. There is a programmable 18 bit interval timer which sets the interval over which power measurements are made. The timer counts in increments of 1024 samples. This allows the user to select intervals from 1 x 1024 samples up to 256 x 1024 samples. For UMTS systems with sample rate rate at 7.68 MHz, the power meter interval range is from 133 S to 34.1 mS. For a CDMA system with the sample rate at 2.4576 MHz, the power meter interval range is 417 S to 107 mS. The power measurement process starts with a sync event. The integration will start at sync event +3 chips + sync_delay. The 8 bit delay register permits delays from 1 to 256 samples after sync. The integration will continue until the integration count is met. At that point, the result in the 55 bit accumulator is transferred to the read holding register and an interrupt is generated indicating the power value is ready to read. The interval counter continues until the programmed interval count is reached. When reached, the integration counter and the interval counter start over again. Each time the integration count is reached, the 55 result bits are again transferred to the read register overwriting the previous value and an interrupt is generated signifying the data is ready to be read. Failure to read the data timely will result in overwriting the previous interval measurement. Sync starts the process. Whenever a sync is received, all the counters are reset to zero no matter what the status. For UMTS, I and Q are calculated and the integrated power is read. When in CDMA mode the power is calculated for both the A ( Signal ) path and the B ( Diversity) signal. As a result, there are two 55-bit words representing the Signal and Diversity when in CDMA mode. The power read is: power = [ (I2 + Q2) x (N x 4 + 1) ] where N is the integration count. PROGRAMMING VARIABLE DESCRIPTION pmeter_result_a(54:0) 55 bit UMTS or CDMA mode A channel power measurement result. pmeter_result_b(54:0) 55 bit CDMA mode B channel power measurement result. pmeter_sqr_sum_ddc(15:0) Integration (square and sum) count in increments of four samples. pmeter_sync_delay_ddc(7:0) Sync delay count in samples. pmeter_interval_ddc(7:0) The measurement interval in increments of 2048 samples. This value must be greater than SQR_SUM. ssel_pmeter(2:0) Sync source selection. pmeter_sync_disable Turns off the sync to the channel power meter. This can be used to individually turn off syncs to a channels power meter while still having syncs to other power meters on the chip. RECEIVE DIGITAL SIGNAL PROCESSING 35 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 3.2.10 DDC AGC I, Q limit & round 18 18 I, Q outputs (up to 25-bits in AGC bypass mode) threshold zero mask 12 integer & 12 fractional 8 24 magnitude 8 4 compare 2 clear Freeze (from register bit shift 29 accumulate ocnt ucnt 8 4 29 amin 16 limit dzro dsat 4 2 under/over detect amax 16 dblw dabv 4 4 4 shift select S=+/-1, D=4-bit shift 5 24 Freeze (from sync source) min limit max limit Gain gain adjust 24 Figure 3-28. DDC AGC Block Diagram The GC5018 automatic gain control circuit is shown above. The basic operation of the circuit is to multiply the 18 bit input data from the PFIR by a 24-bit gain word that represents a gain or attenuation in the range of 0 to 4096. The gain format is mixed integer and fraction. The 12-bit integer allows the gain to be boosted by up to factor of 4096 (72 dB). The 12-bit fractional part allows the gain to be adjusted up or down in steps of one part in 4096, or approximately 0.002 dB. If the integer portion is zero, then the circuit attenuates the signal. The gain adjusted output data is saturated to full scale and then rounded to between 4 and 18 bits in steps of one bit. The AGC portion of the circuit is used to automatically adjust the gain so that the median magnitude of the output data matches a target value, which is performed by comparing the magnitude of the output data with a target threshold. If the magnitude is greater than the threshold, then the gain is decreased, otherwise it is increased. The gain is adjusted as: G(t) = G + A(t), where G is the default, user supplied gain value, and A(t) is the time varying adjustment. A(t) is updated as A(t) = A(t) + G(t)xSx2-D , where S=1 if the magnitude is less than the threshold and is -1 if the magnitude exceeds the threshold, and where D sets the adjustment step size. Note that the adjustment is a fraction of the current gain. This is designed to set the AGC noise level to a known and acceptable level while keeping the AGC convergence and tracking rate constant, independent of the gain level. The AGC noise will be equal to 2-D and the AGC attack and decay rate will be exponential, with a time constant equal to 2-D. Hence, the AGC will increase or decrease by 0.63 times G(t) in 2D updates. If one assumes the data is random with a Gaussian distribution, which is valid for UMTS if more than 12 users with different codes have been overlaid, then the relationship between the RMS level and the median is MEDIAN = 0.6745xRMS, hence the threshold should be set to 0.6745 times the desired RMS level. 36 RECEIVE DIGITAL SIGNAL PROCESSING www.ti.com GC5018 8-CHANNEL WIDEBAND RECEIVER SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 The gain step size can be set using four different values of D, each of which is a 4 bit integer. D can range from 3 to 18. The user can specify values of D for different situations, i.e., when the signal magnitude is below the user-specified threshold (Dblw), is above the threshold (Dabv), is consistently equal to zero (Dzro) or is consistently equal to maximum (Dsat). It is important to note that D represents a gain step size. Smaller values of D represent larger gain steps. The definition of equal to zero is any number when masked by zero_mask is considered to be zero. This permits consistently very small amplitude signals to have their gain increased rapidly. Separate programmable D values allow the user to set different attack and decay time constants, and to set shorter time constants for when the signal falls too low (equal to zero), or is too high (saturates). The magnitude is considered to be consistently equal to zero by using a 4-bit counter that counts up every time the 8-bit magnitude value is zero, and counts down otherwise. If the counter's value exceeds a user specified threshold, then Dabv is used. Similarly the magnitude is considered too high by using a counter that counts up when the magnitude is maximum, and counts down otherwise. If this counter exceeds another user specified threshold, then Dsat is used. As an example, if the AGC's current gain at a particular moment in time is 5.123, and the magnitude of the signal is greater than zero, but less than the user-programmed threshold. Step size Dblw will be used to modify the gain for the next sample. This represents the AGC attack profile. If Dblw is set to a value of 5, then the gain for the next sample will be 5.123 + 5.123 x 2-5 = 5.123 + 0.160 = 5.283. If the signal's magnitude is still less than the user-programmed threshold, then the gain for the next sample will be 5.283 + 5.283 x 2-5 = 5.283 + 0.165 = 5.448. This continues until the signal's magnitude exceeds the user-programmed threshold. When the magnitude exceeds threshold (but is not saturated), then step size Dabv is automatically employed as a size rather than Dblw. The AGC converges linearly in dB with a step size of 40log(1+2-D) when the error is greater than 12 dB (i.e. the gain is off by 12 dB or more). Within 6 dB the behavior is approximately a exponential decay with a time constant of 2(D-0.5) samples. The suggested value of D is 5 or 6 when the error is greater than 12dB (i.e., in the fast range detected by consistently zero or saturated data). This gives a step size of 0.5 or 0.25 dB per sample. The suggested value when the gain is off by less than 12 dB is D=10, giving a exponential time constant for delay of around 724 samples (63% decay every 724 samples). RECEIVE DIGITAL SIGNAL PROCESSING 37 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 AGC GAIN ERROR 7 D=3 6 D=4 D=5 5 D=6 D=7 D=8 D=9 4 D=10 3 D=11 D=12 2 D=13 D=14 D=3 1 D=18 D=15 D=16 D=17 D=18 0 1 10 100 1000 10000 100000 1000000 SAMPLES Figure 3-29. AGC Gain Error Over Time vs. D The AGC noise once the AGC has converged is a random error of amplitude 2-D relative to the RMS signal level. This means that the error level is -6xD dB below the signal RMS level. At D=10 (-60 dB) the error is negligible. The plot above shows the AGC response for vales of D ranging from 3 to 18. Error dB represents the distance the signal level is from the desired target threshold. The AGC is also subject to user specified upper and lower adjustment limits. The AGC stops incrementing the gain if the adjustment exceeds Amax. It stops decrementing the gain if the adjustment is less than Amin. The input data is received with a valid flag that is high when a valid sample is received. For complex data the I and Q samples are on the same data input line and are not treated independently. An adjustment is made for the magnitude of the I sample, and then another adjustment is made for the Q sample. The AGC operates on UMTS and CDMA data. When in UMTS mode the I and Q data are each used to produce the AGC level. There is no separate I path gain and Q path gain. When in CDMA mode there are separate gain levels for the Signal and Diversity I and Q data. The I and Q for A (or the Signal ) pair is calculated and then the I' and Q' for the B (or Diversity) pair is calculated. There is a freeze mode for holding the accumulator at its current level. This will put the AGC in a hold mode using the user-programmed gain along with the current gain_adjust value. To only use the user programmed gain value as the gain, set the freeze bit and then clear the accumulator. When using the freeze bit the full 25 bit output is sent out of the AGC block to support transferring up to 25 bits when the AGC is disabled. For TDD applications, freeze mode can be controlled using a sync source. This allows rxsync_a/b/c/d to be assigned as a AGC hold signal to keep the AGC from responding during the transmit interval and run during the receive interval. The freeze register bit is logically Ored with the freeze sync source. The current AGC gain and state can also be optionally output with the DDCs I and Q output data by setting the gain_mon variable. When in this mode, the top 14 bits of the current AGC gain word are appended to the 8 bit AGC-modified I and Q output data. 38 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Output Bits(17:10) I I output data Q Q output data Bits(9:4) Bits(3:2) Bits(1:0) Gain(23:16) Gain(15:10) "00" AGC State(1:0) "00" PROGRAMMING VARIABLE DESCRIPTION agc_dblw(3:0) Below threshold gain. Sets the value of gain step size Dblw (data x current gain below threshold). Ranges from 3 to 18, and maps to a 4 bit field. For example: 3 = "0000", 4= "0001", ... 18= "1111" agc_dabv(3:0) Above threshold gain. Sets the value of gain step size Dabv (data x current gain above threshold). Ranges from 3 to 18, and maps to a 4 bit field. For example: 3 = "0000", 4= "0001", ... 18= "1111" agc_dzro(3:0) Zero signal gain. Sets the value of gain step size Dzro (data x current gain consistently zero). Ranges from 3 to 18, and maps to a 4 bit field. For example: 3 = "0000", 4= "0001", ... 18= "1111" agc_dsat (3:0) Saturated signal gain. Sets the value of gain step size Dsat (data x current gain consistently saturated). Ranges from 3 to 18, and maps to a 4 bit field. For example: 3 = "0000", 4= "0001", ... 18= "1111" agc_zero_msk(3:0) Masks the lower 4 bits of signal data so as to be considered zeros. agc_md(3:0) AGC rounding. 0000= 18 bits out, 1111= 3 bits out. agc_thresh(7:0) AGC threshold. Compared with magnitude of 8 bits of input x gain. agc_rnd_disable AGC rounding is disabled when this bit is set. agc_freeze The AGC gain adjustment updates are disable when set. agc_clear The AGC gain adjustment accumulator is cleared when set agc_gaina(23:0) 24 bit gain word for DDC A agc_gainb(23:0) 24 bit gain word for DDC B (in CDMA mode) agc_zero_cnt(3:0) When the AGC output (input x gain) is zero value this number of times, the shoft value is changed to agc_dzero. agc_max_cnt(3:0) When the AGC output (input x gain) is zero value this number of times, the shift value is changed to agc_dsat. agc_amax(15:0) The maximum value that gain can be adjusted up to. Top 12 bits are integer, bottom 4 bits are fractional. agc_amin(15:0) The minimum value that gain can be adjusted down to. Top 12 bits are integer, bottom 4 bits are fractional. gain_mon When set, combines current AGC gain with I and Q data. The 18 bit output format thus becomes: I Portion: 8 bits of AGC'd I data - Gain(23:16) - 00 Q Portion: 8 bits of AGC'd Q data - Gain(15:10) - Status(1:0) - 00. Note: Bit 0 of Status, when set, indicates the data is saturated. Bit 1 of Status, when set, indicates the data is zero. ssel_agc_freeze(2:0) Sync selection for freeze mode, 1 of 8 sources. This source is ORed with the freeze register bit ssel_gain(2:0) Sync selection for the double buffered agc_gaina and agc_gainb register. ssel_ddc_agc(2:0) Sync selection used to initialize the AGC, primarily for test purposes. 3.2.11 DDC Output Interface The baseband I/Q sample interface can be configured as serial or parallel formatted data. The serial interface closely matches the GC5316 style interface. The parallel interface is provided to interface directly to the TMS320TCI110 when delayed antenna streams used to implement channel estimation buffering and/or transport format combination indicator (TFCI) buffering are not required. The DDC output data is 2's complement format. RECEIVE DIGITAL SIGNAL PROCESSING 39 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 3.2.11.1 Serial Output Interface Serial Outputs sync 4 clkdiv frame strobe delay DDC Block 1 UMTS mode channel or 2 CDMA mode channels 2 double length PFIR UMTS I msb CDMA UMTS rxout_X_a I ch A I msb rxout_X_b I ch B I msb-1 I msb-1 rxout_X_c Q ch A Q msb I msb-2 rxout_X_d Q ch B Q msb-1 I msb-3 rx_sync_out_X Q msb four outputs from adjacent DDC block Q msb-1 Q msb-2 Q msb-3 Figure 3-30. Serial Output Block Diagram and Output Pins for Each DDC Filter Mode Each DDC block can be assigned four serial output data pins. These pins are used to transfer downconverted I/Q baseband data out of the GC5018 for subsequent processing. The usage of these pins changes depending on how the DDC block is configured. When the block is configured for two CDMA channels, a pair of serial data pins provides separate I and Q data output for the two DDC channels. Word size is selectable from 4 to 25 bits with the most significant bit first. When the DDC block is configured for a single UMTS channel, even and odd I and Q data drive the four serial pins separately, most significant bit first. Four serial pins each for I and Q data can be optionally employed (instead of two for I and two for Q) at half the output rate. This would most likely be used when two DDC channels (2k and 2k + 1, k= 0 to 5) are combined to support double-length PFIR filtering (a channel is sacrificed). Formatting for I data is then: Imsb, Imsb-1, Imsb-2, Imsb-3. Q data formatting is: Qmsb, Qmsb-1, Qmsb-2, Qmsb-3. The frame strobe signal provided on the rx_sync_out_X pins can be programmed to arrive from 0 to 3 bit clocks early via a 2 bit control parameter. The frame interval can be programmed from 1 to 63 bits. A programmable 4-bit clock divider circuit is used to specify the serial bit rate. The clock divider circuit is synchronized using a sync block discussed later in this document. Programming the serial port clock divider requires some thought and depends upon the channel's overall decimation ratio, frame sync interval, number of output bits, and CDMA-UMTS mode. In general: the serial clock divide ratio x the frame sync interval = the total receive decimation The relationship between the number of serial bits output, clock divide ratio, and overall decimation ratio is: CDMA: [overall decimation x (pser_recv_8pin + 1) ] / (pser_recv_clkdiv + 1) > pser_recv_bits + 1 UMTS: 2 x [overall decimation x (pser_recv_8pin + 1) ] / (pser_recv_clkdiv + 1) > pser_recv_bits + 1 where overall decimation = CIC DECIMATION x 2. 40 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Decimation by 2 in the output interface can be achieved by setting the frame strobe interval and clock divider to 1/2 the PFIR output rate. The serial interface samples the PFIR output each time the transfer interval defined by these two settings has completed. The decimation moment can be controlled using the rxsync_X input signal selected as the sync source for the serial interface. The timing diagram below shows the DDC serial output timing. tsetup tpd thold rxclk rxsync_X rxsync_X can be a pulse or level - interface will generate periodic frame strobes using programmed frame sync interval rxsync_out_X Programmed bit time (2 rxclk cycles for this example) 3 rxclk + 1 Programmed bit time MSB rxout_X_Y Figure 3-31. Serial Output Interface Timing Diagram PROGRAMMING VARIABLE DESCRIPTION pser_recv_fsinvl(6:0) Frame sync interval in bits pser_recv_bits(4:0) Number of data output bits - 1. i.e.: 10001= 18 bits pser_recv_clkdiv(3:0) Receive serial interface clock divider rate - 1. 0= rcclk, 15= rxclk/16 pser_recv_8pin When set, configures the serial out pins for 4I and 4Q in UMTS mode. When clear, the mode is 2I and 2Q. Used in conjunction with pser_recv_alt. pser_recv_alt When set, outputs Q data from adjacent DDC channel. pser_recv_fsdel(1:0) Number of bit clocks the frame sync is output early with respect to serial data. ssel_serial(2:0) Sync source selection, 1 of 8. tristate(6:3) Tristate controls for the rx_sync_out_X and rxout_X_X pins. Pins are in tristate when the tristate register bits are set. RECEIVE DIGITAL SIGNAL PROCESSING 41 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 3.2.11.2 Parallel Output Interface DDC7 rx_sync_out_6 par_sync_out DDC6 rxclk _out rxclk _out DDC5 rxout_7_d rxout_7_c rxout_7_b I(15) I(14) I(13) rxout_4_b rxout_4_a I(1) I(0) rxout_3_d rxout_3_c rxout_3_b Q(15) Q(14) Q(13) rxout_0_b rxout_0_a Q(1) Q(0) Output format DDC4 DDC3 DDC2 Parallel I/Q DDC1 DDC0 Figure 3-32. Parallel Output Interface Block Diagram and Output Pins When a parallel I/Q interface is required, a 32 bit time division multiplexed output mode can be selected using the rxout_X_X pins. This interface is provided for direct connection to the TMS320TCI110 Receive Chip Rate ASSP when delayed antenna streams are not required. The output sample rate, rxclk_out clock polarity, par_sync_out position and number of channels to be output are all programmable. rxclk_out par_sync_out Parallel I/Q IQ DDC0 IQ DDC1 IQ DDC2 IQ DDC3 IQ DDC4 IQ DDC5 IQ DDC6 IQ DDC7 Figure 3-33. Parallel Output Interface Timing Diagram The DDC channel serial interface synchronization source selections should all be programmed to the same value when using this parallel output interface (each DDC channel ssel_serial(2:0) in the SYNC_0 register should be programmed to the same rxsync_A/B/C/D value). Decimation by 2 in the output interface can be achieved by setting the frame strobe interval and clock divider to 1/2 the PFIR output rate. The parallel interface samples the PFIR outputs each time the transfer interval defined by these two settings has completed. 42 RECEIVE DIGITAL SIGNAL PROCESSING GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 PROGRAMMING VARIABLE DESCRIPTION par_recv_fsinvl(6:0) rx_sync_out (frame strobe) sync interval. 0 is 1 rxclk cycle and 127 is 128 rxclk cycles. par_recv_clkdiv(6:0) rxclk_out cycles per IQ channel sample; 1 is full rate, 2 is rxclk/2, etc. par_recv_chan(3:0) Number channels to be output. 0 is 1 channel, and 15 is 16 channels. par_recv_sync_del(6:0) Delays the DDC0 pser sync source to establish the timing of IQ DDC0. Increasing the value delays the par_sync_out location. par_recv_syncout_del(3:0) Delays the rx_sync_out position with respect to IQ DDC0. Setting to 0 moves the rx_sync_out pulse one rxclk_out cycle before the IQ DDC0 word, setting to 1 places it as shown above, lined up with IQ DDC0, etc. par_recv_rxclk_pol rxclk_out polarity. Outputs data on falling edges when cleared, rising edges when set. par_recv_sync_pol Parallel interface par_sync_out polarity. 0 is active low, 1 for active high par_recv_ena Parallel TCI110 style interface enabled when set, serial interface enabled when cleared. ssel_serial(2:0) DDC channel serial interface sync source selection. All DDCs should be programmed to the same sync source when using this parallel output interface. gain_mon When set, the parallel output data includes 8b I at I(15:8), 8b Q at Q(15:8), 14b AGC gain at I(7:0) and Q(7:2) and 2b AGC state at Q(1:0). tristate(6:3) 3-state controls for the rx_sync_out_X and rxout_X_X pins. Pins are in 3-state when the 3-state register bits are set. 3.2.12 DDC Checksum Generator The checksum generator is used in conjunction with the input test signal generator to implement a self test capability. sync rxclk rxout_X_a rxout_X_b rxout_X_c rxout_X_d checksum generator 16 checksum read-only results updated on each sync event results register initialized on sync event to "0000 0000 0000 0010" 15 14 1312 11 10 9 3 2 1 0 rxout_X_a rxout_X_b rxout_X_c rxout_X_d Figure 3-34. DDC Checksum Generator Block Diagram The sync for the checksum generator is internally connected to the ddc_counter output. PROGRAMMING VARIABLE ddc_chk_sum(15:0) DESCRIPTION Read only DDC channel checksum results RECEIVE DIGITAL SIGNAL PROCESSING 43 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4 GC5018 GENERAL CONTROL The GC5018 is configured over a bi-directional 16 bit parallel data microprocessor control port. The control port permits access to the control registers which configure the chip. The control registers are organized using a paged-access scheme using 6 address lines. Half of the 64 addresses (Address 32 through Address 63) represent global registers. The other 32 (Address 0 through Address 31) are paged registers. This arrangement permits accessing a large number of control registers using relatively few address lines. Global registers (Address 32 through Address 63) are used to read/write GC5018 parameters that are global in nature and can benefit from single read/write operations. Examples include chip status, reset, sync options, checksum ramp parameters, interrupt sources, interrupt masks, 3-state controls and the page register. Global Address 33 is the page register. Writing a 16 bit value to this register sets the page to which future write or read operations performed. These paged-registers contain the actual parameters that configure the chip and are accessed by writing/reading address 0 through address 31. The global 3-state register can be used to 3-state the output drivers on the GC5018, and also includes the capability of disabling the chip's internal rxclk. PROGRAMMING VARIABLE DESCRIPTION rxclk_ena Enables the internal rxclk when set. When cleared, the GC5018 will ignore the rxclk input signal and hold the internal clock low. 3-state(10:0) Various output pins are forced into tristate mode when these bits are asserted. See the GBL_3-STATE register description for pin groups to bit assignments. arst_func When asserted, the internal datapath is held reset. The control register programming is not affected. 4.1 Microprocessor Interface Control Data, Address, and Strobes The microprocessor control bus consists of 16 bi-directional control data lines d[15:0], 6 address lines a[5:0], a read enable line rd_n, a write enable line wr_n, and a chip enable line ce_n. These lines usually interface to a microprocessor or DSP chip and is intended to look like a block of memory. The interface can be operated in a 3 pin control mode (using rd_n, wr_n and ce_n) or 2 pin control mode (using wr_n and ce_n with rd_n always low). 44 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.1.1 MPU Timing Diagrams tREC ce_n wr_n rd_n tCSU tHIZ a[5:0] tCDLY d[15:0] tCOH valid data Figure 4-1. Read Operation - 3 pin control mode tREC ce_n wr_n tCSU tHIZ a[5:0] tCDLY d[15:0] tCOH valid data Figure 4-2. Read Operation - 2 pin control mode (rd_n tied low) GC5018 GENERAL CONTROL 45 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 tREC ce_n tC CSPW wr_n tCSU rd_n a[5:0] t CHD d[15:0] valid data t EWCSU Figure 4-3. Write Operation - 3 pin control mode tREC ce_n tC CSPW wr_n tCSU a[5:0] t CHD d[15:0] valid data t EWCSU Figure 4-4. Write Operation - 2 pin control mode (rd_n tied low) 4.2 Synchronization Signals Various function blocks within the GC5018 need to be synchronized in order to realize predictable results. The GC5018 provides a flexible system where each function block that requires synchronization can be independently synchronized from either device pins or from a software "one-shot". The one-shot option is setup and triggered through control registers. The four sync input pins, rxsync_a, rxsync_b, rxsync_c and rxsync_d are qualified on the rxclk rising clock edge. Table 4-1 shows the different sync modes available. 46 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Table 4-1. Different Sync Modes Available SYNC SELECT CODE RECEIVE SYNC SOURCE 000 rxsync_a 001 rxsync_b 010 rxsync_c 011 rxsync_d 100 ddc sync counter terminal count 101 ddc sync triggered by s/w oneshot (register bit) 110 0 (always off) 111 1 (always on) Table 4-2 and Table 4-3 summarizes the blocks which have functions that can be synchronized using the above eight sync source options. Table 4-2. Receive Common Syncs Sync Name Purpose sync_ddc_counter Initializes the receive sync counter sync_ddc Initializes the receive ADC interface and clock generation circuits sync_rxsync_out selects sync signal to be output on the rx_sync_out pin. sync_adc_fifo Initializes the input and output pointers in the ADC fifo circuits. sync_tst_decim Initializes the testbus decimation counter. sync_recv_pmeterX Initializes the rxin power meters. {X = 0,1,2 or 3} sync_ragc_interval_X Initializes the rxin receive AGC timers. {X = 0,1,2 or 3} sync_ragc_freeze_X rxin receive AGC freeze mode control. {X = 0,1,2 or 3} sync_ragc_clear_X Initializes the receive AGC error accumulator. {X = 0,1,2 or 3} Table 4-3. DDC Channel Syncs Sync Name Purpose sync_ddc_tadj Selects zero stuff moment in the tadj fine adjustment section. sync_ddc_tadj_reg Updates the tadj output pointer register delay in the tadj coarse adjustment section. sync_ddc_nco Resets the NCO accumulator. sync_ddc_freq Updates the NCO freq registers. sync_ddc_phase Updates the NCO phase register. sync_ddc_dither Initializes the NCO dither circuits. sync_ddc_cic Selects the CIC decimation moment. sync_ddc_pmeter Initializes the receive channel power meters. sync_ddc_gain Updates the DDC channel AGC gain registers sync_ddc_agc Initializes the AGC accumulator. sync_ddc_agc_freeze AGC freeze mode control. sync_ddc_serial Initializes the receive serial interface. A 32-bit general purpose timer is included in the synchronization function. The timer loads the user programmed terminal count on a sync event, and counts down to zero using rxclk. The width of the terminal count pulse can also be programmed up to rxclk cycles. The timers output can be used as a sync source for any other circuits requiring a sync if desired, and can also be routed to the rx_sync_out pin. GC5018 GENERAL CONTROL 47 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 PROGRAMMING VARIABLE DESCRIPTION ddc_counter(31:0) 32-bit programmable terminal count ddc_counter_width(7:0) 8-bit programmable terminal count pulse width ssel_ddc_counter(2:0) Sync source selection for the ddc counter ssel_rxsync_out(2:0) Sync source selection for the rx_sync_out pin tri-state(0) When set, the interrupt and rx_sync_out pins are tri-stated. rx_oneshot Register bit used to generate the S/W oneshot signal for sync. This bit must be programmed from cleared to set in order to generate a rising edge sync signal. 4.3 Interrupt Handling When a GC5018 block sets an interrupt, the interrupt pin will go active if the interrupt source is masked. The microprocessor should then read the interrupt register to determine the source of interrupt. The microprocessor will then have to write the interrupt register to clear the interrupt pin and interrupt source. The interrupt register and interrupt mask are located in the global registers section of control registers. not the the the The GC5018 has 16 interrupt sources; power meters in each of the eight DDC blocks, power meters in the four receive input interface, and four rxin_X_ovr (adc overflow) input pins where X={a,b,c,d}. PROGRAMMING VARIABLE DESCRIPTION pmeterX_im(7:0) Channel pmeter interrupt mask bits. Interrupt source is masked when set. recv_pmeterX_im(3:0) Receive input power meter interrupt masks. rxin_X_ovr_im ADC overflow input pin interrupt masks. pmeterX(7:0) Channel pmeter interrupt status. recv_pmeterX(3:0) Receive input power meter interrupt status. rxin_X_ovr ADC overflow input pin interrupt status. intr_clr When asserted, holds all interrupt status bits cleared. The interrupt pin will be inactive (always low) when this bit is set. Intended for lab/debug use only tri-state(0) When set, the interrupt and rx_sync_out pins are tri-stated. 4.4 GC5018 Programming The GC5018 includes over 3000 internal configuration registers and therefore implements a paged addressing scheme. The register map includes a global control variables register address space that is accessed directly when the a5 signal is high. This global control variables address space includes the page register. All other registers are addressed using a combination of an address comprised of the internal page register contents and the 6-bit external address; a5, a4, a3, a2, a1 and a0. The page register is accessed when the 6-bit address a5:a0 is 0x21 (or binary "100001"). 48 Page Register Contents in Hex Address Pin a5 don't care 1 Global Control Variables 0x00 through 0x1F 0x0000 0 DDC0 PFIR taps 0 through 31 coefficient lsbs (1:0) 0x0020 0 DDC0 PFIR taps 32 through 63 coefficient lsbs (1:0) 0x0040 0 DDC0 PFIR taps 0 through 31 coefficient msbs (17:2) 0x0060 0 DDC0 PFIR taps 32 through 63 coefficient msbs (17:2) 0x0080 0 DDC0 CFIR taps 0 through 31 coefficient lsbs (1:0) GC5018 GENERAL CONTROL Registers Addressed With 5 Bit Address Space, Pins (a4:a0) GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Page Register Contents in Hex Address Pin a5 Registers Addressed With 5 Bit Address Space, Pins (a4:a0) 0x00A0 0 DDC0 CFIR taps 32 through 63 coefficient lsbs (1:0) 0x00C0 0 DDC0 CFIR taps 0 through 31 coefficient msbs (17:2) 0x00E0 0 DDC0 CFIR taps 32 through 63 coefficient msbs (17:2) 0x0100 0 DDC0 Control Registers 0x00 through 0x1F 0x0120 0 DDC0 Control Registers 0x20 through 0x3F 0x0200 0 DDC1 PFIR taps 0 through 31 coefficient lsbs (1:0) 0x0220 0 DDC1 PFIR taps 32 through 63 coefficient lsbs (1:0) 0x0240 0 DDC1 PFIR taps 0 through 31 coefficient msbs (17:2) 0x0260 0 DDC1 PFIR taps 32 through 63 coefficient msbs (17:2) 0x0280 0 DDC1 CFIR taps 0 through 31 coefficient lsbs (1:0) 0x02A0 0 DDC1 CFIR taps 32 through 63 coefficient lsbs (1:0) 0x02C0 0 DDC1 CFIR taps 0 through 31 coefficient msbs (17:2) 0x02E0 0 DDC1 CFIR taps 32 through 63 coefficient msbs (17:2) 0x0300 0 DDC1 Control Registers 0x00 through 0x1F 0x0320 0 DDC1 Control Registers 0x20 through 0x3F 0x0400 0 DDC2 PFIR taps 0 through 31 coefficient lsbs (1:0) 0x0420 0 DDC2 PFIR taps 32 through 63 coefficient lsbs (1:0) 0x0440 0 DDC2 PFIR taps 0 through 31 coefficient msbs (17:2) 0x0460 0 DDC2 PFIR taps 32 through 63 coefficient msbs (17:2) 0x0480 0 DDC2 CFIR taps 0 through 31 coefficient lsbs (1:0) 0x04A0 0 DDC2 CFIR taps 32 through 63 coefficient lsbs (1:0) 0x04C0 0 DDC2 CFIR taps 0 through 31 coefficient msbs (17:2) 0x04E0 0 DDC2 CFIR taps 32 through 63 coefficient msbs (17:2) 0x0500 0 DDC2 Control Registers 0x00 through 0x1F 0x0520 0 DDC2 Control Registers 0x20 through 0x3F 0x0600 0 DDC3 PFIR taps 0 through 31 coefficient lsbs (1:0) 0x0620 0 DDC3 PFIR taps 32 through 63 coefficient lsbs (1:0) 0x0640 0 DDC3 PFIR taps 0 through 31 coefficient msbs (17:2) 0x0660 0 DDC3 PFIR taps 32 through 63 coefficient msbs (17:2) 0x0680 0 DDC3 CFIR taps 0 through 31 coefficient lsbs (1:0) 0x06A0 0 DDC3 CFIR taps 32 through 63 coefficient lsbs (1:0) 0x06C0 0 DDC3 CFIR taps 0 through 31 coefficient msbs (17:2) 0x06E0 0 DDC3 CFIR taps 32 through 63 coefficient msbs (17:2) 0x0700 0 DDC3 Control Registers 0x00 through 0x1F 0x0720 0 DDC3 Control Registers 0x20 through 0x3F 0x0800 0 DDC4 PFIR taps 0 through 31 coefficient lsbs (1:0) 0x0820 0 DDC4 PFIR taps 32 through 63 coefficient lsbs (1:0) 0x0840 0 DDC4 PFIR taps 0 through 31 coefficient msbs (17:2) 0x0860 0 DDC4 PFIR taps 32 through 63 coefficient msbs (17:2) 0x0880 0 DDC4 CFIR taps 0 through 31 coefficient lsbs (1:0) 0x08A0 0 DDC4 CFIR taps 32 through 63 coefficient lsbs (1:0) 0x08C0 0 DDC4 CFIR taps 0 through 31 coefficient msbs (17:2) 0x08E0 0 DDC4 CFIR taps 32 through 63 coefficient msbs (17:2) GC5018 GENERAL CONTROL 49 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 50 Page Register Contents in Hex Address Pin a5 Registers Addressed With 5 Bit Address Space, Pins (a4:a0) 0x0900 0 DDC4 Control Registers 0x00 through 0x1F 0x0920 0 DDC4 Control Registers 0x20 through 0x3F 0x0A00 0 DDC5 PFIR taps 0 through 31 coefficient lsbs (1:0) 0x0A20 0 DDC5 PFIR taps 32 through 63 coefficient lsbs (1:0) 0x0A40 0 DDC5 PFIR taps 0 through 31 coefficient msbs (17:2) 0x0A60 0 DDC5 PFIR taps 32 through 63 coefficient msbs (17:2) 0x0A80 0 DDC5 CFIR taps 0 through 31 coefficient lsbs (1:0) 0x0AA0 0 DDC5 CFIR taps 32 through 63 coefficient lsbs (1:0) 0x0AC0 0 DDC5 CFIR taps 0 through 31 coefficient msbs (17:2) 0x0AE0 0 DDC5 CFIR taps 32 through 63 coefficient msbs (17:2) 0x0B00 0 DDC5 Control Registers 0x00 through 0x1F 0x0B20 0 DDC5 Control Registers 0x20 through 0x3F 0x0C00 0 DDC6 PFIR taps 0 through 31 coefficient lsbs (1:0) 0x0C20 0 DDC6 PFIR taps 32 through 63 coefficient lsbs (1:0) 0x0C40 0 DDC6 PFIR taps 0 through 31 coefficient msbs (17:2) 0x0C60 0 DDC6 PFIR taps 32 through 63 coefficient msbs (17:2) 0x0C80 0 DDC6 CFIR taps 0 through 31 coefficient lsbs (1:0) 0x0CA0 0 DDC6 CFIR taps 32 through 63 coefficient lsbs (1:0) 0x0CC0 0 DDC6 CFIR taps 0 through 31 coefficient msbs (17:2) 0x0CE0 0 DDC6 CFIR taps 32 through 63 coefficient msbs (17:2) 0x0D00 0 DDC6 Control Registers 0x00 through 0x1F 0x0D20 0 DDC6 Control Registers 0x20 through 0x3F 0x0E00 0 DDC7 PFIR taps 0 through 31 coefficient lsbs (1:0) 0x0E20 0 DDC7 PFIR taps 32 through 63 coefficient lsbs (1:0) 0x0E40 0 DDC7 PFIR taps 0 through 31 coefficient msbs (17:2) 0x0E60 0 DDC7 PFIR taps 32 through 63 coefficient msbs (17:2) 0x0E80 0 DDC7 CFIR taps 0 through 31 coefficient lsbs (1:0) 0x0EA0 0 DDC7 CFIR taps 32 through 63 coefficient lsbs (1:0) 0x0EC0 0 DDC7 CFIR taps 0 through 31 coefficient msbs (17:2) 0x0EE0 0 DDC7 CFIR taps 32 through 63 coefficient msbs (17:2) 0x0F00 0 DDC7 Control Registers 0x00 through 0x1F 0x0F20 0 DDC7 Control Registers 0x20 through 0x3F 0x1000 0 Receive Input AGC0 Error RAM addresses 0 through 31 0x1020 0 Receive Input AGC0 Error RAM addresses 32 through 63 0x1040 0 Receive Input AGC0 DVGA RAM addresses 0 through 31 0x1080 0 Receive Input AGC0 Gain RAM addresses 0 through 31 0x10A0 0 Receive Input AGC0 Gain RAM addresses 32 through 63 0x1100 0 Receive Input AGC1 Error RAM addresses 0 through 31 0x1120 0 Receive Input AGC1 Error RAM addresses 32 through 63 0x1140 0 Receive Input AGC1 DVGA RAM addresses 0 through 31 0x1180 0 Receive Input AGC1 Gain RAM addresses 0 through 31 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Page Register Contents in Hex Address Pin a5 0x11A0 0 Receive Input AGC1 Gain RAM addresses 32 through 63 0x1400 0 Receive Input AGC2 Error RAM addresses 0 through 31 0x1420 0 Receive Input AGC2 Error RAM addresses 32 through 63 0x1440 0 Receive Input AGC2 DVGA RAM addresses 0 through 31 0x1480 0 Receive Input AGC2 Gain RAM addresses 0 through 31 0x14A0 0 Receive Input AGC2 Gain RAM addresses 32 through 63 0x1500 0 Receive Input AGC3 Error RAM addresses 0 through 31 0x1520 0 Receive Input AGC3 Error RAM addresses 32 through 63 0x1540 0 Receive Input AGC3 DVGA RAM addresses 0 through 31 0x1580 0 Receive Input AGC3 Gain RAM addresses 0 through 31 0x15A0 0 Receive Input AGC3 Gain RAM addresses 32 through 63 0x1800 0 Receive Input Control Registers 0x00 through 0x1F 0x1820 0 Receive Input Control Registers 0x20 through 0x3F 0x1840 0 Receive Input AGC Control Registers 0x00 through 0x1F 0x1860 0 Receive Input AGC Control Registers 0x20 through 0x3F 4.4.1 Registers Addressed With 5 Bit Address Space, Pins (a4:a0) Control Register Index REGISTER NAME Table 4-4. Control Register Index SECTION GBL_PAR_CONFIG0 Section 4.4.2.4 GBL_PAR_CONFIG1 Section 4.4.2.5 SECTION GBL_TRISTATE Section 4.4.2.6 AGC_AMAX Section 4.4.5.23 NZ_PWR_MASK Section 4.4.3.7 AGC_AMIN Section 4.4.5.24 PAGE Section 4.4.2.2 AGC_CONFIG1 Section 4.4.5.17 PHASE_OFFSETA Section 4.4.5.13 AGC_CONFIG2 Section 4.4.5.18 PHASE_OFFSETB Section 4.4.5.14 AGC_CONFIG3 Section 4.4.5.19 PHASEADD0A Section 4.4.5.9 AGC_GAINA Section 4.4.5.21 PHASEADD0B Section 4.4.5.11 AGC_GAINB Section 4.4.5.22 PHASEADD1A Section 4.4.5.10 AGC_GAINMSB Section 4.4.5.20 PHASEADD1B Section 4.4.5.12 CIC_MODE1 Section 4.4.5.5 PMETER_RESULT_A_LSB Section 4.4.5.32 CIC_MODE2 Section 4.4.5.6 PMETER_RESULT_A_MID Section 4.4.5.33 CONFIG Section 4.4.2.3 PMETER_RESULT_A_MSB Section 4.4.5.34 CONFIG1 Section 4.4.5.15 PMETER_RESULT_B_LSB Section 4.4.5.35 CONFIG2 Section 4.4.5.16 PMETER_RESULT_B_MID Section 4.4.5.36 DDC_CHK_SUM Section 4.4.5.31 PMETER_RESULT_B_MSB Section 4.4.5.37 DDCCONFIG1 Section 4.4.5.27 Section 4.4.5.38 FIR_GAIN Section 4.4.5.2 PMETER_RESULT_AB_UM SB FIR_MODE Section 4.4.5.1 PSER_CONFIG1 Section 4.4.5.25 GBL_IMASK0 Section 4.4.2.8 PSER_CONFIG2 Section 4.4.5.26 GBL_INTERRUPT0 Section 4.4.2.9 RAGC_CONFIG0 Section 4.4.4.1 GBL_ONESHOT Section 4.4.2.7 RAGC_CONFIG1 Section 4.4.4.2 REGISTER NAME GC5018 GENERAL CONTROL 51 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Table 4-4. Control Register Index (continued) REGISTER NAME SECTION RAGC3_ACCUM_LSB Section 4.4.4.63 Section 4.4.4.3 RAGC3_ACCUM_MSB Section 4.4.4.64 RAGC_CONFIG3 Section 4.4.4.4 RAGC3_CLIP_ERROR Section 4.4.4.56 RAGC0_ACCUM_LSB Section 4.4.4.57 RAGC3_CLIP_HITHRESH Section 4.4.4.51 Section 4.4.4.58 RAGC3_CLIP_HITIMER Section 4.4.4.53 RAGC0_CLIP_ERROR Section 4.4.4.17 RAGC3_CLIP_LOTHRESH Section 4.4.4.52 RAGC0_CLIP_HITHRESH Section 4.4.4.12 RAGC3_CLIP_LOTIMER Section 4.4.4.54 Section 4.4.4.14 RAGC3_CLIP_SAMPLES Section 4.4.4.55 RAGC0_CLIP_LOTHRESH Section 4.4.4.13 RAGC3_CONFIG0 Section 4.4.4.46 RAGC0_CLIP_LOTIMER Section 4.4.4.15 RAGC3_CONFIG1 Section 4.4.4.47 Section 4.4.4.16 RAGC3_INTEGINVL_LSB Section 4.4.4.44 Section 4.4.4.7 RAGC3_INTEGINVL_MSB Section 4.4.4.45 RAGC0_CONFIG1 Section 4.4.4.8 RAGC3_SD_SAMPLES Section 4.4.4.50 RAGC0_INTEGINVL_LSB Section 4.4.4.5 RAGC3_SD_THRESH Section 4.4.4.48 Section 4.4.4.6 RAGC3_SD_TIMER Section 4.4.4.49 RAGC0_SD_SAMPLES Section 4.4.4.11 RECV_CONFIG0 Section 4.4.3.5 RAGC0_SD_THRESH Section 4.4.4.9 RECV_CONFIG1 Section 4.4.3.6 Section 4.4.4.10 RECV_PMETER_SYNC Section 4.4.3.8 RAGC1_ACCUM_LSB Section 4.4.4.59 RECV_PMETER0_CONFIG Section 4.4.3.12 RAGC1_ACCUM_MSB Section 4.4.4.60 RECV_PMETER0_LMSB Section 4.4.3.28 Section 4.4.4.30 RECV_PMETER0_LSB Section 4.4.3.26 Section 4.4.4.25 RECV_PMETER0_MID Section 4.4.3.27 RAGC1_CLIP_HITIMER Section 4.4.4.27 RECV_PMETER0_SQR_SU M_LSB Section 4.4.3.9 RAGC1_CLIP_LOTHRESH Section 4.4.4.26 Section 4.4.4.28 RECV_PMETER0_STRT_IN TVL_LSB Section 4.4.3.10 RAGC1_CLIP_LOTIMER RAGC1_CLIP_SAMPLES Section 4.4.4.29 Section 4.4.4.20 RECV_PMETER0_SYNC_D LY Section 4.4.3.11 RAGC1_CONFIG0 RAGC1_CONFIG1 Section 4.4.4.21 RECV_PMETER0_UMSB Section 4.4.3.29 RAGC1_INTEGINVL_LSB Section 4.4.4.18 RECV_PMETER1_CONFIG Section 4.4.3.16 RAGC1_INTEGINVL_MSB Section 4.4.4.19 RECV_PMETER1_LMSB Section 4.4.3.32 Section 4.4.4.24 RECV_PMETER1_LSB Section 4.4.3.30 RAGC1_SD_THRESH Section 4.4.4.22 RECV_PMETER1_MID Section 4.4.3.31 RAGC1_SD_TIMER Section 4.4.4.23 Section 4.4.3.13 RAGC2_ACCUM_LSB Section 4.4.4.61 RECV_PMETER1_SQR_SU M_LSB Section 4.4.4.62 RECV_PMETER1_STRT_IN TVL_LSB Section 4.4.3.14 RAGC2_ACCUM_MSB RAGC2_CLIP_ERROR Section 4.4.4.43 Section 4.4.4.38 RECV_PMETER1_SYNC_D LY Section 4.4.3.15 RAGC2_CLIP_HITHRESH RAGC2_CLIP_HITIMER Section 4.4.4.40 RECV_PMETER1_UMSB Section 4.4.3.33 RAGC2_CLIP_LOTHRESH Section 4.4.4.39 RECV_PMETER2_CONFIG Section 4.4.3.20 RAGC2_CLIP_LOTIMER Section 4.4.4.41 RECV_PMETER2_LMSB Section 4.4.3.36 Section 4.4.4.42 RECV_PMETER2_LSB Section 4.4.3.34 RAGC2_CONFIG0 Section 4.4.4.33 RECV_PMETER2_MID Section 4.4.3.35 RAGC2_CONFIG1 Section 4.4.4.34 Section 4.4.3.17 RAGC2_INTEGINVL_LSB Section 4.4.4.31 RECV_PMETER2_SQR_SU M_LSB RAGC2_INTEGINVL_MSB Section 4.4.4.32 RECV_PMETER2_STRT_IN TVL_LSB Section 4.4.3.18 RAGC2_SD_SAMPLES Section 4.4.4.37 Section 4.4.4.35 RECV_PMETER2_SYNC_D LY Section 4.4.3.19 RAGC2_SD_THRESH RAGC2_SD_TIMER Section 4.4.4.36 RECV_PMETER2_UMSB Section 4.4.3.37 REGISTER NAME RAGC_CONFIG2 RAGC0_ACCUM_MSB RAGC0_CLIP_HITIMER RAGC0_CLIP_SAMPLES RAGC0_CONFIG0 RAGC0_INTEGINVL_MSB RAGC0_SD_TIMER RAGC1_CLIP_ERROR RAGC1_CLIP_HITHRESH RAGC1_SD_SAMPLES RAGC2_CLIP_SAMPLES 52 GC5018 GENERAL CONTROL SECTION GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Table 4-4. Control Register Index (continued) REGISTER NAME SECTION SSEL_DDC_CNTR Section 4.4.3.3 Section 4.4.3.24 SSEL_RX_0 Section 4.4.3.4 RECV_PMETER3_LMSB Section 4.4.3.40 STRT_INTRVL Section 4.4.5.4 RECV_PMETER3_LSB Section 4.4.3.38 SYNC_0 Section 4.4.5.28 Section 4.4.3.39 SYNC_1 Section 4.4.5.29 RECV_PMETER3_SQR_SU M_LSB Section 4.4.3.21 SYNC_2 Section 4.4.5.30 SYNC_DDC_CNTR_LSB Section 4.4.3.1 RECV_PMETER3_STRT_IN TVL_LSB Section 4.4.3.22 SYNC_DDC_CNTR_MSB Section 4.4.3.2 TADJC Section 4.4.5.7 RECV_PMETER3_SYNC_D LY Section 4.4.3.23 TADJF Section 4.4.5.8 VER Section 4.4.2.1 REGISTER NAME RECV_PMETER3_CONFIG RECV_PMETER3_MID SECTION RECV_PMETER3_UMSB Section 4.4.3.41 RECV_SLF_TST_VALUE Section 4.4.3.25 SQR_SUM Section 4.4.5.3 GC5018 GENERAL CONTROL 53 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.2 Global Control Variables These registers are accessed directly without page address extension; when pin a5 is high during a read or write access, this block of 32 registers are accessed. 4.4.2.1 VER Register Register name: VER 54 Address: 0x20 READ_ONLY BIT 15 unused unused unused unused unused unused unused BIT 8 unused 0 0 0 0 0 0 0 0 BIT 7 unused unused unused unused VER3 VER2 VER1 BIT 0 VER0 0 0 0 0 0 0 * * VER(3:0): A hardwired read only register that * valid version codes are "0001" and "0010" GC5018 GENERAL CONTROL returns the version of the chip. GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.2.2 PAGE Register Register name: PAGE Address: 0x21 BIT 15 unused unused unused 0 0 0 0 0 BIT 7 Y(1) Y(0) Zp unused 0 0 0 0 W(2:0) : Selects which dual DDC block to address. X: The DDC modules are configured as dual DDCs; an even numbered DDC and odd numbered DDC are contained in each dual DDC module, the X bit selects which DDC gets address. (DDC0/2/4/6=0, DDC1/3/5/7=1) Y(2:0) : X BIT 8 Y(2) 0 0 0 unused unused unused BIT 0 unused 0 0 0 0 W(2:0) W(2:0) X bit Selected Block 000 0 DDC0 000 1 DDC1 001 0 DDC2 001 1 DDC3 010 0 DDC4 010 1 DDC5 011 0 DDC6 011 1 DDC7 100 0 Receive AGC0/1 RAMs 101 0 Receive AGC2/3 RAMs 110 0 Receive Input Interface Within each major block, there are up to 8 different Zones that can be addressed using the Y bits. Y(2:0) DDC Zone Receive Input Interface Zone Receive AGC RAMs Zone 000 PFIR coeffient lower 2 bits CHIPS control registers RAGC0/2 ERRMAP 001 PFIR coeffient upper 16 bits RAGC control registers RAGC0/2 DVGAMAP 010 CFIR coeffient lower 2 bits Not assigned RAGC0/2 GAINMAP 011 CFIR coeffient upper 16 bits Not assigned Not assigned 100 Control registers Not assigned RAGC1/3 ERRMAP 101 Not assigned Not assigned RAGC1/3 DVGAMAP 110 Not assigned Not assigned RAGC1/3 GAINMAP 111 Not assigned Not assigned Not assigned Zp : The Zp bit is the MSB of the address word sent to the registers and rams. This bit can be thought of as an upper/lower selector of the 64 word addressing. GC5018 GENERAL CONTROL 55 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.2.3 CONFIG Register Register name: CONFIG Address: 0x22 BIT 15 slf_ tst_ena rduz_sens_ena arst_ func BIT 8 0 0 0 BIT 7 par_recv_ena gbl_ ddc_write intr_ clr 0 0 0 tst_rate_sel(4:0) 0 0 0 0 BIT 0 tst_ on tst_select(3:0) 1 1 0 1 0 0 slf_tst_ena : Turns on the checksum LFSR for the receivers. They are located in the RECEIVE INPUT INTERFACE and DDC blocks rduz_sens_ena : When enabled, adds noise to the LSB's of the ADC inputs. arst_func : When asserted, resets the functional portion of the circuits. The MPU registers do not get reset and retain their programmed value tst_rate_sel(4:0) : Sets the rate of the output test data and clock. The length of the clock cycle is the value in tst_rate_sel+1 multiplied by the RXCLK period. par_recv_ena : When asserted, the rxout_*_* serial pins join to form a 32 bit parallel output using 32 pins as a data bus, one pin as a output clock and one pin as a sync. This is used to connect to the TCI110 Chip rate processor from TI. gbl_ddc_write : When asserted, the mpu writes are global. This means that DDC0/2/4/6 or DDC1/3/5/7 can be programmed simultaneously with the same values. This is an effort to reduce the amount of time spent programming the device. A common setup can be used to program the DDC0/2/4/6, then all the DDC1/3/5/7. Afterwards, just individual writes to the registers which differ between DDCs can be done. To use this feature, this bit must be asserted and the DDC0/1 must be addressed. Any other DDC address will not work. intr_clr : When asserted, this bit forces all interrupts to be cleared. To allow the interrupts to be set again, this bit must be programmed to zero. This does not stop blocks from generating interrupts, but rather just keeps the interrupts from being reported. tst_select(3:0) : This selects which block the test output comes from: tst_on : 56 tst_select(3:0) Test Data Sent to Output 0000 DDC 0 0001 DDC 1 0010 DDC 2 0011 DDC 3 0100 DDC 4 0101 DDC 5 0110 DDC 6 0111 DDC 7 1000 rxin_a and rxin_b FIFO outputs others none selected When asserted, the testbus is active. The ADC input ports rxin_c(15:0), rxin_d(15:0), dvga_c(5:0) and dvga_d(5:0) become the testbus output ports. When this bit is set, the rxin_c(15:0) and rxin_d(15:0) ports become chip outputs. The dvga_c(5:0) and dvga_d(5:0) ports are enabled separately using the GBL_TRISTATE register GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.2.4 GBL_PAR_CONFIG0 Register Register name: GBL_PAR_CONFIG0 Address: 0x23 BIT 15 BIT 8 tst_clk_pol par_recv_sync_del(6:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 par_recv_ rxclk_pol par_recv_clkdiv(6:0) 0 0 0 0 0 0 0 0 0 par_recv_sync_del(6:0) : Delays the sync source from the DDC0 AGC output by (par_recv_sync_del+1) rxclk cycles. tst_clk_pol : Selects the polarity of the test clock output at dvga_c(1) when the test bus is enabled; 0 for rising edge in the center of valid data, 1 for falling edge in the center of valid data. No effect when tst_rate_sel is "00000". par_recv_clkdiv(6:0) : Selects the parallel interface output clock rate. par_recv_rxclk_pol : Selects the polarity of the rxclk_out clock output; 0 for rising edge in the center of valid data, 1 for falling edge in the center of valid data. 4.4.2.5 GBL_PAR_CONFIG1 Register Register name: GBL_PAR_CONFIG1 Address: 0x24 BIT 15 BIT 8 par_recv_syncout_del(3:0) 0 0 0 par_recv_chan(3:0) 0 0 0 0 BIT 7 BIT 0 par_recv_ sync_pol par_recv_fsinvl(6:0) 0 0 0 0 0 0 0 0 0 par_recv_syncout_del(3:0) : Changes the rx_sync_out position with respect to IQ DDC0. Setting to 0 causes rx_sync_out to lead IQ DDC0 by 1 output sample, setting to 1 causes rx_sync_out to line up with IQ DDC0, setting to 2 causes rx_sync_out to trail IQ DDC0 by 1 output sample, etc. par_recv_chan(3:0) : Selects the number of channels to be output over the parallel interface, from 1 to 16 channels. par_recv_fsinvl(6:0) : Selects the number of rxclk cycles per parallel interface frame, from 1 to 128 cycles. par_recv_sync_pol : Selects the polarity of the parallel interface sync pulse; 0 for active low, 1 for active high. GC5018 GENERAL CONTROL 57 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.2.6 GBL_TRISTATE Register Register name: GBL_TRISTATE Address: 0x25 BIT 15 rxclk_ena unused unused unused unused tristate(10) tristate(9) BIT 8 tristate(8) 1 0 0 0 0 1 1 1 BIT 7 tristate(7) tristate(6) tristate(5) tristate(4) tristate(3) tristate(2) tristate(1) 1 1 1 1 1 1 1 BIT 0 1 rxclk_ena : Master rxclk enable. When set, the chip's rxclk is enabled, when cleared, rxclk is disabled. All tristates are ACTIVE LOW so a `0' turns on the output and a `1' tristates it. tristate(10) : This bit turns on the dvga_d outputs. tristate(9) : This bit turns on the dvga_c outputs. tristate(8) : This bit turns on the dvga_b outputs. tristate(7) : This bit turns on the dvga_a outputs. tristate(6) : This bit turns on the rx_sync_out_6/7, and the rxout_6/7_a/b/c/d outputs. tristate(5) : This bit turns on the rx_sync_out_4/5, and the rxout_4/5_a/b/c/d outputs. tristate(4) : This bit turns on the rx_sync_out_2/3, and the rxout_2/3_a/b/c/d outputs. tristate(3) : This bit turns on the rx_sync_out_0/1, and the rxout_0/1_a/b/c/d outputs. tristate(2) : unused tristate(1) : rxclk_out tristate(0) : interrupt, and rx_sync_out. 4.4.2.7 GBL_ONESHOT Register Register name: GBL_ONESHOT Address: 0x26 BIT 15 unused unused unused unused unused unused unused BIT 8 unused 0 0 0 0 0 0 0 0 BIT 7 rx_oneshot unused unused unused unused unused unused BIT 0 unused 0 0 0 0 0 0 0 0 rx_oneshot : When set, a one shot pulse is sent to the receive blocks for syncing. This only works if the blocks are programmed to use the oneshot as the sync source. To use the oneshot again, it must be programmed back to a `0' and then back to a `1'. 58 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.2.8 GBL_IMASK0 Register Register name: GBL_IMASK0 Address: 0x27 BIT 15 pmeter7_im pmeter6_im pmeter5_im pmeter4_im pmeter3_im pmeter2_im pmeter1_im BIT 8 pmeter0_im 0 0 0 0 0 0 0 0 BIT 7 recv_ pmeter0_im recv_ pmeter1_im recv_ pmeter2_im recv_ pmeter3_im rxin_a_ ovr_im rxin_b_ ovr_im rxin_c_ ovr_im BIT 0 rxin_d_ ovr_im 0 0 0 0 0 0 0 0 pmeterX_im : When asserted, masks the interrupt for the particular DDC pmeter, X= {0,1,2,3,4,5,6,7}. recv_pmeterX_im : When asserted, masks the interrupt for the particular receive input pmeter, X= {0,1,2,3 }. rxin_X_ovr_im : When asserted, masks the interrupt for the particular rxin overflow, X={a,b,c,d}. 4.4.2.9 GBL_INTERRUPT0 Register Register name: GBL_INTERRUPT0 Address: 0x29 BIT 15 pmeter7 pmeter6 pmeter5 pmeter4 pmeter3 pmeter2 pmeter1 BIT 8 pmeter0 0 0 0 0 0 0 0 0 BIT 7 recv_ pmeter0 recv_ pmeter1 recv_ pmeter2 recv_ pmeter3 rxin_a_ovr rxin_b_ovr rcin_c_ovr BIT 0 rxin_d_ovr 0 0 0 0 0 0 0 0 pmeterX : Asserted when an interrupt has been generated by this DDC pmeterX block, X={1,2,3,4,5,6,7 recv_pmeterX : Asserted when an interrupt has been generated by this receive input pmeter, X= {0,1,2,3 }. rxin_X_ovr : Asserted when a logic high input from the rxin_X_ovr pin occurs, X={a,b,c,d}. GC5018 GENERAL CONTROL 59 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3 Receive Input Interface Controls 4.4.3.1 SYNC_DDC_CNTR_LSB Register Register name: SYNC_DDC_CNTR_LSB Page: 0x1800 Address: 0x00 BIT 15 BIT 8 ddc_counter(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ddc_counter(7:0) 0 4.4.3.2 0 0 0 0 0 0 0 SYNC_DDC_CNTR_MSB Register name: SYNC_DDC_CNTR_MSB Page: 0x1800 Address: 0x01 BIT 15 BIT 8 ddc_counter(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ddc_counter(23:16) 0 0 0 0 0 0 0 0 ddc_counter(32:0) : 32 bit interval timer common to all DDC sync inputs. This timer may be programmed to any interval count, and each DDC synchronization input can select this counter as a source. The value programmed into the counter is: (desired number -1). The counter increments on each RX clock rising edge. 60 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.3 SSEL_DDC_CNTR Register Register name: SSEL_DDC_CNTR Page: 0x1800 BIT 15 rxinab_mux rxincd_mux unused unused unused 0 0 0 0 0 Address: 0x02 BIT 8 ssel_ddc_counter(2:0) 0 0 BIT 7 0 BIT 0 ddc_counter_width(7:0) 0 0 0 0 0 0 0 0 rxinab_mux : When asserted, the rxin_a and rxin_b inputs are internally driven by the rxin_c and rxin_d ports, respectively (Factory test use only). rxincd_mux : When asserted, the rxin_c and rxin_d inputs are internally driven by the rxin_a and rxin_b ports, respectively (Factory test use only). ssel_ddc_counter(2:0) : Selects the sync source for the DDC sync counter. ddc_counter_width(7:0) : Sets the width of the counter generated sync pulse in RX clock cycles, from 1 to 256. Sync sources are contained in this and many of the following registers. For all sync source selections: ssel_ddc_XXXXX(2:0) Selected Sync Source 000 rxsyncA 001 rxsyncB 010 rxsyncC 011 rxsyncD 100 DDC sync counter 101 one shot (register write triggered) 110 always 0 111 always 1 GC5018 GENERAL CONTROL 61 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.4 SSEL_RX_0 Register Register name: SSEL_RX_0 BIT 15 unused 0 Address: 0x03 BIT 8 ssel_adc_fifo(2:0) 0 BIT 7 unused 0 Page: 0x1800 0 unused 0 0 ssel_tst_decim(2:0) 0 0 0 BIT 0 ssel_rxsync_out(2:0) 0 0 unused 0 0 ssel_ddc(2:0) 0 0 0 ssel_adc_fifo(2:0) : Selects the sync source for the adc FIFO blocks. Sync reinitializes the read and write pointers of the FIFO. ssel_tst_decim(2:0) : Selects the sync source for the test bus decimator block. ssel_rxsync_out(2:0) : Selects the sync source for the RXSYNC_OUT pin. ssel_ddc(2:0) : Selects the sync source for the DDC data input mux and mixer. Controls clock generation in each DDC block (before the CIC input) which must match because the FIFO output clock is common for all DDC blocks. 62 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.5 RECV_CONFIG0 Register Register name: RECV_CONFIG0 BIT 15 rate_sel(1:0) 0 0 Page: 0x1800 Address: 0x04 adc_ fifo_strap_cd self_test_ const_ena adc_ fifo_bypass ragc_mpu_ram _read 0 0 0 0 0 0 pmeter3_iq pmeter2_iq pmeter1_iq BIT 0 pmeter0_iq 0 0 0 0 BIT 7 tst_decim_delay(3:0) 0 0 BIT 8 tst_ decim17 adc_ fifo_strap_ab 0 0 rate_sel(1:0) : Tells the RECV_CDRV the input rate. This is the rxin_a/b/c/d input rate and the rate that the RECEIVE INPUT INTERFACE block sends data to the DDCs. rate_sel Input clock rate 00 rxclk 01 rxclk/2 10 rxclk/4 11 rxclk/8 adc_fifo_strap_ab : When asserted, the input pointers of the rxin_a FIFO and rxin_b FIFO are hooked together in lock step configuration. This is used for maintaining FIFO delay consistency when complex inputs are driven on rxin_a(I) and rxin_b(Q). rxin_a is the Master. adc_fifo_strap_cd : When asserted, the input pointers of the rxin_c FIFO and rxin_d FIFO are hooked together in lock step configuration. This is used for maintaining FIFO delay consistency when complex inputs are driven on rxin_c(I) and rxin_d(Q). rxin_c is the Master. self_test_const_ena : When asserted, (with slf_tst_ena also asserted), a constant value is output by the test and noise generator instead of the pseudo random sequence. The constant value is programmable. adc_fifo_bypass : When asserted, the ADC FIFO circuits are bypassed. Input data is then clocked in directly using the rxclk input. The ssel_ddc selection value will control the location of the internally generated sample clock when this bit is asserted where rate_sel is rxclk/2, rxclk/4 or rxclk/8. ragc_mpu_ram_read : When asserted, the RAMs in the RAGC blocks can be read. This bit should only be set when reading the RAGC map rams via the mpu interface and must be cleared for proper RAGC operation. tst_decim17 : When set, the decimation factor of the tst_decimator block is 17X. When cleared, the decimation factor is 1X (no decimation). tst_decim_delay(3:0) : These bits set the delay from the sync occurring until the decimator samples. In other words, the moment of the decimator is set by this delay value. pmeter3_iq : When asserted, the pmeter3 block takes input from both rxin_c and rxin_d as a complex sample pair. When de-asserted, only input from rxin_d is used for the power measurement. pmeter2_iq : When asserted, the pmeter2 block takes input from both rxin_c and rxin_d as a complex sample pair. When de-asserted, only input from rxin_c is used for the power measurement. pmeter1_iq : When asserted, the pmeter1 block takes input from both rxin_a and rxin_b as a complex sample pair. When de-asserted, only input from rxin_b is used for the power measurement. pmeter0_iq : When asserted, the pmeter0 block takes input from both rxin_a and rxin_b as a complex sample pair. When de-asserted, only input from rxin_a is used for the power measurement. GC5018 GENERAL CONTROL 63 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.6 RECV_CONFIG1 Register Register name: RECV_CONFIG1 Page: 0x1800 Address: 0x05 BIT 15 msb_pos_d(2:0) 0 0 offset_bin_d 0 0 BIT 8 offset_bin_c msb_pos_c(2:0) 0 0 0 BIT 7 msb_pos_b(2:0) 0 0 offset_bin_b 0 0 BIT 0 offset_bin_a msb_pos_a(2:0) 0 0 0 0 0 msb_pos_X(2:0) : Places the MSB of the input word from the ADC. The value programmed into the 3 bits is the number of bit positions to the left of bit16 in the input word, that the MSB is located. For example, if a 14bit input word is driving rxin_a input and is aligned with rxin_a_0, then msb_pos_a is programmed to "010" meaning 2 bits shifted down from bit 16 is the MSB. X={a,b,c,d}. offset_bin_X : rxin_X input data is in offset binary and not twos complement. If set, the input value will be converted to 2s complement using the MSB from the corresponding msb_pos_X value. X={a,b,c,d} 4.4.3.7 NZ_PWR_MASK Register Register name: NZ_PWR_MASK Page: 0x1800 Address: 0x06 BIT 15 BIT 8 nz_pwr_mask (15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 nz_pwr_mask (7:0) 0 0 0 0 0 0 0 0 nz_pwr_mask(15:0) : Used with the rduz_sens_ena and selects the noise bits to be added to the ADC input sample when asserted. 64 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.8 RECV_PMETER_SYNC Register Register name: RECV_PMETER_SYNC Page: 0x1800 BIT 15 recv_pmeter0_ ena recv_pmeter1_ ena 0 BIT 8 ssel_recv_pmeter0(2:0) 0 BIT 7 recv_pmeter2_ ena 0 Address: 0x07 0 0 0 ssel_ recv_pmeter1(2:0) 0 0 0 BIT 0 ssel_ recv_pmeter2(2:0) 0 0 recv_pmeter3_ ena 0 0 ssel_ recv_pmeter3(2:0) 0 0 0 recv_pmeter0_ena : Enables the Receive Input Interface pmeter0 block when set recv_pmeter1_ena : Enables the Receive Input Interface pmeter1 block when set recv_pmeter2_ena : Enables the Receive Input Interface pmeter2 block when set recv_pmeter3_ena : Enables the Receive Input Interface pmeter3 block when set ssel_ recv_pmeter0(2:0) : Selects the sync source for the Receive Input Interface pmeter0 block ssel_ recv_pmeter1(2:0) : Selects the sync source for the Receive Input Interface pmeter1 block ssel_ recv_pmeter2(2:0) : Selects the sync source for the Receive Input Interface pmeter2 block ssel_ recv_pmeter3(2:0) : Selects the sync source for the Receive Input Interface pmeter3 block GC5018 GENERAL CONTROL 65 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.9 RECV_PMETER0_SQR_SUM_LSB Register Register name: RECV_PMETER0_SQR_SUM_LSB Page: 0x1800 Address: 0x08 BIT 15 BIT 8 recv_pmeter0_sqr_sum (15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter0_sqr_sum (7:0) 0 0 0 0 0 0 0 0 recv_pmeter0_sqr_sum(15:0) : The sqr_sum register controls the number of samples to accumulate for a power measurement. Ia is (or Ia & Qa if complex mode is selected are) squared and accumulated. Eight Ia samples (or eight sample pairs of Ia and Qa samples) equal to one sqr_sum count. The accumulation interval is initiated when the sync is asserted and the programmed (8*sync_delay+2) samples has expired or when the interval start time is reached. When the (8*sqr_sum+1) sample time is reached, the accumulated powers are made available for MPU access and an interrupt is generated. 4.4.3.10 RECV_PMETER0_STRT_INTVL_LSB Register Register name: RECV_PMETER0_STRT_INTVL_LSB Page: 0x1800 Address: 0x09 BIT 15 BIT 8 recv_pmeter0_strt_intrvl (15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter0_strt_intrvl (7:0) 0 0 0 0 0 0 recv_pmeter0_strt_intrvl(15:0) : The start interval timer is the interval over restarted. The timer value is (8*strt_intrvl + 1) samples and (8*sqr_sum+1) samples. The interval start counter and RMS power at the sync pulse after the programmed delay and every time the reaches its limit. 66 GC5018 GENERAL CONTROL 0 0 which the sqr_sum is must be larger than accumulation is started STRT_INTRVL counter GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.11 RECV_PMETER0_SYNC_DLY Register Register name: RECV_PMETER0_SYNC_DLY Page: 0x1800 Address: 0x0A BIT 15 delay_line_0(5:0) 0 0 0 0 0 0 unused BIT 8 recv_pmeter0_ sync_delay(8) 0 0 BIT 7 BIT 0 recv_pmeter0_sync_delay (7:0) 0 0 0 0 0 0 0 0 delay_line_0(5:0) : Pointer offset for the rxin_a path variable delay line. Larger values result in larger pointer offsets and therefore more path delay. recv_pmeter0_sync_delay(8:0) : Programmable start delay from sync, in eight sample units. The actual value is (8*sync_delay + 2) samples. 4.4.3.12 RECV_PMETER0_CONFIG Register Register name: RECV_PMETER0_CONFIG Page: 0x1800 Address: 0x0B BIT 15 BIT 8 recv_pmeter0_strt_intrvl(20:18) recv_pmeter0_sqr_sum(20:16) 0 0 BIT 7 recv_pmeter0_strt_ intrvl(17:16) 0 0 0 0 0 unused unused unused 0 0 0 0 0 0 BIT 0 ssel_delay_line_0(2:0) 0 0 0 recv_pmeter0_sqr_sum(20:16) : MSBs of sqr_sum value, in 8 sample units recv_pmeter0_strt_intrvl(20:16) : MSBs of start interval value, in 8 sample units. ssel_delay_line_0(2:0) : Sync source selection for the 64 sample delay line pointer value update 4.4.3.13 RECV_PMETER1_SQR_SUM_LSB Register Register name: RECV_PMETER1_SQR_SUM_LSB Page: 0x1800 Address: 0x0C BIT 15 BIT 8 recv_pmeter1_sqr_sum (15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter1_sqr_sum (7:0) 0 0 0 0 0 0 0 0 recv_pmeter1_sqr_sum(15:0) : Lower 16bits of the sqr_sum interval timer, in 8 sample units. GC5018 GENERAL CONTROL 67 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.14 RECV_PMETER1_STRT_INTVL_LSB Register Register name: RECV_PMETER1_STRT_INTVL_LSB Page: 0x1800 Address: 0x0D BIT 15 BIT 8 recv_pmeter1_strt_intrvl (15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter1_strt_intrvl (7:0) 0 0 0 0 0 0 0 0 recv_pmeter1_strt_intrvl(15:0) : Lower 16bits of the interval timer, in 8 sample units. 4.4.3.15 RECV_PMETER1_SYNC_DLY Register Register name: RECV_PMETER1_SYNC_DLY Page: 0x1800 Address: 0x0E BIT 15 delay_line_1(5:0) 0 0 0 0 0 0 unused BIT 8 recv_pmeter1_ sync_ delay(8) 0 0 BIT 7 BIT 0 recv_pmeter1_sync_delay (7:0) 0 0 0 0 0 0 0 0 delay_line_1(5:0) : Pointer offset for the rxin_b path variable delay line. Larger values result in larger pointer offsets and therefore more path delay recv_pmeter1_sync_delay(8:0) : Programmable start delay from sync, in 8 sample units. 68 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.16 RECV_PMETER1_CONFIG Register Register name: RECV_PMETER1_CONFIG Page: 0x1800 Address: 0x0F BIT 15 BIT 8 recv_pmeter1_strt_intrvl(20:18) recv_pmeter1_sqr_sum(20:16) 0 0 BIT 7 recv_pmeter1_strt_ intrvl(17:16) 0 0 0 0 0 unused unused unused 0 0 0 0 0 0 BIT 0 ssel_delay_line_1(2:0) 0 0 0 recv_pmeter1_sqr_sum(20:16) : MSBs of sqr_sum value, in 8 sample units. recv_pmeter1_strt_intrvl(20:16) : MSBs of start interval value, in 8 sample units. ssel_delay_line_1(2:0) : Sync source selection for the 64 sample delay line pointer value update 4.4.3.17 RECV_PMETER2_SQR_SUM_LSB Register Register name: RECV_PMETER2_SQR_SUM_LSB Page: 0x1800 Address: 0x10 BIT 15 BIT 8 recv_pmeter2_sqr_sum (15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter2_sqr_sum (7:0) 0 0 0 0 0 0 0 0 recv_pmeter2_sqr_sum(15:0) : Lower 16bits of the sqr_sum interval timer, in 8 sample units. GC5018 GENERAL CONTROL 69 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.18 RECV_PMETER2_STRT_INTVL_LSB Register Register name: RECV_PMETER2_STRT_INTVL_LSB Page: 0x1800 Address: 0X11 BIT 15 BIT 8 recv_pmeter2_strt_intrvl (15:8) 0 0 0 0 0 0 0 0 BIT 7 BIT 0 recv_pmeter2_strt_intrvl (7:0) 0 0 0 0 0 0 0 0 recv_pmeter2_strt_intrvl(15:0) : Lower 16bits of the interval timer, in 8 sample units. 4.4.3.19 RECV_PMETER2_SYNC_DLY Register Register name: RECV_PMETER2_SYNC_DLY Page: 0x1800 Address: 0x12 BIT 15 delay_line_2(5:0) 0 0 0 0 0 0 unused BIT 8 recv_pmeter2_ sync_ delay(8) 0 0 BIT 7 BIT 0 recv_pmeter2_sync_delay (7:0) 0 0 0 0 0 0 0 0 delay_line_2(5:0) : Pointer offset for the rxin_c path variable delay line. Larger values result in larger pointer offsets and therefore more path delay. recv_pmeter2_sync_delay (8:0) : Programmable start delay from sync, in 8 sample units. 4.4.3.20 RECV_PMETER2_CONFIG Register Register name: RECV_PMETER2_CONFIG Page: 0x1800 Address: 0X13 BIT 15 BIT 8 recv_pmeter2_strt_intrvl(20:18) recv_pmeter2_sqr_sum(20:16) 0 0 BIT 7 recv_pmeter2_strt_ intrvl(17:16) 0 0 0 0 0 unused unused unused 0 0 0 0 0 0 BIT 0 ssel_delay_line_2(2:0) 0 0 recv_pmeter2_sqr_sum(20:16) : MSBs of sqr_sum value, in 8 sample units. recv_pmeter2_strt_intrvl(20:16) : MSBs of start interval value, in 8 sample units. ssel_delay_line_2(2:0) : Sync source selection for the 64 sample delay line pointer value update. 70 GC5018 GENERAL CONTROL 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.21 RECV_PMETER3_SQR_SUM_LSB Register Register name: RECV_PMETER3_SQR_SUM_LSB Page: 0x1800 Address: 0x14 BIT 15 BIT 8 recv_pmeter3_sqr_sum (15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter3_sqr_sum (7:0) 0 0 0 0 0 0 0 0 recv_pmeter3_sqr_sum(15:0) : Lower 16bits of the sqr_sum interval timer, in 8 sample units. 4.4.3.22 RECV_PMETER3_STRT_INTVL_LSB Register Register name: RECV_PMETER3_STRT_INTVL_LSB Page: 0x1800 Address: 0x15 BIT 15 BIT 8 recv_pmeter3_strt_intrvl (15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter3_strt_intrvl (7:0) 0 0 0 0 0 0 0 0 recv_pmeter3_strt_intrvl(15:0) : Lower 16bits of the interval timer, in 8 sample units. 4.4.3.23 RECV_PMETER3_SYNC_DLY Register Register name: RECV_PMETER3_SYNC_DLY Page: 0x1800 Address: 0x16 BIT 15 delay_line_3(5:0) 0 0 0 0 0 0 unused BIT 8 recv_pme ter3_sync_ delay(8) 0 0 BIT 7 BIT 0 recv_pmeter3_sync_delay (7:0) 0 0 0 0 0 0 0 0 delay_line_3(5:0) : Pointer offset for the rxin_d path variable delay line. Larger values result in larger pointer offsets and therefore more path delay. recv_pmeter3_sync_delay(8:0) : Programmable start delay from sync, in 8 sample units. GC5018 GENERAL CONTROL 71 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.24 RECV_PMETER3_CONFIG Register Register name: RECV_PMETER3_CONFIG Page: 0x1800 Address: 0x17 BIT 15 BIT 8 recv_pmeter3_strt_intrvl(20:18) recv_pmeter3_sqr_sum(20:16) 0 0 BIT 7 recv_pmeter3_strt_ intrvl(17:16) 0 0 0 0 unused unused unused 0 0 0 0 0 0 BIT 0 0 ssel_delay_line_3(2:0) 0 0 0 recv_pmeter3_sqr_sum(20:16) : MSBs of sqr_sum value, in 8 sample units recv_pmeter3_strt_intrvl(20:16) : MSBs of start interval value, in 8 sample units ssel_delay_line_3(2:0) : Sync source selection for the 64 sample delay line pointer value update 4.4.3.25 RECV_SLF_TST_VALUE Register Register name: RECV_SLF_TST_VALUE Page: 0x1800 Address: 0x18 BIT 15 BIT 8 self_test_constant(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 self_test_constant(7:0) 0 0 0 0 0 0 0 0 self_test_constant(15:0) : 16 bit constant presented at the test and noise generator output when enabled. Used for test and debug purposes. 4.4.3.26 RECV_PMETER0_LSB Register Register name: RECV_PMETER0_LSB Page: 0x1820 Address: 0x20 READ ONLY BIT 15 BIT 8 recv_pmeter0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter0(7:0) 0 0 0 0 0 recv_pmeter0(15:0) : Lower bits of the power meter 0 measurement 72 GC5018 GENERAL CONTROL 0 0 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.27 RECV_PMETER0_MID Register Register name: RECV_PMETER0_MID Page: 0x1820 Address: 0x21 READ ONLY BIT 15 BIT 8 recv_pmeter0(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter0(23:16) 0 0 0 0 0 0 0 0 recv_pmeter0(31:16) : Mid bits of the power meter 0 measurement 4.4.3.28 RECV_PMETER0_LMSB Register Register name: RECV_PMETER0_LMSB Page: 0x1820 Address: 0x22 READ ONLY BIT 15 BIT 8 recv_pmeter0(47:40) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter0(39:32) 0 0 0 0 0 0 0 0 recv_pmeter0(47:32) : Lower MSB bits of the power meter 0 measurement 4.4.3.29 RECV_PMETER0_UMSB Register Register name: RECV_PMETER0_UMSB Page: 0x1820 Address: 0x23 BIT 15 unused unused unused unused unused unused 0 0 0 0 0 0 READ ONLY BIT 8 recv_pmeter0(57:56) 0 BIT 7 0 BIT 0 recv_pmeter0(55:48) 0 0 0 0 0 0 0 0 recv_pmeter0(57:48) : Upper MSB bits of the power meter 0 measurement GC5018 GENERAL CONTROL 73 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.30 RECV_PMETER1_LSB Register Register name: RECV_PMETER1_LSB Page: 0x1820 Address: 0x24 READ ONLY BIT 15 BIT 8 recv_pmeter1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter1(7:0) 0 0 0 0 0 0 0 0 recv_pmeter1(15:0) : Lower bits of the power meter 1 measurement 4.4.3.31 RECV_PMETER1_MID Register Register name: RECV_PMETER1_MID Page: 0x1820 Address: 0x25 READ ONLY BIT 15 BIT 8 recv_pmeter1(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter1(23:16) 0 0 0 0 0 0 0 0 recv_pmeter1(31:16) : Mid bits of the power meter 1 measurement 4.4.3.32 RECV_PMETER1_LMSB Register Register name: RECV_PMETER1_LMSB Page: 0x1820 Address: 0x26 READ ONLY BIT 15 BIT 8 recv_pmeter1(47:40) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter1(39:32) 0 0 0 0 0 0 recv_pmeter1(47:32) : Lower MSB bits of the power meter 1 measurement 74 GC5018 GENERAL CONTROL 0 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.33 RECV_PMETER1_UMSB Register Register name: RECV_PMETER1_UMSB Page: 0x1820 Address: 0x27 BIT 15 unused unused unused unused unused unused 0 0 0 0 0 0 READ ONLY BIT 8 recv_pmeter1(57:56) 0 BIT 7 0 BIT 0 recv_pmeter1(55:48) 0 0 0 0 0 0 0 0 recv_pmeter1(57:48) : Upper MSB bits of the power meter 1 measurement 4.4.3.34 RECV_PMETER2_LSB Register Register name: RECV_PMETER2_LSB Page: 0x1820 Address: 0x28 READ ONLY BIT 15 BIT 8 recv_pmeter2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter2(7:0) 0 0 0 0 0 0 0 0 recv_pmeter2(15:0) : Lower bits of the power meter 2 measurement 4.4.3.35 RECV_PMETER2_MID Register Register name: RECV_PMETER2_MID Page: 0x1820 Address: 0x29 READ ONLY BIT 15 BIT 8 recv_pmeter2(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter2(23:16) 0 0 0 0 0 0 0 0 recv_pmeter2(31:16) : Mid bits of the power meter 2 measurement GC5018 GENERAL CONTROL 75 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.36 RECV_PMETER2_LMSB Register Register name: RECV_PMETER2_LMSB Page: 0x1820 Address: 0x2A READ ONLY BIT 15 BIT 8 recv_pmeter2(47:40) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter2(39:32) 0 0 0 0 0 0 0 0 recv_pmeter2(47:32) : Lower MSB bits of the power meter 2 measurement 4.4.3.37 RECV_PMETER2_UMSB Register Register name: RECV_PMETER2_UMSB Page: 0x1820 Address: 0x2B READ ONLY BIT 15 unused unused unused unused unused unused 0 0 0 0 0 0 BIT 8 recv_pmeter2(57:56) 0 BIT 7 0 BIT 0 recv_pmeter2(55:48) 0 0 0 0 0 0 recv_pmeter2(57:48) : Upper MSB bits of the power meter 2 measurement 76 GC5018 GENERAL CONTROL 0 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.38 RECV_PMETER3_LSB Register Register name: RECV_PMETER3_LSB Page: 0x1820 Address: 0x2C READ ONLY BIT 15 BIT 8 recv_pmeter3(15:8) 0 0 0 0 0 0 0 BIT 7 0 0 BIT 0 0 0 0 0 0 0 0 recv_pmeter3(15:0) : Lower bits of the power meter 3 measurement 4.4.3.39 RECV_PMETER3_MID Register Register name: RECV_PMETER3_MID Page: 0x1820 Address: 0x2D READ ONLY BIT 15 BIT 8 recv_pmeter3(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter3(23:16) 0 0 0 0 0 0 0 0 recv_pmeter3(31:16) : Mid bits of the power meter 3 measurement 4.4.3.40 RECV_PMETER3_LMSB Register Register name: RECV_PMETER3_LMSB Page: 0x1820 Address: 0x2E READ_ONLY BIT 15 BIT 8 recv_pmeter3(47:40) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter3(39:32) 0 0 0 0 0 0 0 0 recv_pmeter3(47:32) : Lower MSB bits of the power meter 3 measurement GC5018 GENERAL CONTROL 77 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.3.41 RECV_PMETER3_UMSB Register Register name: RECV_PMETER3_UMSB BIT 15 unused unused 0 0 Page: 0x1820 Address: 0x2F READ_ONLY BIT 8 recv_pmeter3(57:56) 0 0 0 0 0 BIT 7 0 BIT 0 recv_pmeter3(55:48) 0 0 0 0 0 0 0 0 recv_pmeter3(57:48) : Upper MSB bits of the power meter 3 measurement 4.4.4 Receive AGC Controls 4.4.4.1 RAGC_CONFIG0 Register Register name: RAGC_CONFIG0 Page: 0x1840 Address: 0x00 BIT 15 hp_ena_0 hp_ena_1 hp_ena_2 hp_ena_3 sd_ena_0 sd_ena_1 sd_ena_2 BIT 8 sd_ena_3 0 0 0 0 0 0 0 0 unused unused unused BIT 0 unused 0 0 0 0 BIT 7 ragc_ bypass_0 ragc_ bypass_1 ragc_ bypass_2 ragc_ bypass_3 0 0 0 0 hp_ena_X : Enables the high pass filter in receive AGC X when set. sd_ena_X : Enables the Signal Detect block in receive AGC X when set. ragc_bypass_X : Bypasses the receive AGC X block when set. 78 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.2 RAGC_CONFIG1 Register Register name: RAGC_CONFIG1 BIT 15 ragc_ freeze_0 ragc_ freeze_1 ragc_ freeze_2 0 0 0 BIT 7 complex01 complex23 0 0 Page: 0x1840 Address: 0x01 ragc_ freeze_3 ragc_ clear_0 ragc_ clear_1 ragc_ clear_2 BIT 8 ragc_ clear_3 0 0 0 0 0 BIT 0 ssel_ragc_interval_0(2:0) 0 0 ssel_ragc_interval_1(2:0) 0 0 0 0 ragc_freeze_X : Freezes the receive AGC block when set. ragc_clear_X : Clears the loop error accumulator when set. complex01 : When set, receive AGC 0 uses complex input with the second sample stream coming from receive AGC 1. The clip detect, high pass, and squarer from receive AGC 1 are used to generate inputs for receive AGC 0. complex23 : When set, receive AGC 2 uses complex input with the second sample stream coming from receive AGC 3. The clip detect, high pass, and squarer from receive AGC 3 are used to generate inputs for receive AGC 2. ssel_ragc_interval_0(2:0) : Selects the sync source for receive AGC 0. After a programmed delay from sync, the interval update timer is started. ssel_ragc_interval_1(2:0) : Selects the sync source for receive AGC 1. After a programmed delay from sync, the interval update timer is started. 4.4.4.3 RAGC_CONFIG2 Register Register name: RAGC_CONFIG2 Page: 0x1840 Address: 0x02 BIT 15 ssel_ragc_freeze_0(2:0) 0 0 BIT 7 ssel_ragc_freez e_2(0) 0 BIT 8 ssel_ragc_ freeze_2(2:1) ssel_ragc_freeze_1(2:0) 0 0 0 0 0 0 BIT 0 ssel_ragc_freeze_3(2:0) 0 0 unused 0 0 ssel_ragc_interval_2(2:0) 0 0 0 ssel_ragc_freeze_X(2:0) : Selects the sync source that will freeze the receive AGC loop when asserted. ssel_ragc_interval_2(2:0) : Selects the sync source for receive AGC 2. After a programmed delay from sync, the interval update timer is started. GC5018 GENERAL CONTROL 79 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.4 RAGC_CONFIG3 Register Register name: RAGC_CONFIG3 Page: 0x1840 Address: 0x03 BIT 15 ssel_ragc_clear_0(2:0) 0 0 BIT 7 ssel_ragc_ clear_2(0) 0 BIT 8 ssel_ragc_ clear_2(2:1) ssel_ragc_clear_1(2:0) 0 0 0 0 0 0 BIT 0 ssel_ragc_clear_3(2:0) 0 0 unused 0 0 ssel_ragc_interval_3(2:0) 0 0 0 ssel_agc_clear_X(2:0 : Controls the selection of the sync that will clear the receive AGC error accumulator. ssel_agc_interval_3(2:0) : Selects the sync source for receive AGC 3. After a programmed delay from sync, the interval update timer is started. 4.4.4.5 RAGC0_INTEGINVL_LSB Register Register name: RAGC0_INTEGINVL_LSB Page: 0x1840 Address: 0x04 BIT 15 BIT 8 integ_interval_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 integ_interval_0(7:0) 0 0 0 0 0 0 0 0 integ_interval_0(15:0) : The 16 LSBs of the integration time for receive AGC 0. 4.4.4.6 RAGC0_INTEGINVL_MSB Register Register name: RAGC0_INTEGINVL_MSB Page: 0x1840 Address: 0x05 BIT 15 BIT 8 ragc_update_0(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 integ_interval_0(23:16) 0 0 0 0 0 0 0 ragc_update_0(7:0) : Sets the number of receive AGC updates per sync event (0x00 is infinite). integ_interval_0(23:16) : The eight MSBs of the integration time for receive AGC 0. 80 GC5018 GENERAL CONTROL 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.7 RAGC0_CONFIG0 Register Register name: RAGC0_CONFIG0 Page: 0x1840 Address: 0x06 BIT 15 BIT 8 ragc_sync_delay_0(7:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 hp_corner_0(2:0) 0 0 0 acc_shift_0(4:0) 0 0 0 0 0 0 ragc_sync_delay_0(7:0) : The input sync to the receive AGC block is delayed by this number of samples. hp_corner_0(2:0) : Sets the corner frequency of the high pass filter. Larger values result in higher corner frequencies acc_shift_0(4:0) : Selects the integrated power measurements result bits to be used as the error lookup table address. A larger number means fewer samples will have to be integrated to achieve the same result. 4.4.4.8 RAGC0_CONFIG1 Register Register name: RAGC0_CONFIG1 Page: 0x1840 Address: 0x07 BIT 15 BIT 8 err_shift_0(4:3) acc_offset_0(5:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 err_shift_0(2:0) 0 0 0 delay_adj_0(4:0) 0 0 0 0 0 0 acc_offset_0(5:0) : Constant subtracted from the integrated power measurement result before the error lookup table. err_shift_0(4:0) : Adjusts the loop gain by controlling the amount of shifting applied to the error lookup table output. Larger values result in higher gain. delay_adj_0(4:0) : Sets the delay difference, in samples, between the DVGA outputs and the value applied to the sample multiplier. GC5018 GENERAL CONTROL 81 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.9 RAGC0_SD_THRESH Register Register name: RAGC0_SD_THRESH Page: 0x1840 Address: 0x08 BIT 15 BIT 8 sd_thresh_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_thresh_0(7:0) 0 0 0 0 0 0 0 0 sd_thresh_0(15:0) : This is the threshold used by the Signal Detect block to determine if there is signal on the inputs. The comparison is done to the output of the squarer block, which is a 32 bit word. Because of this, these bits are aligned with bits 24 down to 8 of the 32 bit squared value. 4.4.4.10 RAGC0_SD_TIMER Register Register name: RAGC0_SD_TIMER Page: 0x1840 Address: 0x09 BIT 15 BIT 8 sd _timer_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd _timer_0(7:0) 0 0 0 0 0 0 0 0 sd_timer_0(15:0) : Qualification window timer for loss of input signal. 4.4.4.11 RAGC0_SD_SAMPLES Register Register name: RAGC0_SD_SAMPLES Page: 0x1840 Address: 0x0A BIT 15 BIT 8 sd_samples_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_samples_0(7:0) 0 0 0 0 0 0 0 0 sd_samples_0(15:0) : Number of samples that must be below the sd_thresh_X within the sd_timer_X timer value for the loss of signal condition to occur. 82 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.12 RAGC0_CLIP_HITHRESH Register Register name: RAGC0_CLIP_HITHRESH Page: 0x1840 Address: 0x0B BIT 15 BIT 8 clip_hi_thresh_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_hi_thresh_0(7:0) 0 0 0 0 0 0 0 0 clip_hi_thresh_0(15:0) : The high threshold value for clip detection. 4.4.4.13 RAGC0_CLIP_LOTHRESH Register Register name: RAGC0_CLIP_LOTHRESH Page: 0x1840 Address: 0x0C BIT 15 BIT 8 clip_lo_thresh_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_thresh_0(7:0) 0 0 0 0 0 0 0 0 clip_lo_thresh_0(15:0) : The low threshold value for clip detection. 4.4.4.14 RAGC0_CLIP_HITIMER Register Register name: RAGC0_CLIP_HITIMER Page: 0x1840 Address: 0x0D BIT 15 BIT 8 clip_hi_timer_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_hi_timer_0(7:0) 0 0 0 0 0 0 0 0 clip_hi_timer_0(15:0) : The high timer value in Samples GC5018 GENERAL CONTROL 83 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.15 RAGC0_CLIP_LOTIMER Register Register name: RAGC0_CLIP_LOTIMER Page: 0x1840 Address: 0x0E BIT 15 BIT 8 clip_lo_timer_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_timer_0(7:0) 0 0 0 0 0 0 0 0 clip_lo_timer_0(15:0) : The low timer value in Samples. 4.4.4.16 RAGC0_CLIP_SAMPLES Register Register name: RAGC0_CLIP_SAMPLES Page: 0x1840 Address: 0x0F BIT 15 BIT 8 clip_hi_samples_0(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_samples_0(7:0) 0 0 0 0 0 0 0 0 clip_hi_samples_0(7:0) : Number of samples above the high threshold within the clip high time to enable the clip event. clip_lo_samples_0(7:0) : Number of samples below the low threshold within the clip low time to disable the clip event. 4.4.4.17 RAGC0_CLIP_ERROR Register Register name: RAGC0_CLIP_ERROR Page: 0x1840 Address: 0x10 BIT 15 BIT 8 clip_error_0(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_error_0(7:0) 0 0 0 0 0 0 0 0 clip_error_0(15:0) : This is the error value that is added into the loop accumulator when a clip is detected. 84 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.18 RAGC1_INTEGINVL_LSB Register Register name: RAGC1_INTEGINVL_LSB Page: 0x1840 Address: 0x11 BIT 15 BIT 8 integ_interval_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 integ_interval_1(7:0) 0 0 0 0 0 0 0 0 integ_interval_1(15:0) : The LSBs of the integration time for receive AGC 1 4.4.4.19 RAGC1_INTEGINVL_MSB Register Register name: RAGC1_INTEGINVL_MSB Page: 0x1840 Address: 0x12 BIT 15 BIT 8 ragc_update_1(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 integ_interval_1(23:16) 0 0 0 0 0 0 0 0 ragc_update_1(7:0) : Sets the number of receive AGC updates per sync event (0x00 is infinite). integ_interval_1(23:16) : The MSBs of the integration time for receive AGC 1 4.4.4.20 RAGC1_CONFIG0 Register Register name: RAGC1_CONFIG0 Page: 0x1840 Address: 0x13 BIT 15 BIT 8 ragc_sync_delay_1(7:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 hp_corner_1(2:0) 0 0 0 acc_shift_1(4:0) 0 0 0 0 0 0 ragc_sync_delay_1(7:0) : The input sync to the receive AGC block is delayed by this value of samples. hp_corner_1(2:0) : This sets the corner frequency of the High Pass filter. Larger values result in higher corner frequencies. acc_shift_1(4:0) : Selects the integrated power measurements result bits to be used as the error lookup table address. A larger number means fewer samples will have to be integrated to achieve the same result. GC5018 GENERAL CONTROL 85 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.21 RAGC1_CONFIG1 Register Register name: RAGC1_CONFIG1 Page: 0x1840 Address: 0x14 BIT 15 BIT 8 err_shift_1(4:3) acc_offset_1(5:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 err_shift_1(2:0) 0 0 0 delay_adj_1(4:0) 0 0 0 0 0 0 acc_offset_1(5:0) : Constant subtracted from the integrated power measurement result before the error lookup table err_shift_1(4:0) : Controls the loop gain by left shifting the error output. Larger values result in higher gain. delay_adj_1(4:0) : Sets the delay difference, in samples, between the DVGA outputs and the value applied to the sample multiplier. 4.4.4.22 RAGC1_SD_THRESH Register Register name: RAGC1_SD_THRESH Page: 0x1840 Address: 0x15 BIT 15 BIT 8 sd_thresh_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_thresh_1(7:0) 0 0 0 0 0 0 0 0 sd_thresh_1(15:0) : This is the threshold used by the Signal Detect block to determine if there is signal on the inputs. The comparison is done to the output of the squarer block, which is a 32 bit word. Because of this, these bits are aligned with bits 24 down to 8 of the 32 bit squared value. 4.4.4.23 RAGC1_SD_TIMER Register Register name: RAGC1_SD_TIMER Page: 0x1840 Address: 0x16 BIT 15 BIT 8 sd_timer_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_timer_1(7:0) 0 0 0 0 0 0 0 0 sd_timer_1(15:0) : After the first no signal sample occurs, this is the amount of samples that control the length of time to determine the loss of signal condition. 86 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.24 RAGC1_SD_SAMPLES Register Register name: RAGC1_SD_SAMPLES Page: 0x1840 Address: 0x17 BIT 15 BIT 8 sd_samples_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_samples_1(7:0) 0 0 0 0 0 0 0 0 sd_samples_1(15:0) : Number of samples that must be below the sd_thresh_X threshold within the sd_timer_X timer value for the loss of signal condition to occur. 4.4.4.25 RAGC1_CLIP_HITHRESH Register Register name: RAGC1_CLIP_HITHRESH Page: 0x1840 Address: 0x18 BIT 15 BIT 8 clip_hi_thresh_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_hi_thresh_1(7:0) 0 0 0 0 0 0 0 0 clip_hi_thresh_1(15:0) : The high threshold value for clip detection. 4.4.4.26 RAGC1_CLIP_LOTHRESH Register Register name: RAGC1_CLIP_LOTHRESH Page: 0x1840 Address: 0x19 BIT 15 BIT 8 clip_lo_thresh_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_thresh_1(7:0) 0 0 0 0 0 0 0 0 clip_lo_thresh_1(15:0) The low threshold value for clip detection. GC5018 GENERAL CONTROL 87 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.27 RAGC1_CLIP_HITIMER Register Register name: RAGC1_CLIP_HITIMER Page: 0x1840 Address: 0x1A BIT 15 BIT 8 clip_hi_timer_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_hi_timer_1(7:0) 0 0 0 0 0 0 0 0 clip_hi_timer_1(15:0) : The high timer value in samples. 4.4.4.28 RAGC1_CLIP_LOTIMER Register Register name: RAGC1_CLIP_LOTIMER Page: 0x1840 Address: 0x1B BIT 15 BIT 8 clip_lo_timer_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_timer_1(7:0) 0 0 0 0 0 0 0 0 clip_lo_timer_1(15:0) : The low timer value in samples. 4.4.4.29 RAGC1_CLIP_SAMPLES Register Register name: RAGC1_CLIP_SAMPLES Page: 0x1840 Address: 0x1C BIT 15 BIT 8 clip_hi_samples_1(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_samples_1(7:0) 0 0 0 0 0 0 0 0 clip_hi_samples_1(7:0) : Number of samples above the high threshold within the clip high time to enable the clip event. clip_lo_samples_1(7:0) : Number of samples below the low threshold within the clip low time to disable the clip event. 88 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.30 RAGC1_CLIP_ERROR Register Register name: RAGC1_CLIP_ERROR Page: 0x1840 Address: 0x1D BIT 15 BIT 8 clip_error_1(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_error_1(7:0) 0 0 0 0 0 0 0 0 clip_error_1(15:0) : This is the error value that is added into the loop accumulator when a clip is detected. 4.4.4.31 RAGC2_INTEGINVL_LSB Register Register name: RAGC2_INTEGINVL_LSB Page: 0x1840 Address: 0x1E BIT 15 BIT 8 integ_interval_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 integ_interval_2(7:0) 0 0 0 0 0 0 0 0 integ_interval_2(15:0) : The LSBs of the integration time for receive AGC 2 4.4.4.32 RAGC2_INTEGINVL_MSB Register Register name: RAGC2_INTEGINVL_MSB Page: 0x1840 Address: 0x1F BIT 15 BIT 8 ragc_update_2(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 integ_interval_2(23:16) 0 0 0 0 0 0 0 0 ragc_update_2(7:0) : Sets the number of receive AGC updates per sync event (0x00 is infinite). integ_interval_2(23:16) : The MSBs of the integration time for receive AGC 2 GC5018 GENERAL CONTROL 89 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.33 RAGC2_CONFIG0 Register Register name: RAGC2_CONFIG0 Page: 0x1860 Address: 0x20 BIT 15 BIT 8 ragc_sync_delay_2(7:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 hp_corner_2(2:0) 0 0 0 acc_shift_2(4:0) 0 0 0 0 0 0 ragc_sync_delay_2(7:0) : The input sync to the receive AGC block is delayed by this value of samples. hp_corner_2(2:0) : This sets the corner frequency of the High Pass filter. Larger values result in higher corner frequencies. acc_shift_2(4:0) : Selects the integrated power measurements result bits to be used as the error lookup table address. A larger number means fewer samples will have to be integrated to achieve the same result. 4.4.4.34 RAGC2_CONFIG1 Register Register name: RAGC2_CONFIG1 Page: 0x1860 Address: 0x21 BIT 15 BIT 8 err_shift_2(4:3) acc_offset_2(5:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 err_shift_2(2:0) 0 0 0 delay_adj_2(4:0) 0 0 0 0 0 0 acc_offset_2(5:0) : Constant subtracted from the integrated power measurement result before the error lookup table. err_shift_2(4:0) : Controls the loop gain by left shifting the error output. Larger values result in higher gain.. delay_adj_2(4:0) : Sets the delay difference, in samples, between the DVGA outputs and the value applied to the sample multiplier. 90 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.35 RAGC2_SD_THRESH Register Register name: RAGC2_SD_THRESH Page: 0x1860 Address: 0x22 BIT 15 BIT 8 sd_thresh_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_thresh_2(7:0) 0 0 0 0 0 0 0 0 sd_thresh_2(15:0) : This is the threshold used by the Signal Detect block to determine if there is signal on the inputs. The comparison is done to the output of the squarer block, which is a 32 bit word. Because of this, these bits are aligned with bits 24 down to 8 of the 32 bit squared value. 4.4.4.36 RAGC2_SD_TIMER Register Register name: RAGC2_SD_TIMER Page: 0x1860 Address: 0x23 BIT 15 BIT 8 sd_timer_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_timer_2(7:0) 0 0 0 0 0 0 0 0 sd_timer_2(15:0) : After the first no signal sample occurs, this is the amount of samples that control the length of time to determine the loss of signal condition. 4.4.4.37 RAGC2_SD_SAMPLES Register Register name: RAGC2_SD_SAMPLES Page: 0x1860 Address: 0x24 BIT 15 BIT 8 sd_samples_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_samples_2(7:0) 0 0 0 0 0 0 0 0 sd_samples_2(15:0) : Number of samples that must be below the sd_thresh_X threshold within the sd_timer_X timer value for the loss of signal condition to occur. GC5018 GENERAL CONTROL 91 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.38 RAGC2_CLIP_HITHRESH Register Register name: RAGC2_CLIP_HITHRESH Page: 0x1860 Address: 0x25 BIT 15 BIT 8 clip_hi_thresh_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_hi_thresh_2(7:0) 0 0 0 0 0 0 0 0 clip_hi_thresh_2(15:0) : The high threshold value for clip detection. 4.4.4.39 RAGC2_CLIP_LOTHRESH Register Register name: RAGC2_CLIP_LOTHRESH Page: 0x1860 Address: 0x26 BIT 15 BIT 8 clip_lo_thresh_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_thresh_2(7:0) 0 0 0 0 0 0 0 0 clip_lo_thresh_2(15:0) : The low threshold value for clip detection. 4.4.4.40 RAGC2_CLIP_HITIMER Register Register name: RAGC2_CLIP_HITIMER Page: 0x1860 Address: 0x27 BIT 15 BIT 8 clip_hi_timer_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_hi_timer_2(7:0) 0 0 0 0 clip_hi_timer_2(15:0) : The high timer value in samples 92 GC5018 GENERAL CONTROL 0 0 0 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.41 RAGC2_CLIP_LOTIMER Register Register name: RAGC2_CLIP_LOTIMER Page: 0x1860 Address: 0x28 BIT 15 BIT 8 clip_lo_timer_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_timer_2(7:0) 0 0 0 0 0 0 0 0 clip_lo_timer_2(15:0) : The low timer value in samples. 4.4.4.42 RAGC2_CLIP_SAMPLES Register Register name: RAGC2_CLIP_SAMPLES Page: 0x1860 Address: 0x29 BIT 15 BIT 8 clip_hi_samples_2(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_samples_2(7:0) 0 0 0 0 0 0 0 0 clip_hi_samples_2(7:0) : Number of samples above the high threshold within the clip high time to enable the clip event. clip_lo_samples_2(7:0) : Number of samples below the low threshold within the clip low time to disable the clip event. 4.4.4.43 RAGC2_CLIP_ERROR Register Register name: RAGC2_CLIP_ERROR Page: 0x1860 Address: 0x2A BIT 15 BIT 8 clip_error_2(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_error_2(7:0) 0 0 0 0 0 0 0 0 clip_error_2(15:0) : This is the error value that is added into the loop accumulator when a clip is detected. GC5018 GENERAL CONTROL 93 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.44 RAGC3_INTEGINVL_LSB Register Register name: RAGC3_INTEGINVL_LSB Page: 0x1860 Address: 0x2B BIT 15 BIT 8 integ_interval_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 integ_interval_3(7:0) 0 0 0 0 0 0 0 0 integ_interval_3(15:0) : The LSBs of the integration time for receive AGC 3 4.4.4.45 RAGC3_INTEGINVL_MSB Register Register name: RAGC3_INTEGINVL_MSB Page: 0x1860 Address: 0x2C BIT 15 BIT 8 ragc_update_3(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 integ_interval_3(23:16) 0 0 0 0 0 0 0 0 ragc_update_3(7:0) : Sets the number of receive AGC updates per sync event (0x00 is infinite). integ_interval_3(23:16) : The MSBs of the integration time for receive AGC 3 4.4.4.46 RAGC3_CONFIG0 Register Register name: RAGC3_CONFIG0 Page: 0x1860 Address: 0x2D BIT 15 BIT 8 ragc_sync_delay_3(7:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 hp_corner_3(2:0) 0 0 0 acc_shift_3(4:0) 0 0 0 0 0 0 ragc_sync_delay_3(7:0) : The input sync to the receive AGC block is delayed by this value of samples. hp_corner_3(2:0) : This sets the corner frequency of the High Pass filter. Larger values result in higher corner frequencies. acc_shift_3(4:0) : Selects the integrated power measurements result bits to be used as the error lookup table address. A larger number means fewer samples will have to be integrated to achieve the same result. 94 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.47 RAGC3_CONFIG1 Register Register name: RAGC3_CONFIG1 Page: 0x1860 Address: 0x2E BIT 15 BIT 8 err_shift_3(4:3) acc_offset_3(5:0) 0 0 0 0 0 0 0 BIT 7 BIT 0 err_shift_3(2:0) 0 0 0 delay_adj_3(4:0) 0 0 0 0 0 0 acc_offset_3(5:0) : Constant subtracted from the integrated power measurement result before the error lookup table err_shift_3(4:0) : Controls the loop gain by left shifting the error output. Larger values result in higher gain. delay_adj_3(4:0) : Sets the delay difference, in samples, between the DVGA outputs and the value applied to the sample multiplier. 4.4.4.48 RAGC3_SD_THRESH Register Register name: RAGC3_SD_THRESH Page: 0x1860 Address: 0x2F BIT 15 BIT 8 sd_thresh_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_thresh_3(7:0) 0 0 0 0 0 0 0 0 sd_thresh_3(15:0) : This is the threshold used by the Signal Detect block to determine if there is signal on the inputs. The comparison is done to the output of the squarer block, which is a 32 bit word. Because of this, these bits are aligned with bits 24 down to 8 of the 32 bit squared value. 4.4.4.49 RAGC3_SD_TIMER Register Register name: RAGC3_SD_TIMER Page: 0x1860 Address: : 0x30 BIT 15 BIT 8 sd_timer_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_timer_3(7:0) 0 0 0 0 0 0 0 0 sd_timer_3(15:0) : After the first no signal sample occurs, this is the amount of samples that control the length of time to determine the loss of signal condition. GC5018 GENERAL CONTROL 95 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.50 RAGC3_SD_SAMPLES Register Register name: RAGC3_SD_SAMPLES Page: 0x1860 Address: 0x31 BIT 15 BIT 8 sd_samples_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 sd_samples_3(7:0) 0 0 0 0 0 0 0 0 sd_samples_3(15:0) : Number of samples that must be below the sd_thresh_X threshold within the sd_timer_X timer value for the loss of signal condition to occur. 4.4.4.51 RAGC3_CLIP_HITHRESH Register Register name: RAGC3_CLIP_HITHRESH Page: 0x1860 Address: 0x32 BIT 15 BIT 8 clip_hi_thresh_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_hi_thresh_3(7:0) 0 0 0 0 0 0 0 0 clip_hi_thresh_3(15:0) : The high threshold value for clip detection. 4.4.4.52 RAGC3_CLIP_LOTHRESH Register Register name: RAGC3_CLIP_LOTHRESH Page: 0x1860 Address: 0x33 BIT 15 BIT 8 clip_lo_thresh_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_thresh_3(7:0) 0 0 0 0 0 clip_lo_thresh_3(15:0) : The low threshold value for clip detection. 96 GC5018 GENERAL CONTROL 0 0 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.53 RAGC3_CLIP_HITIMER Register Register name: RAGC3_CLIP_HITIMER Page: 0x1860 Address: 0x34 BIT 15 BIT 8 clip_hi_timer_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_hi_timer_3(7:0) 0 0 0 0 0 0 0 0 clip_hi_timer_3(15:0) : The clip high timer value in samples 4.4.4.54 RAGC3_CLIP_LOTIMER Register Register name: RAGC3_CLIP_LOTIMER Page: 0x1860 Address: 0x35 BIT 15 BIT 8 clip_lo_timer_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_timer_3(7:0) 0 0 0 0 0 0 0 0 clip_lo_timer_3(15:0) : The clip low timer value in samples. 4.4.4.55 RAGC3_CLIP_SAMPLES Register Register name: RAGC3_CLIP_SAMPLES Page: 0x1860 Address: 0x36 BIT 15 BIT 8 clip_hi_samples_3(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_lo_samples_3(7:0) 0 0 0 0 0 0 0 0 clip_hi_samples_3(7:0) : Number of samples above the high threshold within the clip high time to enable a clip event. clip_lo_samples_3(7:0) : Number of samples below the low threshold within the clip low time to disable a clip event. GC5018 GENERAL CONTROL 97 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.56 RAGC3_CLIP_ERROR Register Register name: RAGC3_CLIP_ERROR Page: 0x1860 Address: 0x37 BIT 15 BIT 8 clip_error_3(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 clip_error_3(7:0) 0 0 0 0 0 0 0 0 clip_error_3(15:0) : Error value that is added into the loop accumulator when a clip is detected. 4.4.4.57 RAGC0_ACCUM_LSB Register Register name: RAGC0_ACCUM_LSB Page: 0x1860 Address: 0x38 READ ONLY BIT 15 BIT 8 ragc0_accum(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ragc0_accum (7:0) 0 0 0 0 0 0 0 0 ragc0_accum(15:0) : lower 16 bits of the ragc0 error accumulator. 4.4.4.58 RAGC0_ACCUM_MSB Register Register name: RAGC0_ACCUM_MSB Page: 0x1860 Address: 0x39 READ ONLY BIT 15 BIT 8 ragc0_accum(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ragc0_accum (23:16) 0 0 0 0 0 ragc0_accum(31:16) : upper 16 bits of the ragc0 error accumulator. 98 GC5018 GENERAL CONTROL 0 0 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.59 RAGC1_ACCUM_LSB Register Register name: RAGC1_ACCUM_LSB Page: 0x1860 Address: 0x3A READ ONLY BIT 15 BIT 8 ragc1_accum(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ragc1_accum (7:0) 0 0 0 0 0 0 0 0 ragc1_accum(15:0) : lower 16 bits of the ragc1 error accumulator. 4.4.4.60 RAGC1_ACCUM_MSB Register Register name: RAGC1_ACCUM_MSB Page: 0x1860 Address: 0x3B READ ONLY BIT 15 BIT 8 ragc1_accum(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ragc1_accum (23:16) 0 0 0 0 0 0 0 0 ragc1_accum(31:16) : upper 16 bits of the ragc1 error accumulator. 4.4.4.61 RAGC2_ACCUM_LSB Register Register name: RAGC2_ACCUM_LSB Page: 0x1860 Address: 0x3C READ ONLY BIT 15 BIT 8 ragc2_accum(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ragc2_accum (7:0) 0 0 0 0 0 0 0 0 ragc2_accum(15:0) : lower 16 bits of the ragc2 error accumulator. GC5018 GENERAL CONTROL 99 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.4.62 RAGC2_ACCUM_MSB Register Register name: RAGC2_ACCUM_MSB Page: 0x1860 Address: 0x3D READ ONLY BIT 15 BIT 8 ragc2_accum(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ragc2_accum (23:16) 0 0 0 0 0 0 0 0 ragc2_accum(31:16) : upper 16 bits of the ragc2 error accumulator. 4.4.4.63 RAGC3_ACCUM_LSB Register Register name: RAGC3_ACCUM_LSB Page: 0x1860 Address: 0x3E READ ONLY BIT 15 BIT 8 ragc3_accum(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ragc3_accum (7:0) 0 0 0 0 0 0 0 0 ragc3_accum(15:0) : lower 16 bits of the ragc3 error accumulator. 4.4.4.64 RAGC3_ACCUM_MSB Register Register name: RAGC3_ACCUM_MSB Page: 0x1860 Address: 0x3F READ ONLY BIT 15 BIT 8 ragc3_accum(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ragc3_accum (23:16) 0 0 0 0 0 ragc3_accum(31:16) : upper 16 bits of the ragc3 error accumulator. 100 GC5018 GENERAL CONTROL 0 0 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5 DDC Channel Controls 4.4.5.1 FIR_MODE Register Register name: FIR_MODE Page: 0x0%00 where: % = 2x(DDC channel #)+1 BIT 15 cdma_mode unused unused 0 0 0 Address: 0x00 BIT 8 crastarttap_pfir(4:0) 0 0 0 0 0 unused unused BIT 0 unused 0 0 0 BIT 7 crastarttap_cfir(4:0) 0 0 0 0 0 cdma_mode : When asserted the DDC block is in CDMA mode (2 streams per DDC block). crastarttap_pfir : These bits define the number of taps that PFIR will use for the filtering. crastarttap_cfir : These bits define the number of taps that CFIR will use for the filtering. Formulas for the number of taps, in the different FIR's, using the crastarttap word. DDC PFIR: 4*(crastarttap_pfir+1) DDC PFIR long mode: 8*(crastarttap_pfir+1) DDC CFIR: 2*(crastarttap_cfir+1) 4.4.5.2 FIR_GAIN Register Register name: FIR_GAIN Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x01 BIT 15 pfir_gain(2:0) 0 unused unused unused unused BIT 8 unused 0 0 0 0 0 0 0 unused unused unused unused unused unused unused unused 0 0 0 0 0 0 0 0 BIT 7 BIT 0 pfir_gain(2:0) : PFIR gain, from 2e-19 to 2e-12 for the receive PFIR. ("000" = 2e-19 and "111" = 2e-12) cfir_gain : When `0' then the gain of the CFIR is 2e-19, otherwise when set to `1' the gain is 2e-18. GC5018 GENERAL CONTROL 101 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.3 SQR_SUM Register Register name: SQR_SUM Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x02 BIT 15 BIT 8 pmeter_sqr_sum_ddc(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_sqr_sum_ddc(7:0) 0 0 0 0 0 0 0 0 pmeter_sqr_sum_ddc(15:0): The sqr_sum register is the number of 4 sample sets to accumulate for a power measurement. In CDMA mode, one sample set is the I & Q of the signal and diversity. Ia & Qa (signal) are each squared and accumulated and Ib & Qb (diversity) are squared and accumulated. In UMTS mode, each I and Q pair are squared and accumulated. 4 samples is equal to one SQR_SUM count. The count is initiated when the sync is asserted or when the interval start time is reached. When the SQR_NUM number is reached, the accumulated powers are made available for MPU access and an interrupt is generated. 4.4.5.4 STRT_INTRVL Register Register name: STRT_INTRVL Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x03 BIT 15 BIT 8 pmeter_sync_delay_ddc(7:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_interval_ddc(7:0) 0 0 0 0 0 0 0 0 pmeter_sync_delay_ddc(7:0) : The delay from selected sync source to when the power calculation starts. The actual value is sync_delay + 1. pmeter_interval_ddc(7:0) : The start interval timer is the interval over which the SQR_SUM is restarted and must be greater than the SQR_SUM. The actual interval is interval +1, and must be greater than the sqr_sum interval. The interval start counter and RMS power accumulation is started at the sync pulse after the programmed delay and every time the interval counter reaches its limit. This value is in 1024 sample units. 102 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.5 CIC_MODE1 Register Register name: CIC_MODE1 Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x04 BIT 15 BIT 8 cic_scale_a(4:0) 0 0 BIT 7 cic_scale_b(1:0) 0 0 cic_scale_b(4:2) 0 0 0 0 0 BIT 0 cic_gain_ ddc 0 0 cic_decim(4:0) 0 0 0 0 0 cic_scale_a(4:0) : This sets the gain shift at the output of the A channel CIC. 0x00 is no shift, each increment by 1 increases the signal amplitude by 2X. cic_scale_b(4:0) : This sets the gain shift at the output of the B channel CIC. 0x00 is no shift, each increment by 1 increases the signal amplitude by 2X. cic_gain_ddc : Adds a fixed gain of 12dB at the CIC output when asserted. cic_decim(4:0) : Sets the CIC decimation rate, where decimation is cic_decim + 1. 4.4.5.6 CIC_MODE2 Register Register name: CIC_MODE2 Page: 0x0%00 where: % = 2x(DDC channel #)+1 BIT 15 BIT 8 cic_m2_ena_b(5:4) cic_m2_ena_a(5:0) 0 0 0 0 Address: 0x05 0 0 0 0 unused unused unused BIT 0 unused 0 0 0 0 BIT 7 cic_m2_ena_b(3:0) 0 0 0 0 cic_m2_ena_a(5:0) : Programs the A channel CIC fir sections M value to 2 when set, 1 when cleared. cic_m2_ena_a(0) controls the M value for the first comb section and cic_m2_ena_a(5) controls the M value for the last comb section. cic_m2_ena_b(5:0) : Programs the B channel CIC fir sections M value to 2 when set, 1 when cleared. cic_m2_ena_b(0) controls the M value for the first comb section and cic_m2_ena_b(5) controls the M value for the last comb section. GC5018 GENERAL CONTROL 103 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.7 TADJC Register Register name: TADJC Page: 0x0%00 where: % = 2x(DDC channel #)+1 BIT 15 unused unused unused 0 0 0 BIT 7 unused 0 0 DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING tadj_offset_coarse_a(2:0) 0 tadj_offset_coarse_b(2:0) 0 Address: 0x06 0 unused BIT 8 unused 0 0 0 0 unused unused unused BIT 0 unused 0 0 0 0 tadj_offset_coarse_a(2:0) : This is the coarse time adjustment offset and acts as an offset from the write address in the delay ram. This value affects the A data in the path if CDMA mode is being used. Each LSB is one more offset between input to the course delay block and the output of the course block. dj_offset_coarse_b(2:0) : Effects the B channel in CDMA, just as the above effects the A channel. 4.4.5.8 TADJF Register Register name: TADJF Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x07 DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 tadj_offset_fine_a(2:0) BIT 8 tadj_interp(2:1) tadj_offset_fine_b(2:0) 0 0 0 0 0 0 0 0 BIT 7 tadj_interp(0) unused unused unused unused unused unused BIT 0 unused 0 0 0 0 0 0 0 0 tadj_offset_fine_a(2:0) : This is the fine adjust (zero stuff offset) value. It adjusts the time delay at the rxclk rate. This value affects the A channel data in the path if CDMA mode is being used. tadj_offset_fine_b(2:0) : Same as above except this value affects the B channel data in CDMA mode. tadj_interp(2:0) : This is the interpolation (zero stuff) value for the fine time adjust block. Interpolation can be from 1 to 8 (tadj_interp + 1). This value affects the A and B data in the path if CDMA mode is being used. 104 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.9 PHASEADD0A Register Register name: PHASEADD0A Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x08 DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 phase_add_a(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 phase_add_a(7:0) 0 0 0 0 0 0 0 0 phase_add_a(15:0) This 32 bit word is used to control the frequency of the NCO. This value is added to the frequency accumulator every clock cycle (UMTS mode and Main channel in CDMA mode). 4.4.5.10 PHASEADD1A Register Register name: PHASEADD1A Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x09 DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 phase_add_a(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 phase_add_a(23:16) 0 0 0 0 0 0 0 0 phase_add_a(31:16) : This 32 bit word is used to control the frequency of the NCO. This value is added to the frequency accumulator every clock cycle (UMTS mode and A channel in CDMA mode). 4.4.5.11 PHASEADD0B Register Register name: PHASEADD0 Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x0A DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 phase_add_b(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 phase_add_b(7:0) 0 0 0 0 0 0 0 0 phase_add_b(15:0) : This 32 bit word is used to control the frequency of the NCO. This value is added to the frequency accumulator every clock cycle (B channel in CDMA mode). GC5018 GENERAL CONTROL 105 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.12 PHASEADD1B Register Register name: PHASEADD1B Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x0B DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 phase_add_b(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 phase_add_b(23:16) 0 0 0 0 0 0 0 0 phase_add_b(31:16) : This 32 bit word is used to control the frequency of the NCO. This value is added to the frequency accumulator every clock cycle (B channel in CDMA mode). 4.4.5.13 PHASE_OFFSETA Register Register name: PHASE_OFFSETA Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x0C DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 phase_offset_a(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 phase_offset_a(7:0) 0 0 0 0 0 0 0 0 phase_offset_a(15:0) : This is the fixed phase offset added to the output of the frequency accumulator for sinusoid generation in the NCO. (UMTS mode and A channel in CDMA mode). 4.4.5.14 PHASE_OFFSETB Register Register name: PHASE_OFFSETB Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x0D DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 phase_offset_b(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 phase_offset_b(7:0) 0 0 0 0 0 0 0 0 phase_offset_b(15:0) : This is the fixed phase offset added to the output of the frequency accumulator for sinusoid generation in the NCO. (B channel in CDMA mode) 106 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.15 CONFIG1 Register Register name: CONFIG1 BIT 15 dither_ena Page: 0x0%00 where: % = 2x(DDC channel #)+1 dither_mask(1:0) Address: 0x0E pmeter_ sync_ disable ddc_ena muxed _data mixer_gain BIT 8 mpu_ram_read 0 0 0 0 0 0 0 0 BIT 7 unused unused unused unused unused zero_ qsample mux_pos BIT 0 mux_factor 0 0 0 0 0 0 0 0 dither_ena : This bit controls whether dither is turned on(1) or off(0). dither_mask(1) : This bit controls the MASKing of the dither word's MSB. (1= MASKed, 0=used in dither word) dither_mask(0) : This bit controls the MASKing of the dither word's MSB-1. (1= MASKed, 0=used in dither word) pmeter_sync_disable : Turns off the sync to the channel power meter. This can be used to individually turn off syncs to a channels power meter while still having syncs to other power meters available. ddc_ena : When set this turns on the DDC. When cleared, the clocks to this block are turned off. For the DDC blocks used as the second half in the long PFIR configuration, this bit should be cleared. muxed_data : When asserted the DDC mux block assumes that multiple channels are muxed together on one input data stream. For factory use only. For a 2X muxed stream it would look like: Sa0, Sb0, Sa1, Sb1, Sa2, Sb2 .... etc... mixer_gain : Adds a fixed 6 dB of gain to the mixer output(before round and limiting) when asserted. mpu_ram_read : (TESTING PURPOSES) Allows the coefficient RAMs in the PFIR/CFIR to be read out the mpu data bus. Unfortunately, this cannot be done during normal operation and must be done when the state of the output data is not important. THIS BIT MUST ONLY BE SET DURING THE MPU READ OPERATION AND MUST BE CLEARED FOR NORMAL DDC OPERATION. zero_qsample : When asserted, the Q sample into the mixer is held to zero. For UMTS mode at any input rate, and CDMA mode with input rates of rxclk/2 or lower, this bit must be set for real only input data mode (also for muxed input data stream modes). For real only inputs at the full rxclk rate in CDMA mode, the remix_only bit must be set in the DDCCONFIG1 register. mux_pos : These bits set the position for selection in the muxed data stream. This value must be less than or equal to the mux_factor bits. mux_factor : These two bits set the number of channels in the data stream. 0=1 stream, 1=2 streams. The ch_rate_sel bits for the DDC should be programmed to rxclk/2 for the 2 streams mode. GC5018 GENERAL CONTROL 107 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.16 CONFIG2 Register Register name: CONFIG2 Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x0F BIT 15 unused unused unused unused unused unused unused BIT 8 unused 0 0 0 0 0 0 0 0 BIT 7 unused unused 0 0 BIT 0 ddc_tst_sel(5:0) 0 0 0 0 0 0 ddc_tst_sel(5:0) : This is the selection of which signal comes out the test bus. When a constant `0' is selected this also reduces power by preventing the data at the input of the tst_blk from changing. It does not stop the clock however. The 36 bits for the testbus are routed to the rxin_c, rxin_d, dvga_c and dvga_d pins on the chip. 108 SYNC on dvga_c(0) AFLAG on dvga_d(5) ddc_tst_sel(5:0) N 000000 constant 0 Y 000001 pfir output - (35:18) I and (17:0) Q Y 000010 cfir output - (35:18) I and (17:0) Q N 000011 tadj A output - (35:18) I and (17:0) Q N 000100 tadj B output - (35:18) I and (17:0) Q N 000101 nco SINE output - (35:20) zeroed (19:0) SINE N 000110 nco COSINE output - (35:20) zeroed (19:0) COSINE N 000111 cic output - (35:18) I and (17:0) Q Y 001000 agc output - (35:11) I and (10:0) Q {full 25b I result and upper 11b Q result} N 001001 mix A output - (35:18) i*cos-q*sin and (17:0) i*sin+q*cos N 001010 mix B output - (35:18) i*cos-q*sin and (17:0) i*sin+q*cos N 001011 DDC MUX A output (35:18) I and (17:0) Q N 001100 DDC MUX B output (35:18) I and (17:0) Q GC5018 GENERAL CONTROL Data selected for output (36 bits total) rxin_d(15:0), dvga_c(3:2), rxin_c(15:0), dvga_c(5:4) GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.17 AGC_CONFIG1 Register Register name: AGC_CONFIG1 Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x10 BIT 15 BIT 8 agc_dblw(3:0) 0 0 agc_dabv(3:0) 0 0 0 0 0 BIT 7 BIT 0 agc_dzro(3:0) 0 0 0 agc_dsat(3:0) 0 0 0 0 0 0 agc_dblw(3:0) : The value to shift the gain that is then added to the accumulator when the value of the incoming data * current gain value is below the Threshold. agc_dabv(3:0) : The value to shift the gain that is then subtracted from the accumulator when the value of the incoming data * the current gain value is above the Threshold. agc_dzro(3:0) : The value to shift the gain that is then added to the accumulator when the value of the incoming data * current gain values consistently equal to zero. (Usually a smaller number than agc_dblw). agc_dsat(3:0) : The value to shift the gain that is then subtracted form the accumulator when the value of the incoming data * the current gain value is consistently equal to maximum (saturation). NOTE: The larger the number in the above words, the smaller the step size. The above values control the AGC gain shifting (range is from 3 to 18). 4.4.5.18 AGC_CONFIG2 Register Register name: AGC_CONFIG2 Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x11 BIT 15 BIT 8 zero_msk(3:0) 0 0 agc_rnd(3:0) 0 0 0 0 0 BIT 7 0 BIT 0 agc_thresh(7:0) 0 0 0 0 0 0 0 0 zero_msk(3:0) : Masks the lower 4 bits of the magnitude of the input signal so that they are counted as zeros. agc_rnd(3:0) : Determines where to round the output of the AGC; the number of bits output is (18 - agc_rnd). For example, 0000 is 18 bits. agc_thresh(7:0) : Threshold for (input * gain) comparison. This value is compared to the magnitude of the upper eight bits of the agc output. GC5018 GENERAL CONTROL 109 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.19 AGC_CONFIG3 Register Register name: AGC_CONFIG3 Page: 0x0%00 where: % = 2x(DDC channel #)+1 BIT 15 unused unused unused agc_ freeze 0 0 0 0 BIT 7 unused unused unused agc_ clear 0 0 0 0 Address: 0x12 BIT 8 agc_max_cnt(3:0) 0 0 0 0 BIT 0 agc_zero_cnt(3:0) 0 0 0 0 agc_freeze : Freezes the agc when set. This should be asserted when the AGC algorithm is bypassed or held constant. agc_max_cnt(3:0) : When the agc_output (input * gain) is at full scale for this number of samples, then the gain shift value is changed to agc_dsat. agc_clear : Clears the AGC accumulator when set. Assert this when the AGC is in bypass mode. agc_zero_cnt(3:0) : when the agc_output (input * gain) is zero value for this number of samples, then the gain shift value is changed to agc_dzro. 4.4.5.20 AGC_GAINMSB Register Register name: AGC_GAINMSB Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x13 DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 agc_gaina(23:16) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 agc_gainb(23:16) 0 0 0 0 agc_gaina(23:16) : MSBs of the agc_gaina word. agc_gainb(23:16) : MSBs of the agc_gainb word. 110 GC5018 GENERAL CONTROL 0 0 0 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.21 AGC_GAINA Register Register name: AGC_GAINA Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x14 DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 agc_gaina(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 agc_gaina(7:0) 0 0 0 0 0 0 0 0 agc_gaina(15:0) : This is the lower 16 bits of the total 24 bits of programmable gain. The gain value is always positive with the upper 12 bits being the integer value and the lower 12 bits being the fractional. This gain value is used for all UMTS operations and for A channel data when in CDMA mode. A 24-bit value of 00000000001.000000000000 is unity gain. 4.4.5.22 AGC_GAINB Register Register name: AGC_GAINB Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x15 DOUBLE BUFFERED, REQUIRES SYNC FOR LOADING BIT 15 BIT 8 agc_gainb(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 agc_gainb(7:0) 0 0 0 0 0 0 0 0 agc_gainb(15:0) : This is the lower 16 of the total of 24 bit of programmable gain. The gain value is always positive with the upper 12 bits being the integer value and the lower 12 bits being the fractional. This gain value is used for B channel data when in CDMA. A 24-bit value of 00000000001.000000000000 is unity gain. GC5018 GENERAL CONTROL 111 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.23 AGC_AMAX Register Register name: AGC_AMAX Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x16 BIT 15 BIT 8 agc_amax(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 agc_amax(7:0) 0 0 0 0 0 0 0 0 agc_amax(15:0) : This is the maximum gaina or gainb can be adjusted up. The value programmed is a positive value that is used to generate the most positive AGC gain adjust. For example, if 512 is programmed, the maximum gain will be the programmed gain (AGC_GAINA/B) + 512. 4.4.5.24 AGC_AMIN Register Register name: AGC_AMIN Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x17 BIT 15 BIT 8 agc_amin(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 agc_amin(7:0) 0 0 0 0 0 0 0 0 agc_amin(15:0) : This is the minimum gaina or gainb can be adjusted down. The value programmed is a positive value that is inverted internally to generate the most negative AGC gain adjust. For example, if 512 is programmed, the minimum gain will be the programmed gain (AGC_GAINA/B) - 512. 112 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.25 PSER_CONFIG1 Register Register name: PSER_CONFIG1 Page: 0x0%00 where: % = 2x(DDC channel #)+1 BIT 15 unused Address: 0x18 BIT 8 pser_recv_fsinvl(6:0) 0 0 0 BIT 7 unused unused unused 0 0 0 0 0 0 0 0 BIT 0 pser_recv_bits(4:0) 0 0 0 0 0 pser_recv_fsinvl(6:0) : Receive serial interface frame sync interval in bit clocks. pser_recv_bits(4:0) : Number of output bits per sample-1; for 18 bits, this is set to {10001}. 4.4.5.26 PSER_CONFIG2 Register Register name: PSER_CONFIG2 Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x19 BIT 15 pser_recv_clkdiv(3:0) unused unused unused BIT 8 unused 0 0 0 0 0 0 0 0 BIT 7 pser_recv_8pin pser_recv_alt unused unused unused unused 0 0 0 0 0 0 BIT 0 pser_recv_fsdel(1:0) 0 0 pser_recv_clkdiv(3:0) : Receive serial interface clock divider rate-1; 0 is full rate and 15 divides the clock by 16. For example, to run the receive serial interface at 1/4 the GC5018 clock, set pser_recv_clkdiv(3:0) = 0011. pser_recv_8pin : When set, 4 pins are used for I and 4 pins for Q in UMTS mode. When cleared, 2 pins are used for I and 2 pins for Q. This is used in combination with the pser_recv_alt bit. When this bit is set, it would be set in 2 adjacent DDC channels; one would also set the pser_recv_alt bit in the adjacent DDC. This will cause the I channel to be serialized on 4 pins and the Q channel to be serialized on the adjacent channels 4 pins. pser_recv_alt : When set, this channel's receive serial interface will output the Q data from the adjacent DDC channel. pser_recv_fsdel(1:0) : Delay between the receive frame sync output and the MSB of serial data {3,2,1,0}. This number is in serial output bit times, not rxclk periods. GC5018 GENERAL CONTROL 113 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.27 DDCCONFIG1 Register Register name: DDCCONFIG1 Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x1A BIT 15 ddcmux_sel_a (3:0) 0 0 0 agc_rnd_ disable gain_mon 0 0 remix_only cic_ bypass 0 0 0 BIT 8 ch_rate_sel(1:0) 0 BIT 7 ddcmux_sel_b(3:0) 0 0 0 0 0 BIT 0 double_tap(1:0) 0 0 ddcmux_sel_X(3:0) : Controls which samples go to the mixer for I/Q. Since in CDMA there are two streams, an A and B stream, two mux select values are used. Select Value I data from X input Q data from X input 0000 RXINA RXINA 0001 RXINB RXINB 0010 RXINC RXINC 0011 RXIND RXIND 0100 RXINA RXINB 0101 RXINA RXINC 0110 RXINA RXIND 0111 RXINB RXINA 1000 RXINB RXINC 1001 RXINB RXIND 1010 RXINC RXINA 1011 RXINC RXINB 1100 RXINC RXIND 1101 RXIND RXINA 1110 RXIND RXINB 1111 RXIND RXINC agc_rnd_disable : When set, the agc_rnd bits have no effect. The whole 29 bits are used in the rounding and the round bit is bit4. gain_mon : Combines the gain with the I/Q output signals when asserted. OUTPUT Bits(17:10) Bits(9:4) I Gained I value Q Gained Q value Bits(3:2) Gain(18:11) Gain(10:5) Bits(1:0) "00" Shift status(1:0) "00" ch_rate_sel(1:0) : Sets the DDC channel input data rate. The value set here should match the value in the Receive Input Interface rate select bits (rate_sel). ch_rate_sel 114 GC5018 GENERAL CONTROL Input data rate 00 rxclk 01 rxclk/2 10 rxclk/4 11 rxclk/8 www.ti.com GC5018 8-CHANNEL WIDEBAND RECEIVER SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 When muxed_data is set (Factory Use Only) rate_sel should be set to rxclk "00" and ch_rate_sel should be set to rxclk/2 "01". remix_only : Assert this when real only, full rxclk rate input data is used in CDMA mode. The signal on the Q bus selected by the ddcmux_sel_X(3:0) bits above is ignored (functions as if the Q data is 0). cic_bypass : Factory Use Only. If asserted then the data from the rxin_a(15:0) and rxin_b(15:0) are fed directly into the cfir input as I and Q respectively. rxin_a(0) also functions as the "sync_cfir" signal and should rise at the beginning of input data. ONLY DDC0, DDC2, DDC4 and DDC6 can be the UMTS double tap (64 to 128 tap) PFIR Mode. DDC1, DDC3, DDC5 and DDC7 PFIRs are used to lengthen the DDC0, DDC2, DDC4 and DDC6 PFIRs. double_tap(1) : When set, the DDC is in double length PFIR mode which sends the data out of the last PFIR sample ram in this DDC (DDC0, DDC2, DDC4, DDC6) to the adjacent secondary DDC (DDC1, DDC3, DDC5, DDC7) PFIR forming a 128-tap delay line. Output data received from the adjacent secondary DDC PFIR summer is added into the Main DDC's PFIR sum to form the final output. double_tap(0) : When set, the PFIR input comes from the adjacent(Main) PFIR. When cleared, PFIR input is from the CFIR connected directly to this PFIR. Only valid in DDC1, DDC3, DDC5 and DDC7. The ddc_ena bit in the CONFIG1 register should be cleared for the DDC1, DDC3, DDC5 and DDC7 when double_tap(0) is set. NOTE: to put 2 DDCs in to 128 tap mode: Program DDC0/DDC2/DDC4/DDC6 double_tap(1:0) to "10" and ddc_ena to "1". Program DDC1/DDC3/DDC5/DDC7 double_tap(1:0) to "01" and ddc_ena to "0". GC5018 GENERAL CONTROL 115 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.28 SYNC_0 Register Register name: SYNC_0 BIT 15 unused 0 Address: 0x1B BIT 8 ssel_cic(2:0) 0 BIT 7 unused 0 Page: 0x0%00 where: % = 2x(DDC channel #)+1 0 unused 0 0 ssel_pmeter(2:0) 0 0 0 BIT 0 ssel_agc_freeze(2:0) 1 1 unused 0 0 ssel_serial(2:0) 0 0 0 ssel_cic(2:0) : Selects the sync source for the DDC CIC filter, thus setting the decimation moment. ssel_pmeter(2:0) : Selects the sync source for the channel power meter. ssel_agc_freeze(2:0) : Selects the sync that is used to hold the AGC in freeze mode. With this functionality the user can program the AGC freeze control to look at the state of an input sync, or the one shots. It defaults to being off or not looking at any syncs and not driving the freeze control. This way, upon startup, the chip looks at the MPU register bit for AGC freezing and not the syncs. ssel_serial(2:0) : Selects the sync source for the DDC serial interface state machines. Sync sources are contained in this and many of the following registers. For all sync source selections: 116 ssel_XXXX(2:0) Selected sync source for DDC 000 rxsyncA 001 rxsyncB 010 rxsyncC GC5018 GENERAL CONTROL 011 rxsyncD 100 DDC sync counter 101 one shot (register write triggered) 110 always 0 111 always 1 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.29 SYNC_1 Register Register name: SYNC_1 BIT 15 unused Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x1C BIT 8 ssel_tadj_fine(2:0) 0 0 BIT 7 unused 0 unused 0 ssel_tadj_reg(2:0) 0 0 0 0 BIT 0 ssel_gain(2:0) 0 0 0 unused 0 ssel_ddc_agc(2:0) 0 0 0 0 ssel_tadj_fine(2:0) : Selects the sync source for the fine time adjust zero stuff moment. ssel_tadj_reg(2:0) : Selects the sync source for the fine and coarse time adjust register updates. ssel_gain(2:0) : Selects the sync source for the DDC AGC gain register. ssel_ddc_agc(2:0) : Selects the sync source to initialize the AGC, primarily for test purposes. 4.4.5.30 SYNC_2 Register Register name: SYNC_2 BIT 15 unused 0 Page: 0x0%00 where: % = 2x(DDC channel #)+1 Address: 0x1D BIT 8 ssel_nco(2:0) 0 BIT 7 unused 0 unused 0 0 ssel_dither(2:0) 0 0 0 BIT 0 ssel_freq(2:0) 0 unused 0 0 0 0 ssel_nco : Selects the sync source for the NCO accumulator reset. ssel_phase (2:0) 0 0 0 ssel_dither : Selects the sync source for the NCO phase dither generator reset. ssel_freq : Selects the sync source for the NCO frequency register. ssel_phase : Selects the sync source for the NCO phase offset register. GC5018 GENERAL CONTROL 117 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.31 DDC_CHK_SUM Register Register name: DDC_CHK_SUM Page: 0x0%20 where: % = 2x(DDC channel #)+1 Address: 0x20 READ ONLY BIT 15 BIT 8 ddc_chk_sum(15:0) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 ddc_chk_sum(7:0) 0 0 0 0 0 0 0 0 ddc_chk_sum : The DDC self test checksum value 4.4.5.32 PMETER_RESULT_A_LSB Register Register name: PMETER_RESULT_A_LSB Page: 0x0%20 where: % = 2x(DDC channel #)+1 Address: 0x21 READ ONLY BIT 15 BIT 8 pmeter_result_a(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_result_a(7:0) 0 0 0 0 0 0 0 0 pmeter_result_a(15:0) : Lower 16 bits of the UMTS mode or CDMA mode A channel power measurement. 118 GC5018 GENERAL CONTROL GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.33 PMETER_RESULT_A_MID Register Register name: PMETER_RESULT_A_MID Page: 0x0%20 where: % = 2x(DDC channel #)+1 Address: 0x22 READ ONLY BIT 15 BIT 8 pmeter_result_a(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_result_a(23:16) 0 0 0 0 0 0 0 0 pmeter_result_a(31:16) : Mid 16 bits of the UMTS mode or CDMA mode A channel power measurement. 4.4.5.34 PMETER_RESULT_A_MSB Register Register name: PMETER_RESULT_A_MSB Page: 0x0%20 where: % = 2x(DDC channel #)+1 Address: 0x23 READ ONLY BIT 15 BIT 8 pmeter_result_a(47:40) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_result_a(39:32) 0 0 0 0 0 0 0 0 pmeter_result_a(47:32) : Upper mid 16 bits of the UMTS mode or CDMA mode A channel power measurement. 4.4.5.35 PMETER_RESULT_B_LSB Register Register name: PMETER_RESULT_B_LSB Page: 0x0%20 where: % = 2x(DDC channel #)+1 Address: 0x24 READ ONLY BIT 15 BIT 8 pmeter_result_b(15:8) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_result_b(7:0) 0 0 0 0 0 0 0 0 pmeter_result_b(15:0) : Lower 16 bits of the CDMA mode B channel power measurement GC5018 GENERAL CONTROL 119 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.36 PMETER_RESULT_B_MID Register Register name: PMETER_RESULT_B_MID Page: 0x0%20 where: % = 2x(DDC channel #)+1 Address: 0x25 READ ONLY BIT 15 BIT 8 pmeter_result_b(31:24) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_result_b(23:16) 0 0 0 0 0 0 0 0 pmeter_result_b(31:16) : Mid 16 bits of the CDMA mode B channel power measurement. 4.4.5.37 PMETER_RESULT_B_MSB Register Register name: PMETER_RESULT_B_MSB Page: 0x0%20 where: % = 2x(DDC channel #)+1 Address: 0x26 READ ONLY BIT 15 BIT 8 pmeter_result_b(47:40) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_result_b(39:32) 0 0 0 0 0 0 0 pmeter_result_b(47:32) : Upper mid 16 bits of the CDMA mode B channel power measurement. 120 GC5018 GENERAL CONTROL 0 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 4.4.5.38 PMETER_RESULT_AB_UMSB Register Register name: PMETER_RESULT_AB_UMSB Page: 0x0%20 where: % = 2x(DDC channel #)+1 Address: 0x27 READ ONLY BIT 15 BIT 8 pmeter_result_a(54:48) 0 0 0 0 0 0 0 BIT 7 0 BIT 0 pmeter_result_b(54:48) 0 0 0 0 0 0 0 0 pmeter_result_a(54:48) : Most Significant 7 bits of the 55-bit UMTS or CDMA mode A channel power measurement. pmeter_result_b(54:48) : Most Significant 7 bits of the 55-bit CDMA mode B channel power measurement. GC5018 GENERAL CONTROL 121 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 5 GC5018 PINS 5.1 Digital Receive Section Signals Ball Type rxclk Signal Name U8 input receive digital section clock input adcclk_a B10 input rxin_a_x input clock adcclk_b A10 input rxin_b_x input clock adcclk_c F2 input rxin_c_x input clock adcclk_d E4 input rxin_d_x input clock rxin_a_ovr B16 input adc overflow/overrange bit for rxin_a rxin_b_ovr C9 input adc overflow/overrange bit for rxin_b rxin_c_ovr A3 input adc overflow/overrange bit for rxin_c rxin_d_ovr E1 input adc overflow/overrange bit for rxin_d dvga_a_5 A17 output Digital VGA control output for ADC0 MSB dvga_a_4 B17 output Digital VGA control output for ADC0 dvga_a_3 C16 output Digital VGA control output for ADC0 dvga_a_2 C17 output Digital VGA control output for ADC0 dvga_a_1 D16 output Digital VGA control output for ADC0 dvga_a_0 D17 output Digital VGA control output for ADC0 LSB dvga_b_5 B18 output Digital VGA control output for ADC1 MSB dvga_b_4 E16 output Digital VGA control output for ADC1 dvga_b_3 E17 output Digital VGA control output for ADC1 dvga_b_2 C18 output Digital VGA control output for ADC1 dvga_b_1 D15 output Digital VGA control output for ADC1 dvga_b_0 D18 output Digital VGA control output for ADC1 LSB dvga_c_5 M1 output Digital VGA control output for rxin_c MSB, test bus bit 1 dvga_c_4 L2 output Digital VGA control output for rxin_c, test bus bit 0 dvga_c_3 L3 output Digital VGA control output for rxin_c, test bus bit 19 dvga_c_2 M4 output Digital VGA control output for rxin_c, test bus bit 18 dvga_c_1 N4 output Digital VGA control output for rxin_c, test bus CLK dvga_c_0 M2 output Digital VGA control output for rxin_c LSB, test bus SYNC dvga_d_5 M3 output Digital VGA control output for rxin_d MSB, test bus AFLAG dvga_d_4 P1 output Digital VGA control output for rxin_d dvga_d_3 P4 output Digital VGA control output for rxin_d dvga_d_2 N2 output Digital VGA control output for rxin_d dvga_d_1 R1 output Digital VGA control output for rxin_d dvga_d_0 N3 output Digital VGA control output for rxin_d LSB rxin_a_15 C15 input receive input data bus a bit 15 (MSB) rxin_a_14 B15 input receive input data bus a rxin_a_13 C14 input receive input data bus a 122 GC5018 PINS Description GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Ball Type rxin_a_12 Signal Name B14 input receive input data bus a Description rxin_a_11 A16 input receive input data bus a rxin_a_10 A15 input receive input data bus a rxin_a_9 C13 input receive input data bus a rxin_a_8 B13 input receive input data bus a rxin_a_7 A14 input receive input data bus a rxin_a_6 C12 input receive input data bus a rxin_a_5 B12 input receive input data bus a rxin_a_4 A12 input receive input data bus a rxin_a_3 C11 input receive input data bus a rxin_a_2 B11 input receive input data bus a rxin_a_1 D11 input receive input data bus a rxin_a_0 C10 input receive input data bus a bit 0 (LSB) rxin_b_15 B9 input receive input data bus b bit 15 (MSB) rxin_b_14 D9 input receive input data bus b rxin_b_13 A9 input receive input data bus b rxin_b_12 C8 input receive input data bus b rxin_b_11 B8 input receive input data bus b rxin_b_10 D8 input receive input data bus b rxin_b_9 C7 input receive input data bus b rxin_b_8 B7 input receive input data bus b rxin_b_7 A7 input receive input data bus b rxin_b_6 B6 input receive input data bus b rxin_b_5 C6 input receive input data bus b rxin_b_4 A5 input receive input data bus b rxin_b_3 B5 input receive input data bus b rxin_b_2 C5 input receive input data bus b rxin_b_1 A4 input receive input data bus b rxin_b_0 B4 input receive input data bus b bit 0 (LSB) rxin_c_15 A2 input/output receive input data bus c bit 15 (MSB), test bus bit 17 rxin_c_14 B3 input/output receive input data bus c bit 14, test bus bit 16 rxin_c_13 B2 input/output receive input data bus c bit 13, test bus bit 15 rxin_c_12 C3 input/output receive input data bus c bit 12, test bus bit 14 rxin_c_11 C2 input/output receive input data bus c bit 11, test bus bit 13 rxin_c_10 A1 input/output receive input data bus c bit 10, test bus bit 12 rxin_c_9 D3 input/output receive input data bus c bit 9, test bus bit 11 rxin_c_8 D2 input/output receive input data bus c bit 8, test bus bit 10 rxin_c_7 B1 input/output receive input data bus c bit 7, test bus bit 9 rxin_c_6 C4 input/output receive input data bus c bit 6, test bus bit 8 rxin_c_5 E3 input/output receive input data bus c bit 5, test bus bit 7 rxin_c_4 C1 input/output receive input data bus c bit 4, test bus bit 6 rxin_c_3 E2 input/output receive input data bus c bit 3, test bus bit 5 rxin_c_2 D4 input/output receive input data bus c bit 2, test bus bit 4 rxin_c_1 D1 input/output receive input data bus c bit 1, test bus bit 3 rxin_c_0 F3 input/output receive input data bus c bit 0 (LSB), test bus bit 2 rxin_d_15 G3 input/output receive input data bus d bit 15 (MSB), test bus bit 35 GC5018 PINS 123 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Signal Name Ball Type G2 input/output receive input data bus d bit 14, test bus bit 34 rxin_d_13 F4 input/output receive input data bus d bit 13, test bus bit 33 rxin_d_12 G4 input/output receive input data bus d bit 12, test bus bit 32 rxin_d_11 G1 input/output receive input data bus d bit 11, test bus bit 31 rxin_d_10 H3 input/output receive input data bus d bit 10, test bus bit 30 rxin_d_9 H2 input/output receive input data bus d bit 9, test bus bit 29 rxin_d_8 H4 input/output receive input data bus d bit 8, test bus bit 28 rxin_d_7 J3 input/output receive input data bus d bit 7, test bus bit 27 rxin_d_6 J2 input/output receive input data bus d bit 6, test bus bit 26 rxin_d_5 J1 input/output receive input data bus d bit 5, test bus bit 25 rxin_d_4 K1 input/output receive input data bus d bit 4, test bus bit 24 rxin_d_3 K2 input/output receive input data bus d bit 3, test bus bit 23 rxin_d_2 K3 input/output receive input data bus d bit 2, test bus bit 22 rxin_d_1 K4 input/output receive input data bus d bit 1, test bus bit 21 rxin_d_0 L4 input/output receive input data bus d bit 0 (LSB), test bus bit 20 rx_synca T8 input receive sync input rx_syncb V10 input receive sync input rx_syncc R10 input receive sync input rx_syncd U9 input receive sync input rx_sync_out U16 output receive general purpose output sync rxclk_out U17 output receive clock output rx_sync_out_7 U15 output receive serial interface frame strobe for rxout_7_x rx_sync_out_6 T18 output receive serial interface frame strobe for rxout_6_x, frame strobe (rx_sync_out signal) for parallel interface. rx_sync_out_5 P15 output receive serial interface frame strobe for rxout_5_x rx_sync_out_4 M15 output receive serial interface frame strobe for rxout_4_x rx_sync_out_3 K16 output receive serial interface frame strobe for rxout_3_x rx_sync_out_2 J16 output receive serial interface frame strobe for rxout_2_x rx_sync_out_1 G15 output receive serial interface frame strobe for rxout_1_x rx_sync_out_0 E15 output receive serial interface frame strobe for rxout_0_x rxout_7_a R16 output DDC 7 serial out data. CDMA A: I data UMTS: Imsb DDC Parallel Interface I(12) rxout_7_b R17 output DDC 7 serial out data. CDMA B: I data UMTS: Imsb - 1 DDC Parallel Interface I(13) rxout_7_c U18 output DDC 7 serial out data. CDMA A: Q data UMTS: Qmsb DDC Parallel Interface I(14) rxout_7_d P16 output DDC 7 serial out data. CDMA B: Q data UMTS: Qmsb -1 DDC Parallel Interface I(15) rxout_6_a P17 output DDC 6 serial out data. CDMA A: I data UMTS: Imsb DDC Parallel Interface I(8) rxout_6_b T15 output DDC 6 serial out data. CDMA B: I data UMTS: Imsb - 1 DDC Parallel Interface I(9) rxout_6_c R15 output DDC 6 serial out data. CDMA A: Q data UMTS: Qmsb DDC Parallel Interface I(10) rxout_6_d N16 output DDC 6 serial out data. CDMA B: Q data UMTS: Qmsb -1 DDC Parallel Interface I(11) rxout_5_a N17 output DDC 5 serial out data. CDMA A: I data UMTS: Imsb Parallel Interface I(4) rxout_5_b R18 output DDC 5 serial out data. CDMA B: I data UMTS: Imsb - 1 Parallel Interface I(5) rxout_5_c P18 output DDC 5 serial out data. CDMA A: Q data UMTS: Qmsb Parallel Interface I(6) rxout_5_d M16 output DDC 5 serial out data. CDMA B: Q data UMTS: Qmsb -1 Parallel Interface I(7) rxin_d_14 124 GC5018 PINS Description GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Signal Name Ball Type Description rxout_4_a M17 output DDC 4 serial out data. CDMA A: I data UMTS: Imsb Parallel Interface I(0) rxout_4_b N15 output DDC 4 serial out data. CDMA B: I data UMTS: Imsb - 1 Parallel Interface I(1) rxout_4_c L16 output DDC 4 serial out data. CDMA A: Q data UMTS: Qmsb Parallel Interface I(2) rxout_4_d L17 output DDC 4 serial out data. CDMA B: Q data UMTS: Qmsb -1 Parallel Interface I(3) rxout_3_a M18 output DDC 3 serial out data. CDMA A: I data UMTS: Imsb Parallel Interface Q(12) rxout_3_b L15 output DDC 3 serial out data. CDMA B: I data UMTS: Imsb - 1 Parallel Interface Q(13) rxout_3_c K17 output DDC 3 serial out data. CDMA A: Q data UMTS: Qmsb Parallel Interface Q(14) rxout_3_d K18 output DDC 3 serial out data. CDMA B: Q data UMTS: Qmsb -1 Parallel Interface Q(15) rxout_2_a J18 output DDC 2 serial out data. CDMA A: I data UMTS: Imsb Parallel Interface Q(8) rxout_2_b J17 output DDC 2 serial out data. CDMA B: I data UMTS: Imsb - 1 Parallel Interface Q(9) rxout_2_c H15 output DDC 2 serial out data. CDMA A: Q data UMTS: Qmsb Parallel Interface Q(10) rxout_2_d G18 output DDC 2 serial out data. CDMA B: Q data UMTS: Qmsb -1 Parallel Interface Q(11) rxout_1_a H17 output DDC 1 serial out data. CDMA A: I data UMTS: Imsb Parallel Interface Q(4) rxout_1_b H16 output DDC 1 serial out data. CDMA B: I data UMTS: Imsb - 1 Parallel Interface Q(5) rxout_1_c F15 output DDC 1 serial out data. CDMA A: Q data UMTS: Qmsb Parallel Interface Q(6) rxout_1_d G17 output DDC 1 serial out data. CDMA B: Q data UMTS: Qmsb -1 Parallel Interface Q(7) rxout_0_a G16 output DDC 0 serial out data. CDMA A: I data UMTS: Imsb Parallel Interface Q(0) rxout_0_b E18 output DDC 0 serial out data. CDMA B: I data UMTS: Imsb - 1 Parallel Interface Q(1) rxout_0_c F17 output DDC 0 serial out data. CDMA A: Q data UMTS: Qmsb Parallel Interface Q(2) rxout_0_d F16 output DDC 0 serial out data. CDMA B: Q data UMTS: Qmsb -1 Parallel Interface Q(3) GC5018 PINS 125 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 5.2 Microprocessor Signals Ball Type d0 Signal Name V3 input/output MPU register interface data bus bit 0 (LSB) d1 U3 input/output MPU register interface data bus d2 V2 input/output MPU register interface data bus d3 U2 input/output MPU register interface data bus d4 T3 input/output MPU register interface data bus d5 T2 input/output MPU register interface data bus d6 V4 input/output MPU register interface data bus d7 R3 input/output MPU register interface data bus d8 U4 input/output MPU register interface data bus d9 R2 input/output MPU register interface data bus d10 U1 input/output MPU register interface data bus d11 P3 input/output MPU register interface data bus d12 T4 input/output MPU register interface data bus d13 T1 input/output MPU register interface data bus d14 P2 input/output MPU register interface data bus d15 R4 input/output MPU register interface data bus bit 15 (MSB) a0 V7 input MPU register interface address bus bit 0 (LSB) a1 T6 input MPU register interface address bus a2 U6 input MPU register interface address bus a3 V5 input MPU register interface address bus a4 T5 input MPU register interface address bus a5 U5 input MPU register interface address bus bit 5 (MSB) rd_n T7 input MPU register interface read - active low wr_n V9 input MPU register interface write - active low ce_n U7 input MPU register interface chip enable - active low reset_n R9 input chip reset - active low interrupt T9 output 126 GC5018 PINS Description chip interrupt GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 5.3 JTAG Signals Signal Name Ball Type U11 input JTAG test data in tms T11 input JTAG test mode select trst_n U12 input JTAG test reset (same as trst; the "_n" is for consistency - being active low) tdi Description Note: the trst_n pin should be asserted low after power up to insure the JTAG logic is properly initialized. tck V12 input tdo U10 output 5.4 JTAG test clock JTAG test data out Factory Test and No Connect Signals Signal Name Ball Type U14 input Do not connect; internal pull down testmode1 T13 input Do not connect; internal pull down scanen U13 input Do not connect; internal pull down fa002_scan V14 input Do not connect; internal pull down fa002_clk V15 input Do not connect; internal pull down fa002_out T12 output zero T14 input fuse_out V16 output testmode0 5.5 Description Do not connect Do not connect; internal pull down Do not connect Power and Ground Signals Signal Name Ball Description VDDS A6, D5, D6, D10, D13, D14, E5, E13, E14, F1, F5, F14, F18, J4, J15, K15, L5, M5, N5, N14, P5, P13, P14, R5, R6, R7, R8, R12, R13, R14, T16, V6, V17 Digital I/O Power (3.3 V), also called Vpad DVDD A13, D7, D12, E6, E7, E8, E10, E11, E12, G5, G14, H5, H14, J5, J14, L14, M14, N1, N18, P6, P7, P8, P10, P11, P12, V13 Digital Core Power (1.5 V), also called Vcore GC5018 PINS 127 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 Signal Name Ball Description DVSS A8, A11, E9, F6, F7, F8, F9, F10, F11, F12, F13, G6, G7, G8, G9, G10, G11, G12, G13, H1, H6, H7, H12, H13, H18 J6, J7, J12, J13, K5, K6, K7, K12, K13, K14, L1, L6, L7, L12, L13, L18, M6, M7, M8, M9, M10, M11, M12, M13, N6, N7, N8, N9, N10, N11, N12, N13, T17, P9, V8, V11 Digital Ground 5.6 Digital Supply Monitoring Signal Name Ball Description dvddmon T10 It is recommended that this pin be brought to a probe point for monitoring and debugging purposes. dvssmon R11 It is recommended that this pin be brought to a probe point for monitoring and debugging purposes. 5.7 JTAG The JTAG standard for boundary scan testing will be implemented for board testing purposes. Internal scan test will not be supported. Five device pins are dedicated for JTAG support: tdi, tdo, tms, tck, and trst_n. The JTAG bsdl configuration file is available at www.ti.com. NOTE The trst_n pin should be asserted after power up to insure the JTAG logic is properly initialized. 6 SPECIFICATIONS NOTE: These numbers are engineering estimates prior to first silicon. They will change after we have characterized the parts. ABSOLUTE MAXIMUM RATINGS (1) 6.1 UNIT VDDS Pad ring supply voltage DVDD Core supply voltage - 0.3 V to 1.8 V Digital input voltage - 0.3 V to VDDS+0.3 V Clamp current for an input or output -0.3 V to 3.7 V - 20 mA to +20 mA TSTG Storage temperature - 65C to 140C TJ Junction temperature 105C Lead soldering temperature (10 seconds) 300C Class 2 ESD classification (tested to EIA/JESD22-A114-B) Class 2 Moisture sensitivity (1) 128 Stresses beyond those listed under absolute maximum ratings may cause permanent damage to the device. These are stress ratings only, and functional operation of the device at these or any other conditions beyond those indicated under recommended operating conditions is not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability. SPECIFICATIONS GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 6.2 RECOMMENDED OPERATING CONDITIONS MIN VDDS Digital chip, I/O ring supply voltage DVDD Digital chip, core supply voltage NOM Temperature ambient, no air flow TJ (2) Junction temperature (1) (2) UNIT 3.6 V 1.425 1.575 V Digital chip, supply voltage difference, VDDS - DVDD TA (1) MAX 3 -40 2.0 V 85 C 105 C Chips specifications in Tables 6.4 and 6.5 are production tested to 90C case temperature. QA tests are performed at 85C. Thermal management will be required for full rate operation, See table below and Section 8.4. The circuit is designed for junction temperatures up to 125C. Sustained operation at elevated temperatures will reduce long-term reliability. Lifetime calculations based on maximum junction temperature of 105C. 6.3 THERMAL CHARACTERISTICS (1) JA Theta Junction to Ambient (still air) 15.3 C/W JA2m Theta Junction to Ambient (2m/s estimated) 12.4 C/W JC Theta Junction to Case 4.5 C/W THERMAL CONDUCTIVITY (1) MIN TYP MAX UNIT Air flow will reduce JA and is highly recommended. 6.3.1 POWER CONSUMPTION The maximum power consumption is largely a function of the operating mode of the chip. IDVDD = proportional to filter lengths, supply, frequency, and number of channels active. Current consumption on the pad supply is primarily due to the external loads and follows CxVxF. Internal loads are estimated at 2 pF per pin. Data outputs have a transition density of going from a zero to a one once per four clocks, while clock outputs transition every cycle. The rx_sync_out_X frame strobes consume negligible power due to the low transition frequency. In general, IVDDS = DataPad/4xCxFxV + ClockPadxCxFxV DC CHARACTERISTICS (1) (2) (3) 6.4 -40C to 85C case (unless otherwise noted) VDDS=3 V to 3.6 V PARAMETER VIL Voltage input low VIH Voltage input high VOL Voltage output low (IOL = 2 mA) (4) VOH Voltage output high (IOH = -2 mA) (4) MIN MAX 0.8 2.0 A) (4) UNIT V V 0.5 V 2.4 VDDS V 5 35 A 5 35 A 20 A 8 mA |IPU| Pullup current (VIN = 0 V) (tdi, tms, trst_n, ce_n, wr_n, rd_n, reset_n ) (nominal 20 |IPD| Pulldown current (VIN = VDDS) (all other inputs and bidirectionals) (nominal 20 A) (4) |IIN| Leakage current (VIN = 0V or VDDS), Outputs in 3-state condition ICCQ Quiescent supply current, IDVDD or IVDDS (VIN = 0 for pads with pulldowns, VIN = VDDS for inputs with pullups) (4) CIN Capacitance for inputs (5) Typical 5 Typical 5 pF CBI Capacitance for bidirectionals (5) Typical 5 Typical 5 pF (1) (2) (3) (4) (5) (4) Voltages are measured at low speed. Output voltages are measured with the indicated current load. Currents are measured at nominal voltages, high temperature (90C for production test, 85C for QA). reset_n and interrupt have no timing specifications since they are asynchronous signals. die_id pins fa002_out, fa002_clk and fa002_scan will not be specified and are for factory use only. fuse pin fuse_out will not be specified and is for factory use only. test pins zero, scanen, testmode0 and testmode1 will not be specified and are for factory use only. Each part is tested at 90C case temperature for the given specification. Lots are sample tested at -40C. Controlled by design and process and not directly tested. SPECIFICATIONS 129 GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 AC TIMING CHARACATERISTICS (1) 6.5 -40C to 85C case supplies across recommended range (unless otherwise noted) PARAMETER MIN TYP MAX UNIT 160 MHz FCK Clock frequency (adcclk_a/b/c/d, rxclk) (2) tCKL Clock low period (below VIL) (adcclk_a/b/c/d, rxclk) (2) 2 ns tCKH Clock high period (above VIH) (adcclk_a/b/c/d, rxclk) (2) 2 ns tRF Clock rise and fall times (VIL to VIH) (adcclk_a/b/c/d, rxclk) (3) 2 Input setup (rxsync_a/b/c/d) before rxclk rises (2) Input setup (rxin_a/b/c/d_[0-15] ) before rxclk rises (adc fifo blocks bypassed) (2) tSU Input setup (rxin_a/b/c/d_[0-15] ) before adcclk_a/b/c/d rises (adc fifo blocks 2 enabled) (2) Input hold (rxsync_a/b/c/d) after rxclk rises (2) ns 2 1 Input hold (rxin_a/b/c/d_[0-15] ) after rxclk rises (adc fifo blocks bypassed) (2) tHD ns 2 Input hold (rxin_ a/b/c/d_[0-15] ) after adcclk_a/b/c/d rises (adc fifo blocks 2.5 enabled) (2) ns 1 tDLY Data output delay (rx_sync_out_[0-7], rxout_[0-7]_a/b/c/d, rxclk_out, rx_sync_out, dvga_[a-d]_[5-0]) after rxclk rises. (2) tOHD Data output hold (rx_sync_out_[0-7], rxout_[0-7]_a/b/c/d, rxclk_out, rx_sync_out, dvga_[a-d]_[5-0]) after rxclk rises. (2) FJCK JTAG Clock frequency (tck) (2) tJCKL JTAG Clock low period (below VIL) (tck) (2) 10 ns tJCKH JTAG Clock high period (above VIH) (tck ) (2) 10 ns tJSU JTAG Input (tdi or tms) setup before tck goes high (2) 2 ns tJHD JTAG Input (tdi or tms) hold time after tck goes high (2) tJDLY JTAG output (tdo) delay from falling edge of tck. (2) tCSU Control setup during reads or writes 3 pin mode: a[5:0] valid before rd_n, wr_n or ce_n falling edge 2 pin mode: a[5:0] and wr_n valid before ce_n falling edge (2) tEWCSU Control setup during writes 3 pin mode: d[15:0] valid before wr_n and ce_n rising edge 2 pin mode: d[15:0] valid before ce_n rising edge (2) tCHD Control hold during writes. 3 pin mode: a[5:0] and d[15:0] valid after wr_n and ce_n rise 2 pin mode: a[5:0], d[15:0] and wr_n valid after ce_n rise (2) tCSPW Control strobe (ce_n and wr_n low) pulse width during write. tCDLY Control output delay ce_n and rd_n low and a[5:0] stable to d[15:0] during read. MHz ns 10 (2) 6 ns 10 ns 6 ns 25 ns (2) (2) Control recovery time between reads or writes. Control end of read to Hi-Z. rd_n and ce_n rise to d[15:0] 3-state tCOH Control read d[15:0] output hold time (4) ICDYN Core dynamic supply current ,nominal voltages, 160 MHz, (specific conditions, typical app with chip busy within capability of the tester, high temperature.) (4) 130 ns ns 10 tHIZ (2) (3) (4) 0.5 40 tREC (1) 7 (4) 1 25 ns 6 ns 10 ns ns 1700 mA Timing is measured from the respective clock at VDDS/2 to input or output at VDDS/2. Output loading is a 50 transmission line whose delay is calibrated out. Each part is tested at 90C case temperature for the given specification. Lots are sample tested at -40C. Recommended practice. Controlled by design and process and not directly tested. Verified on initial part evaluation. SPECIFICATIONS GC5018 8-CHANNEL WIDEBAND RECEIVER www.ti.com SLWS169A - MAY 2005 - REVISED NOVEMBER 2005 7 Revision History DATE REV ?? OCT 05 A PAGE SECTION DESCRIPTION DC Characterisitcs Changed Note 1 AC Timing Characterisitcs Changed note references at section header tHD Input hold (rxin_a/b/c/d_[0-15] ) after rxclk rises (adc fifo blocks bypassed) changed from 2.0 to 2.5 ns min tDLY changed from 6.5 ns to 7.0 ns tJSU JTAG input (tdi or tms) setup from 1.0 ns to 2.0 ns min tCSPW control strobe pulse width during write from 15 ns to 25 ns min tHIZ control end of read Hi-Z needs note (4) instead of note (1) tCOH control read d(15:0) hold time from 3 ns to 1 ns, and note changed from (1) to (4) ICDYN needs to be 1700 mA typical (not a maximum, but a typical value) and note changed from (1) to (4) 25 MAR 05 * - - Original version Revision History 131 PACKAGE OPTION ADDENDUM www.ti.com 17-Mar-2009 PACKAGING INFORMATION Orderable Device Status (1) Package Type Package Drawing GC5018IZDL ACTIVE BGA ZDL Pins Package Eco Plan (2) Qty 305 84 Green (RoHS & no Sb/Br) Lead/Ball Finish SNAGCU MSL Peak Temp (3) Level-3-260C-168 HR (1) The marketing status values are defined as follows: ACTIVE: Product device recommended for new designs. LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect. NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in a new design. PREVIEW: Device has been announced but is not in production. Samples may or may not be available. OBSOLETE: TI has discontinued the production of the device. (2) Eco Plan - The planned eco-friendly classification: Pb-Free (RoHS), Pb-Free (RoHS Exempt), or Green (RoHS & no Sb/Br) - please check http://www.ti.com/productcontent for the latest availability information and additional product content details. TBD: The Pb-Free/Green conversion plan has not been defined. Pb-Free (RoHS): TI's terms "Lead-Free" or "Pb-Free" mean semiconductor products that are compatible with the current RoHS requirements for all 6 substances, including the requirement that lead not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered at high temperatures, TI Pb-Free products are suitable for use in specified lead-free processes. Pb-Free (RoHS Exempt): This component has a RoHS exemption for either 1) lead-based flip-chip solder bumps used between the die and package, or 2) lead-based die adhesive used between the die and leadframe. The component is otherwise considered Pb-Free (RoHS compatible) as defined above. Green (RoHS & no Sb/Br): TI defines "Green" to mean Pb-Free (RoHS compatible), and free of Bromine (Br) and Antimony (Sb) based flame retardants (Br or Sb do not exceed 0.1% by weight in homogeneous material) (3) MSL, Peak Temp. -- The Moisture Sensitivity Level rating according to the JEDEC industry standard classifications, and peak solder temperature. Important Information and Disclaimer:The information provided on this page represents TI's knowledge and belief as of the date that it is provided. TI bases its knowledge and belief on information provided by third parties, and makes no representation or warranty as to the accuracy of such information. Efforts are underway to better integrate information from third parties. TI has taken and continues to take reasonable steps to provide representative and accurate information but may not have conducted destructive testing or chemical analysis on incoming materials and chemicals. TI and TI suppliers consider certain information to be proprietary, and thus CAS numbers and other limited information may not be available for release. In no event shall TI's liability arising out of such information exceed the total purchase price of the TI part(s) at issue in this document sold by TI to Customer on an annual basis. Addendum-Page 1 IMPORTANT NOTICE Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications, enhancements, improvements, and other changes to its products and services at any time and to discontinue any product or service without notice. Customers should obtain the latest relevant information before placing orders and should verify that such information is current and complete. All products are sold subject to TI's terms and conditions of sale supplied at the time of order acknowledgment. TI warrants performance of its hardware products to the specifications applicable at the time of sale in accordance with TI's standard warranty. Testing and other quality control techniques are used to the extent TI deems necessary to support this warranty. Except where mandated by government requirements, testing of all parameters of each product is not necessarily performed. TI assumes no liability for applications assistance or customer product design. Customers are responsible for their products and applications using TI components. To minimize the risks associated with customer products and applications, customers should provide adequate design and operating safeguards. TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right, copyright, mask work right, or other TI intellectual property right relating to any combination, machine, or process in which TI products or services are used. Information published by TI regarding third-party products or services does not constitute a license from TI to use such products or services or a warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of the third party, or a license from TI under the patents or other intellectual property of TI. Reproduction of TI information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied by all associated warranties, conditions, limitations, and notices. Reproduction of this information with alteration is an unfair and deceptive business practice. TI is not responsible or liable for such altered documentation. Information of third parties may be subject to additional restrictions. Resale of TI products or services with statements different from or beyond the parameters stated by TI for that product or service voids all express and any implied warranties for the associated TI product or service and is an unfair and deceptive business practice. TI is not responsible or liable for any such statements. TI products are not authorized for use in safety-critical applications (such as life support) where a failure of the TI product would reasonably be expected to cause severe personal injury or death, unless officers of the parties have executed an agreement specifically governing such use. Buyers represent that they have all necessary expertise in the safety and regulatory ramifications of their applications, and acknowledge and agree that they are solely responsible for all legal, regulatory and safety-related requirements concerning their products and any use of TI products in such safety-critical applications, notwithstanding any applications-related information or support that may be provided by TI. Further, Buyers must fully indemnify TI and its representatives against any damages arising out of the use of TI products in such safety-critical applications. TI products are neither designed nor intended for use in military/aerospace applications or environments unless the TI products are specifically designated by TI as military-grade or "enhanced plastic." Only products designated by TI as military-grade meet military specifications. Buyers acknowledge and agree that any such use of TI products which TI has not designated as military-grade is solely at the Buyer's risk, and that they are solely responsible for compliance with all legal and regulatory requirements in connection with such use. TI products are neither designed nor intended for use in automotive applications or environments unless the specific TI products are designated by TI as compliant with ISO/TS 16949 requirements. Buyers acknowledge and agree that, if they use any non-designated products in automotive applications, TI will not be responsible for any failure to meet such requirements. Following are URLs where you can obtain information on other Texas Instruments products and application solutions: Products Amplifiers Data Converters DLP(R) Products DSP Clocks and Timers Interface Logic Power Mgmt Microcontrollers RFID RF/IF and ZigBee(R) Solutions amplifier.ti.com dataconverter.ti.com www.dlp.com dsp.ti.com www.ti.com/clocks interface.ti.com logic.ti.com power.ti.com microcontroller.ti.com www.ti-rfid.com www.ti.com/lprf Applications Audio Automotive Broadband Digital Control Medical Military Optical Networking Security Telephony Video & Imaging Wireless www.ti.com/audio www.ti.com/automotive www.ti.com/broadband www.ti.com/digitalcontrol www.ti.com/medical www.ti.com/military www.ti.com/opticalnetwork www.ti.com/security www.ti.com/telephony www.ti.com/video www.ti.com/wireless Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265 Copyright (c) 2009, Texas Instruments Incorporated