CMX882E1 - CML Microcircuits - Datasheet.Directory

CML Microcircuits

COMMUNICATION SEMICONDUCTOR

CMX882

Baseband Processor for

‘Leisure’ Radios with Data

D/882/8 October 2004 Provisional Issue

Full-Feature Audio-Processing, Signalling and GPS Data for

Half Duplex FRS, MURS, PMR446 and GMRS ‘Leisure’ Radios

Features

• Automatic signal type scanning and IRQ on

detection of valid Rx signals, level or RSSI • Voice processing facilities, including Tx and Rx

gain setting and voice/subaudio filtering

• Tone generator for caller recognition tunes • C-BUS serial host interface

• Programmable power down control • RF interface allowing 1 or 2 point modulation

• Programmable signal detection thresholds • Programmable soft limiter

• Low Power operation with Zero Power mode • Enhanced CTCSS and 23/24 bit DCS codecs

• Uncommitted Aux ADC with switchable input

to monitor signals • Zero ‘talk down’ CTCSS decoder performance

prevents dropouts

• Silent operation by removal of unwanted calls • ‘All call code’ and ‘monitor’ modes for CTCSS

• Selectable voice companding • Audio scrambling

• XTCSS channel and data signalling

• Robust Half Duplex FFSK/MSK modem,

1200/2400 bps, gets data through when signal

is too degraded for voice – for text

messaging/paging, location transfer, etc.

applications.

• Robust automatic packet data protocol with:

Error correction and detection, Interleaving and

data scrambling/privacy coding.

FRS Signalling Processor CMX882

1.1 Brief Description

CMX882, a full-function half-duplex audio and signalling processor IC for FRS and PMR446 type facilities

suitable for both complex and simple end-designs. Under the control of the host µC, all voiceband

requirements are catered for: voiceband and sub-audio filtering, pre/de-emphasis, compression and

expansion and audio routing and global level setting with single or two-point modulation in the transmit

path.

The combination of new and standard signalling functions of this product offer, under software control,

increased functionality, versatility and privacy. Standard Extended-Code CTCSS and DCS functions are

integrated with the new XTCSS code implementation. XTCSS provides additional and improved squelch-

centred privacy codes with the added advantage of ‘silent operation’; no annoying interference from other

sub-audio users. XTCSS fitted radios enjoy more privacy and flexibility of operation.

For advanced and enhanced radio operation, the CMX882 provides a 1200/2400 bps free-format and

packet data FFSK/MSK modem for text messaging/paging, passing GPS location data (compatible with

NMEA 0183) and other data applications.

With ultra low power requirements and graduated powersave, this product only requires a smaller, lower-

power µC than existing FRS/PMR446 solutions. It is available in compact SSOP and TSSOP packages.

FRS Signalling Processor CMX882

CMX882 Functions and Facilities

Half Duplex Operation

Working in a half duplex mode, when the product is in Tx the Rx sections can be powered down to extend

battery life, conversely in Rx major sections of the Tx can be treated in the same manner.

Serial Control and Data Interfaces

C-BUS: Serial control, data and command program interface compatible with SCI, SPI and Microwire type

interfaces.

Power Requirements and Economy

With an ultra low power requirement, the CMX882 operates from a single 2.7 to 3.6 Volt supply with

graduated ‘Sleep Mode’ powersaving facilities for both Rx and Tx modes.

FRS Signalling Processor CMX882

Signalling:

XTCSS

A state-of-the-art (squelch) signalling format, employing both sub-audio (CTCSS) and in-band (XTC)

signalling concurrently, which offers more than twice as many privacy codes as standard CTCSS

operation and completely eliminates interference caused by other traffic on the channel (quiet operation).

Additionally the XTCSS signalling can be employed as an over-air control for such features as voice-

compression. XTCSS is fully compatible with both conventional and enhanced CTCSS signalling

operations and will implement the All Call Code function.

CTCSS

’Zero talkdown’ performance eliminates unwanted breaks in communication. The CMX882 is pre-

programmed with 39 standard CTCSS (+ Notone and DCS ‘turn off’ tone) and 12 additional ‘split-tone’

frequencies. Any one of these can be selected for reception or transmission. Decoding is aided by the

use of adjustable decode bandwidths and threshold levels. Decoding is carried out rapidly thus avoiding

the loss of the beginning of speech or data signals. A CTCSS configuration of this product enables ‘Tone

Cloning’.

Two unique features of this product are its CTCSS ‘All Call Code’ and ‘All Codes Monitor’ modes:

All Call Code – transmissions using this code will be heard by all CMX882 enhanced radios regardless of

their selected CTCSS code. This provides an important benefit to both safety and convenience.

All Codes Monitor – selection of this code at the receiver enables all transmissions that are using a

CTCSS tone to be heard, and the tone number to be reported. Open channel noise or calls lacking

coding, will go unheard. This is a superior method of ‘channel monitoring’, which allows miscoded calls

from conventional CTCSS-party radios to be heard and directly responded to.

FRS Signalling Processor CMX882

DCS

The DCS code is in NRZ format and is transmitted at 134.4b/s in either 23 or 24 bit patterns. The code,

for transmission or reception is programmed via the host µC with the ‘turn off’ tone being supplied from

the CTCSS facility. Decoding is carried out rapidly thus avoiding the loss of the beginning of speech or

data signals.

FFSK/MSK Modem

This high performance modem facilitates data transfers (e.g. text messaging/paging caller identification,

caller location, digital poll of remote radio location, GPS information, etc.) by providing a 1200/2400 bps

MSK data packet encoder (suitable for NMEA 0183 data transfer). Software selectable interleaving, error

correction (FEC), error checking (CRC) and data scrambling functions provide multiple communication

types to select from robustness, speed and privacy options. In the Rx path, a 1200/2400 bps data packet

decoder features automatic bit-rate recognition, 16-bit frame-sync detector, error correction and checking,

data de-scrambling and packet disassembly. High performance allows the modem to be used even when

radio link quality is too degraded for voice communications.

Modulation type is optimised to maximise robustness and data rate and simplify hardware connections to

voice radio Tx and Rx circuits with virtually no modification. The modulation type also is appropriate to

develop F2D type emissions, as required by certain regulations. (F2D = frequency modulation, digital

modulation with modulated subcarrier, transmitted information is data/telemetry/telecommand).

Signal Monitor

An auxiliary circuit intended for the monitoring of any signal or level; both internal and external. This

function can be used in conjunction with the host µC to allow such activities as: VOX operation and/or the

‘wake-up’ of powered-down circuitry.

Audio Processing:

Adjustable Gain Input Amplifiers

Selectable, component adjustable inputs are available for microphone or line voiceband or discriminator

inputs. In either mode (Tx or Rx) the selected input can be further level adjusted under the control of the

host µC prior to signal or audio- processing.

Voiceband and Sub-Audio Filtering with Limiting

Both Rx and Tx paths present voiceband filtering; the Tx path filter can be configured to either 12.5 or 25

kHz channel spacing whilst the Rx path also includes a sub-audio passband filter.

Voiceband Pre-emphasis and De-emphasis

Voiceband pre-emphasis is selectable to either 12.5 or 25 kHz channel configurations in the Tx path; de-

emphasis at -6dB/ octave is selectable in the Rx path.

Software Adjustable Gains, Volume, Mixing and Routing

Providing total flexibility of operation, this product, under µC control has the ability to select and route

functions and audio and signal paths, set bandwidths and threshold levels, mix audio and sub bands and

vary both input and out gain/attenuation levels. Output levels from all analogue ports can be ‘ramped’ up

and down at independently programmed rates.

Attenuation-Adjustable Single/Two-Point Modulation Outputs

To facilitate a wide range of transmitter types, the CMX882 has the ability to provide, independently

programmable, modulation outputs; for single or two-point modulation schemes.

Scrambler

An optional frequency inversion scrambler is provided in both transmit and receive modes.

FRS Signalling Processor CMX882

CONTENTS

Section Page

1.0 Features.............................................................................................................................2

1.1 Brief Description.................................................................................................................2

1.2 Block Diagram....................................................................................................................8

1.3 Signal List...........................................................................................................................9

1.4 External Components.......................................................................................................11

1.4.1 PCB Layout Guidelines and Power Supply Decoupling.....................................12

1.4.2 Modulator Outputs...............................................................................................13

1.5 General Description .........................................................................................................14

1.5.1 Sleep Mode and Auto Start Up...........................................................................16

1.5.2 Auxiliary ADC......................................................................................................16

1.5.3 Receive Mode.....................................................................................................17

1.5.4 Transmit Mode....................................................................................................26

1.5.5 FFSK/MSK Data Packeting.................................................................................30

1.5.6 XTCSS Coding....................................................................................................32

1.5.7 C-BUS Operation................................................................................................35

1.6 C-BUS Register Description ............................................................................................36

1.7 Application Notes.............................................................................................................59

1.7.1 CRC and FEC encoding information ..................................................................59

1.7.2 Data Interleaving.................................................................................................60

1.7.3 Scrambler Seed Transfer....................................................................................60

1.8 Performance Specification...............................................................................................61

1.8.1 Electrical Performance........................................................................................61

1.8.2 Packaging ...........................................................................................................71

Table Page

Table 1 Voice Processing Combinations.......................................................................................15

Table 2 Concurrent Rx Signalling Modes Supported by the CMX882..........................................17

Table 3 CTCSS Tones..................................................................................................................21

Table 4 DCS Modulation Modes...................................................................................................22

Table 5 DCS 23 Bit Codes............................................................................................................23

Table 6 In-band Tones..................................................................................................................24

Table 7 Concurrent Tx Modes Supported by the CMX882...........................................................26

Table 8 Data Frequencies for each Baud Rate ............................................................................30

Table 9 Maximum Data Transfer Latency.....................................................................................32

Figure Page

Figure 1 Block Diagram...................................................................................................................8

Figure 2 Recommended External Components............................................................................11

Figure 3 Power Supply Connections and De-coupling.................................................................12

Figure 4 Modulator output components to achieve -100dB/decade roll-off..................................13

Figure 5 Rx Audio Filter Frequency Response.............................................................................19

Figure 6 De-emphasis Curve for TIA/EIA-603 Compliance..........................................................19

Figure 7 Low Pass Sub-Audio Band Filter for CTCSS and DCS..................................................20

FRS Signalling Processor CMX882

Figure 8 25kHz Channel Audio Filter Response Template ..........................................................27

Figure 9 12.5kHz Channel Audio Filter Response Template .......................................................27

Figure 10 Audio Frequency Pre-emphasis Template...................................................................28

Figure 11 Modulating Waveforms for 1200 and 2400 Baud FFSK/MSK Signals.........................29

Figure 12 C-BUS Transactions.....................................................................................................35

Figure 13 Possible FRS + GPS Configuration..............................................................................59

Figure 14 C-BUS Timing...............................................................................................................69

Figure 15 Typical FFSK/MSK Bit Error Rate Graph .....................................................................70

Figure 16 Mechanical Outline of 28-pin SSOP (D6): Order as part no. CMX882D6..................71

Figure 17 Mechanical Outline of 28-pin TSSOP (E1): Order as part no. CMX882E1................71

It is always recommended that you check for the latest product datasheet version from the Datasheets

page of the CML website: [www.cmlmicro.com].

FRS Signalling Processor CMX882

1.2 Block Diagram

Figure 1 Block Diagram

FRS Signalling Processor CMX882

1.3 Signal List

Package

D6, E1 Signal Description

Pin No. Name Type

23 VDD(D) Power The digital positive supply rail. This pin should be decoupled

to VSS(D) by a capacitor mounted close to the device pins.

5 VSS(D) Power The negative supply rail (digital ground).

18 VDD(A) Power The analogue positive supply rail. Levels and thresholds

within the device are proportional to this voltage. This pin

should be decoupled to VSS(A) by a capacitor mounted

close to the device pins.

9, 21 VSS(A) Power The negative supply rail. Both pins must be connected to

analogue ground.

1, 2 NC No connection should be made to these pins.

3 IRQN O/P A 'wire-ORable' output for connection to the Interrupt

Request input of the host. This output is pulled down to

VSS(D) when active and is high impedance when inactive.

An external pull-up resistor is required.

4 REPLY_DATA T/S The C-BUS serial data output to the host. This output is held

at high impedance when not sending data to the host.

6 SERIAL_CLOCK I/P The C-BUS serial clock input from the host.

7 CMD_DATA I/P The C-BUS serial data input from the host.

8 CSN I/P The C-BUS data loading control function. Data transfer

sequences are initiated, and completed by the CSN signal.

FRS Signalling Processor CMX882

1.3 Signal List (continued)

Package

D6, E1 Signal Description

Pin No. Name Type

10 VBIAS O/P Internally generated bias voltage of approximately VDD(A)/2,

except when bias is power-saved when VBIAS will discharge

to VSS(A). This pin should be decoupled to VSS(A) by a

capacitor mounted close to the device pins.

11 DISC I/P Input terminal of discriminator input amplifier.

12 DISC_FB O/P Output / feedback terminal of discriminator input amplifier.

13 INPUT_2 I/P Input terminal of amplifier 2, for either a second microphone

or discriminator input.

14 INPUT_2_FB O/P Output / feedback terminal of input amplifier 2.

15 MIC I/P Input terminal of microphone input amplifier.

16 MIC_FB O/P Output / feedback terminal of microphone input amplifier.

17 SIG_MONITOR I/P Signal Monitor input to the internal level detecting circuit.

19 MOD_1 O/P Modulator 1 output.

20 MOD_2 O/P Modulator 2 output.

22 AUDIO O/P Output of the audio section.

24 CLOCK/XTAL I/P The input to the on-chip oscillator for an external crystal or a

clock circuit.

25 CLOCK_OUT O/P Buffered (un-inverted) clock output available for use by other

devices in the system.

26 I/P Test input, connect to VSS(D).

27, 28 NC No connection should be made to these pins.

Notes: I/P = Input

O/P = Output

T/S = 3-state Output

NC = No Connection

FRS Signalling Processor CMX882

1.4 External Components

VDD(D)

R1 NC 128

NC 227

IRQN 326

VSS(D) X1

REPLY_DATA 425

CLOCK_OUT

C- BUS VSS(D) 524

CLOCK/XTAL VSS(D)

Interface SERIAL_CLOCK 623

VDD(D) C2

CMD_DATA 722

AUDIO R2 Loudspeaker amp

Discriminator CSN 821

VSS(A)

VSS(A) 920

MOD_2 R3 Modulator reference

C8 R5 C5 VBIAS 10 19 MOD_1 R4 Modulator control

DISC 11 18 VDD(A) C1 C3 C4

DISC_FB 12 17 SIG_MONITOR

R6 INPUT_2 13 16 MIC_FB VSS(A)

INPUT_2_FB 14 15 MIC

C7 C9* D1* R10* C10*

C6 R7 R8 R9*

Microphone VSS(A)

CMX882

Figure 2 Recommended External Components

R1 100kΩ R9 See note 6 C6 See note 4

R2 100kΩ R10 See note 6 C7 200pF

R3 100kΩ C8 See note 4

R4 100kΩ C1 100pF C9/10 See note 6

R5 See note 2 C2 1nF

R6 100kΩ C3 100pF X1 18.432MHz See note 1

R7 See note 3 C4 100pF

R8 100kΩ C5 100pF D1 See note 6

Resistors ±5%, capacitors and inductors ±20% unless otherwise stated.

Notes:

1. X1 can be a crystal or an external clock generator; this will depend on the application. The clock drift

requirement is defined in section 1.8.1. The tracks between the crystal and pin 24 and pin 5 should

be as short as possible to achieve maximum stability and best start up performance.

2. R5 should be selected to provide the desired dc gain (assuming C8 is not present) of the

discriminator input, as follows:

⏐GAINDisc⏐ = 100kΩ / R5

The gain should be such that the resultant output at the DISC_FB pin is within the discriminator input

signal range specified in section 1.8.1.

3. R7 should be selected to provide the desired dc gain (assuming C6 is not present) of the microphone

input as follows: ⏐GAINMic⏐ = 100kΩ / R7

The gain should be such that the resultant output at the MIC_FB pin is within the microphone input

signal range specified in section 1.8.1.

4. C6 and C8 should be selected to maintain the lower frequency roll-off of the microphone and

discriminator inputs as follows: C6 = 30nF × ⏐GAINMic⏐

C8 = 1µF × ⏐GAINDisc⏐

FRS Signalling Processor CMX882

5. INPUT_2 and INPUT_2_FB connections allow the user to have a second discriminator or

microphone input. Component connections and values are as for the networks around pins 11 and

12 or pins 15 and 16 respectively. If this input is not required pin 13 must be connected to pin 14.

6. The circuit formed by D1, C9, C10, R9 and R10 is a peak detector, this is only required when the

signal monitor is connected to an ac signal (e.g. microphone or received signal). For a dc type signal

(e.g. RSSI) these components are not required. The values of C9 and R10 set the attack time, C9

and R9 set the decay time. D1 can be any suitable small signal diode. R10 should be a high enough

value so as not to distort the signal source.

1.4.1 PCB Layout Guidelines and Power Supply Decoupling

Digital G round

128 Digital

227C11C12

ground plane

326 +

425 Cloc k O ut put

VSS(D) 524 L1

623

V DD(D) Digital + ve Suppl y

provis ion for 7 22

wire link 821

V S S (A) A nalogue Ground

VSS(A) 920

C13 C14

VBIAS 10 19 +L2

C15 11 18 VDD(A ) Anal ogue + ve S upply

12 17

13 16 Analogue

14 15 ground plane

CMX882

Figure 3 Power Supply Connections and De-coupling

C11 10nF C14 10µF L1 100nH See note 7

C12 10µF C15 100nF L2 100nH See note 7

C13 10nF

Resistors ±5%, capacitors and inductors ±20% unless otherwise stated.

Notes:

7. The inductors L1 and L2 can be omitted but this may degrade system performance.

It is important to protect the analogue pins from extraneous inband noise and to minimise the impedance

between the CMX882 and the supply and bias de-coupling capacitors. The de-coupling capacitors C11,

C12, C13 and C14 should be as close as possible to the CMX882, particularly C11 and C13. It is

therefore recommended that the printed circuit board is laid out with separate ground planes for the

VSS(A) and VSS(D) in the area of the CMX882, with provision to make a link between them close to the

CMX882.

VBIAS is used as an internal reference for detecting and generating the various analogue signals. It must

be carefully decoupled, to ensure its integrity, so apart from the decoupling capacitor shown, no other

loads should be connected. If VBIAS needs to be used to set the discriminator mid-point reference, it must

be buffered with a high input impedance buffer.

The single ended microphone input(s) and audio output must be ac coupled as shown, so their return

paths can be connected to VSS(A) without introducing dc offsets. Further buffering of the audio output is

advised.

The crystal X1 can be replaced with an external clock source if required/desired. The internal clock

generating circuit can be placed in power-save mode if the clock is provided externally.

FRS Signalling Processor CMX882

1.4.2 Modulator Outputs

The combination of CMX882 and the modulator output components, R3/C3 and R4/C4, achieve roll-off

rates better than –60dB/decade. If required this can be increased to better than –100dB/decade by

replacing R3/C3 and R4/C4 with the active filter circuit shown in Figure 4.

C17/C19

MOD_1/2 +

R11/R13 R12/R14 C16/C18 Modul ator 1 & 2

VSS(A)

Figure 4 Modulator output components to achieve -100dB/decade roll-off

R11 120kΩ C16 220pF

R12 120kΩ C17 440pF (2 x C16)

R13 120kΩ C18 220pF

R14 120kΩ C19 440pF (2 x C18)

Resistors ±5%, capacitors and inductors ±20% unless otherwise stated.

Notes:

8. The external op-amp must be chosen to ensure that the required signal output level can be driven

within acceptable distortion limits.

FRS Signalling Processor CMX882

1.5 General Description

The CMX882 is intended for use in half duplex analogue two way land mobile radio (LMR) equipment and

is particularly suited to enhanced MURS / GMRS / FRS with GPS terminal designs. The CMX882

provides radio signal encoder and decoder functions for: Voice, in-band tones, XTCSS, CTCSS, DCS and

FFSK/MSK data permitting simple to sophisticated levels of tone control and data transfer. Power control

facilities allow the device to be placed in varying levels of sleep allowing the user to fine tune the power

depending on system requirements. The CMX882 includes a crystal clock generator, with buffered

output, to provide a common system clock if required. A block diagram of the CMX882 is shown in Figure

Tx functions

Audio

o Single/dual microphone inputs with input amplifier and programmable gain adjustment

o Filtering selectable for 12.5kHz and 25kHz channels

o Selectable pre-emphasis

o Selectable compression

o Selectable frequency inversion voice scrambling

o 2-point modulation outputs with programmable level adjustment

Signalling

o Pre-programmed 51 tone CTCSS encoder

o Programmable 23/24bit DCS encoder

o Programmable audio tone generator (for custom audio tones)

o Pre-programmed XTCSS and in-band tone encoder

o 1200/2400bps MSK data packet encoder (suitable for text messaging/paging, caller

identification, caller location, digital poll of remote radio location, GPS information via

NMEA 0183 data transfer etc.), incorporating interleaving, FEC, CRC and data scrambler

Rx functions

Audio

o Single/dual demodulator inputs with input amplifier and programmable gain adjustment

o Voice-band and sub-audio rejection filtering

o Selectable de-emphasis

o Selectable expansion

o Selectable frequency inversion voice de-scrambling

o Software volume control

Signalling

o 1 from 51 CTCSS decoder + Tone Clone mode

o 23/24bit DCS decoder

o Pre-programmed in-band tone decode with XTCSS 4 tone addressing

o 1200/2400bps MSK data packet decoder with automatic bit rate recognition, 16 bit frame

sync detector, error correction, data de-scrambler and packet disassembly

o Signal Monitor (RSSI / Microphone / Rx channel level monitor)

FRS Signalling Processor CMX882

Voice Processing Combinations

Table 1 shows the valid voice processing combinations.

Table 1 Voice Processing Combinations

TX RX

Compress Scramble Pre-Emphasis Filter Filter De-Emphasis De-Scramble Expand

1 3 3

2 3 3 3 3

3 3 3 3 3 3 3

4 3 3 3 3

5 3 3 3 3 3 3

6 3 3 3 3

Not Valid

7 3 3 3 3 3 3 3 3

Not Recommended

81 3 3 3 3 3 3

Host Interface

A serial data interface (C-BUS) is used for command, status and data transfers between the CMX882 and

the host µC; this interface is compatible with microwire, SPI etc. Interrupt signals notify the host µC when

a change in status has occurred and the µC should read the status register across the C-BUS and

respond accordingly. Interrupts only occur if the appropriate mask bit has been set. See section 1.6.15.

Auxiliary (Signal Monitor) analogue signal

The CMX882 includes a Signal level monitor. This is an 8-bit successive approximation ADC and a two

level signal sensor. The two level sensor facility can be used in conjunction with the power saving mode

to wake up powered down blocks, and issue an interrupt on the IRQN line when the Signal exceeds the

preset threshold level. The auxiliary ADC voltage reference is taken directly from the VDD(A) supply, so

the Signal level being monitored should be derived from this supply voltage.

1 Audio quality is somewhat degraded.

FRS Signalling Processor CMX882

1.5.1 Sleep Mode and Auto Start Up

Power-on reset or C-BUS general reset places the CMX882 into sleep mode, which results in all internal

blocks, except the xtal clock circuit, being placed in power-saved mode. The xtal clock circuit can be

power-saved but this must be done by an explicit C-BUS command. Power saving is achieved by turning

off bias current sources or disabling local clocks, as appropriate.

During system standby periods, parts of the device can be put into sleep mode by the host to conserve

power. The Auxiliary ADC can be programmed so that when the level exceeds a threshold, an interrupt is

issued over the C-BUS and the selected mode (Tx or Rx) “woken up” within 400µs. If this time is too long

to ensure no part of the signal is lost, the DISC or MIC input and ADC path can be kept powered up whilst

in standby mode. The receive modes and transmit modes can also be activated by commands from the

C-BUS. On wake up, activation of the various signal path stages are phased appropriately to avoid

causing unwanted transients. More details are provided in section 1.6.4 on Signal Routing.

The CMX882 can be programmed to wake up its receive path automatically (automatic start-up) when the

DISC input level exceeds the ‘high’ level threshold. While the CMX882 is in automatic receive start-up

mode the DISC input must also be selected for the signal path. When not in automatic start-up mode it is

recommended that the required input is selected during Auxiliary ADC operation to avoid subsequent

switching of the input signal source.

1.5.2 Auxiliary ADC

This section of the CMX882 operates in both Tx and Rx modes and can be used to monitor one of 4

signal sources: Sig_Monitor pin, MIC1, Input_2 or DISC inputs. Activity on the selected input will

optionally issue an interrupt if host intervention is required. During idle periods the majority of the

CMX882 can be placed into low power mode. If monitoring ac signals connected to the Sig_Monitor pin

they must be rectified and filtered using passive external circuitry.

The Auxiliary ADC facility comprises an 8-bit ADC, a comparator, an 8-bit result data word and two 8-bit

threshold registers, one defining the ‘Signal high’ level and the other the ‘Signal low’ level. The two

threshold registers are combined into one 16-bit C-BUS register word. The ADC measures the Signal

level at intervals that are set by C-BUS command. It is advised that the interval be set to <125µs while

waiting for a new incoming signal so that the CMX882 and host µC can be powered up and put into the

correct mode in time to avoid missing any part of the signal. The default interval period following a reset

is 20.8µs. Power dissipation of the Auxiliary ADC can be reduced by increasing the conversion interval

time.

The result of the most recent Auxiliary ADC measurement can be read over the C-BUS whenever the

Signal Processing and Aux ADC circuits are powered up.

The Auxiliary ADC compares each conversion result with the values in the ‘Signal high’ or the ‘Signal low’

threshold registers. The CMX882 can, for example, issue an interrupt to the host µC to wake up the

receive path when the Auxiliary ADC input exceeds the ‘high’ level threshold. The CMX882 can also

issue an interrupt to the host µC to indicate a weak or absent signal when it falls below the ‘low’ level

threshold. This provides a user programmable hysteresis facility. The host must ensure that the value in

the ‘low’ register is always less than that of the ‘high’ register. The options for issuing interrupts and for

automatic start-up are selected by C-BUS command.

The Auxiliary ADC options are controlled by the $B2, $B3 and $C0 C-BUS registers.

The Auxiliary ADC requires the Auxiliary ADC, BIAS and Xtal clock to all be enabled in the Power Down

Control register.

FRS Signalling Processor CMX882

1.5.3 Receive Mode

The CMX882 can receive voice and various signal formats: CTCSS tone, DCS code, XTCSS / In-band

tones and FFSK/MSK data at 1200 and 2400bps. Reception of each of these signal types can be

independently enabled/disabled by C-BUS command. If enabled, an interrupt will be issued to notify the

host µC of the presence and type of the incoming signal.

In receive mode the CMX882 performs signal type identification in 2 frequency bands, sub-audio (60 -

260Hz) and voice band (300 - 3kHz), to determine what type of signal is being received. When an

enabled signal is detected this will be indicated to the host over the C-BUS and the CMX882 will continue

to process the received signal in its band. Identification / process mode will continue in the other band.

The CMX882 can process voice and simultaneously identify and process at least 2 other signal types

(one in the sub-audio in parallel with one in the voice band). See Table 2 for valid combinations. These

combinations can be used with Voice Processing, if desired.

The receive gain and audio output amplifier gain can be adjusted by the host µC, via C-BUS command, to

provide receive signal level adjustment and output volume control or muting.

Table 2 Concurrent Rx Signalling Modes Supported by the CMX882

Sub-Audio

All combinations of: Voice band signalling

Any one of A - C:

With Rx Voice

Processing1 or

Audio Tone

generation

DCS

Inverted DCS

CTCSS

None

XTCSS3, 4

1200bps FFSK/MSK

Sub-Audio

All combinations of: Voice band signalling

Any one of A:

With Rx Voice

Processing2

DCS

Inverted DCS

CTCSS A: None

Sub-Audio

All combinations of: Voice band signalling

Any one of A - G:

No Voice

Processing or

Audio Tone

generation

DCS

Inverted DCS

CTCSS

None

XTCSS3

1200bps FFSK/MSK

2400bps FFSK/MSK

1200 & 2400 bps FFSK/MSK

XTCSS3 & 1200bps FFSK/MSK

XTCSS3 & 2400bps FFSK/MSK

Sub-Audio

All combinations of: Voice band signalling

Any one of A - H:

No Voice

Processing or

Audio Tone

generation

No Subaudio

processing

None

XTCSS3

1200bps FFSK/MSK

2400bps FFSK/MSK

1200 & 2400 bps FFSK/MSK

XTCSS3 & 1200bps FFSK/MSK

XTCSS3 & 2400bps FFSK/MSK

XTCSS3 & 1200bps & 2400bps FFSK/MSK

1 Including optional de-emphasis, but excluding voice companding and scrambling

2 Including voice companding and scrambling

3 XTCSS or In-band tone signalling.

FRS Signalling Processor CMX882

4 XTCSS or In-band tone can only be used with DCS disabled and, if enabled, CTCSS not operating in

tone-clone mode.

By disabling all the decoding modes, the device can be configured to receive voice only signals with no

decoding of the voice band, CTCSS or DCS signalling. This will result in reception of all signals as if they

are voice. In this case it is up to the user/host µC to respond appropriately to incoming signals.

The CMX882 operates in half duplex, so whilst in receive mode the transmit path (microphone input and

modulator output amplifiers) can be disabled and powered down if required. The AUDIO output signal

level is equalised (to VBIAS) before switching between the audio port and the modulator ports, to minimise

unwanted audible transients. The Off/Power-save level of the modulator outputs is the same as the VBIAS

pin, so the audio output level must also be at this level before switching.

FRS Signalling Processor CMX882

1.5.3.1 Receiving Voice Band Signals

When a voice based signal is being received, it is up to the µC, in response to signal status information

provided by the CMX882, to control muting/enabling of the voice band signal to the AUDIO output.

The discriminator path through the device has a programmable gain stage. Whilst in receive mode this

should normally be set to 0dB (the default) gain.

Receive Filtering

The incoming signal is filtered, as shown in Figure 5, to remove sub-audio components and to minimise

high frequency noise. When appropriate the voice signal can then be routed to the AUDIO output.

Frequency (Hz)

-60

-50

-40

-30

-20

-10

10Hz 100Hz 1000Hz 10000Hz 100000Hz

Tem

late

Filter response

250Hz

300Hz 3kHz

Figure 5 Rx Audio Filter Frequency Response

De-emphasis

Optional de-emphasis at -6dB per octave from 300Hz to 3000Hz (shown in Figure 6) can be selected to

facilitate compliance with TIA/EIA-603.

-20

-16

-12

-8

-4

100 1000 10000

Frequency (Hz)

0dB/octave

+1dB

0dB

3dB

-18dB/octave

-6dB/octave

Figure 6 De-emphasis Curve for TIA/EIA-603 Compliance

FRS Signalling Processor CMX882

Rx Companding (Expanding)

The CMX882 incorporates an optional syllabic compandor in both transmit and receive modes. This

expands received voice band signals that have been similarly compressed in the transmitter to enhance

dynamic range. The compandor attack, decay and 0dB point are defined in section 1.8.1. See section

1.6.9 for details of how to control this function.

Audio De-Scrambling

The CMX882 incorporates an optional frequency inversion de-scrambler in receive mode. This de-

scrambles received voice band signals that have been scrambled in the transmitter. See section 1.6.9 for

details of how to control this function.

Voice Processing Combinations

Table 1 shows the valid voice processing combinations. (See section 1.5).

1.5.3.2 Receiving and Decoding CTCSS Tones

The CMX882 is able to accurately detect valid CTCSS tones quickly to avoid losing the beginning of voice

or possibly data transmissions, and is able to continuously monitor the detected tone with minimal

probability of falsely dropping out. The received signal is filtered in accordance with the template shown

in Figure 7, to prevent signals outside the sub-audio range from interfering with the sub-audio tone

detection.

Figure 7 Low Pass Sub-Audio Band Filter for CTCSS and DCS

Once a valid CTCSS tone has been detected, the voice band signal can be passed to the audio output.

The voice band signal is extracted from the received signal by band pass filtering as shown in Figure 5.

To help decode received CTCSS tones adjustable decoder bandwidths and threshold levels permit

decode certainty and signal to noise performance to be traded when congestion or range limits the

system performance. This entails setting the tone decoder bandwidth and threshold level in P2.1 of the

Programming register ($C8) and programming the Audio & Tx CTCSS Control register with the desired

tone.

Tone Cloning™:

Tone CloningTM facilitates the detection of CTCSS tones 1 to 39 in receive mode. This allows the device

to non-predictively detect any tone in this range. The received tone number will be reported in the Tones

Status register. This tone code can then be programmed into the ‘Audio and Device Address Control’

™ Tone Cloning is a trademark of CML Microsystems Plc.

-70

-60

-50

-40

-30

-20

-10

0 200 400 600 800 1000

Frequency (Hz)

Gain (dB)

FRS Signalling Processor CMX882

Tone cloning should not be used in systems where tones 41 to 51 or other split tones (tones between the

frequencies of tones 1 to 40) may be received. The all call tone 40 can still be used after tone cloning has

been performed.

CTCSS Tones

Table 3 lists the CTCSS tones available. The tone numbers are decimal equivalents of the numbers

written to the Audio & Device Address Control register ($C2) and reported in the Tone Status register

($CC). Table 3 CTCSS Tones

Tone

Number Freq.

(Hz) Tone

Number Freq.

(Hz) Tone

Number Freq.

(Hz)

001 No Tone 20 131.8 402 62.5 *

01 67.0 21 136.5 41 159.8 *

02 71.9 22 141.3 42 165.5 *

03 74.4 23 146.2 43 171.3 *

04 77.0 24 151.4 44 177.3 *

05 79.7 25 156.7 45 183.5 *

06 82.5 26 162.2 46 189.9 *

07 85.4 27 167.9 47 196.6 *

08 88.5 28 173.8 48 199.5 *

09 91.5 29 179.9 49 206.5 *

10 94.8 30 186.2 50 229.1 *

11 97.4 31 192.8 51 254.1 *

12 100.0 32 203.5 52-54 Reserved

13 103.5 33 210.7 553 Invalid

14 107.2 34 218.1 tone

15 110.9 35 225.7 >=56 Reserved

16 114.8 36 233.6

17 118.8 37 241.8

18 123.0 38 250.3

19 127.3 39 69.3

* Subaudio tone not in TIA-603A standard.

Notes:

1 Tone number 00 in the Tone Status register ($CC) indicates that none of the above subaudio

tones is being detected - see also section 1.6.19. If tone number 00 is programmed into the

Audio & Device Address Control register ($C2) only tone 40 will be scanned for - see note 2. If

CTCSS transmit is selected this tone setting will cause the CTCSS generator to output no signal.

2 Tone number 40 provides an all user CTCSS tone option; regardless of the subaudio tones set,

the CMX882 will indicate to the host when this tone is detected whenever the CTCSS detector is

enabled. This feature is useful for implementing emergency type calls e.g. all call.

3 Tone number 55 is reported in the Tone Status register ($CC), when CTCSS receive is enabled

and a subaudio tone is detected that does not correspond to the selected tone or the all-call tone

(tone number 40). This could be a tone in the subaudio band which is not in the table or a tone

in the table which is not the selected tone or all-call tone.

FRS Signalling Processor CMX882

1.5.3.3 Receiving and Decoding DCS Codes

DCS Code is in NRZ format and transmitted at 134.4 ±0.4bps. The CMX882 is able to decode any 23 or

24 bit pattern in either of the two DCS modulation modes defined by TIA/EIA-603 and described in Table

4. The CMX882 can detect a valid DCS Code quickly enough to avoid losing the beginning of voice

transmissions.

Table 4 DCS Modulation Modes

Modulation Type: Data Bit: FM Frequency Change:

A 0 Minus frequency shift

1 Plus frequency shift

B 0 Plus frequency shift

1 Minus frequency shift

The CMX882 detects the DCS code that matches the programmed code defined in the ‘DCS Code’ words

(P2.2-3) in the Programming register ($C8).

To detect the pre-programmed DCS code the signal is low pass filtered to suppress all but the sub-audio

band using the filter shown in Figure 7. Further equalisation filtering, signal slicing and level detection are

done to extract the code being received. The extracted code is then matched with the programmed 23 or

24-bit DCS code to be recognised, in the order least significant first through to most significant DCS code

bit last. Table 5 shows a selection of valid 23-bit DCS codes, this does not preclude other codes being

programmed. Recognition of a valid DCS Code will be flagged if the decode is successful (3 or less

errors). A failure to decode is indicated by a '0' flag. This flag is updated after the decoding of every 4th

bit of the incoming signal.

Once a valid DCS Code has been detected, the voice band signal can be passed to the AUDIO output

under the control of the host µC. The voice signal is extracted from the received input signal by band

pass filtering; see Figure 5. More details for programming DCS Codes are provided in section 1.6.20.3.

The end of DCS transmissions is indicated by a 134.4 ±0.5Hz tone for 150-200ms. To detect the DCS

turn off tone while receiving DCS, the DCS turn off tone option must be selected in the Audio and CTCSS

Control ($C2) register and CTCSS receive must also be enabled. When a DCS turn off tone is detected it

will cause a DCS interrupt; the receiver audio output can then be muted by the host.

FRS Signalling Processor CMX882

Table 5 DCS 23 Bit Codes

DCS

Code

DCS

bits

22-12

DCS

bits

11-0 DCS

Code

DCS

bits

22-12

DCS

bits

11-0 DCS

Code

DCS

bits

22-12

DCS

bits

11-0

023 763 813 174 18B 87C 445 7B8 925

025 6B7 815 205 6E9 885 464 27E 934

026 65D 816 223 68E 893 465 60B 935

031 51F 819 226 7B0 896 466 6E1 936

032 5F5 81A 243 45B 8A3 503 3C6 943

043 5B6 823 244 1FA 8A4 506 2F8 946

047 0FD 827 245 58F 8A5 516 41B 94E

051 7CA 829 251 627 8A9 532 0E3 95A

054 6F4 82C 261 177 8B1 546 19E 966

065 5D1 835 263 5E8 8B3 565 0C7 975

071 679 839 265 43C 8B5 606 5D9 986

072 693 83A 271 794 8B9 612 671 98A

073 2E6 83B 306 0CF 8C6 624 0F5 994

074 747 83C 311 38D 8C9 627 01F 997

114 35E 84C 315 6C6 8CD 631 728 999

115 72B 84D 331 23E 8D9 632 7C2 99A

116 7C1 84E 343 297 8E3 654 4C3 9AC

125 07B 855 346 3A9 8E6 662 247 9B2

131 3D3 859 351 0EB 8E9 664 393 9B4

132 339 85A 364 685 8F4 703 22B 9C3

134 2ED 85C 365 2F0 8F5 712 0BD 9CA

143 37A 863 371 158 8F9 723 398 9D3

152 1EC 86A 411 776 909 731 1E4 9D9

155 44D 86D 412 79C 90A 732 10E 9DA

156 4A7 86E 413 3E9 90B 734 0DA 9DC

162 6BC 872 423 4B9 913 743 14D 9E3

165 31D 875 431 6C5 919 754 20F 9EC

172 05F 87A 432 62F 91A

1.5.3.4 Receiving and Decoding In-band Tones

In-band tones can be used to flag the start of a call or to confirm the end of a call. If they occur during a

call the tone may be audible at the receiver. When enabled, an interrupt will be issued when a signal

matching a valid In-band tone is detected and when a present In-band tone turns off or changes (i.e. at

the start and at the end of each In-band tone).

The CMX882 implements QTC coding using the EEA tone set. Other addressing and data formats can

be implemented but will require more host intervention. The custom tones (1-4) permit other audio tones

to be encoded or decoded, the frequency of each tone is defined in the program registers P1.2-5.

In receive the CMX882 scans through the tone table sequentially, the code reported will be the first one

that matches the incoming frequency.

Adjustable decoder bandwidths, threshold levels are programmable via the Programming register and

permits certainty of detection and signal to noise performance to be traded when congestion or range

limits the system performance. The In-band signal is derived from the received input signal after the band

pass filtering shown in Figure 5.

FRS Signalling Processor CMX882

Table 6 In-band Tones

Special / Information Tones

(5th bit = 0) Normal Tones

(5th bit = 1)

4 bit Code Freq. 4 bit Code Freq.

Dec Hex (Hz) Dec Hex (Hz)

0 0 No Tone 0 0 1981

1 1 Custom Tone 0 P1.21 1 1 1124

2 2 Custom Tone 1 P1.31 2 2 1197

3 3 Custom Tone 2 P1.41 3 3 1275

4 4 Custom Tone 3 P1.51 4 4 1358

5 5 5 5 1446

6 6 6 6 1540

7 7 7 7 1640

8 8 8 8 1747

9 9 9 9 1860

10 A 10 A 1055

11 B 11 B 930

12 C 12 C 2247

13 D 13 D 991

14 E

Reserved

14 E 21102

15 F Unrecognised tone 15 F Unrecognised tone

Notes:

1 Special tones 1-4 provide user programmable tone options for both transmit and receive modes

as set in the indicated Program register, for programming information see section 1.6.20.2.

2 Normal tone 14 is the repeat tone, this code must be used in transmit when the new code to be

sent is the same as the previous one. e.g. to send ‘333’ the sequence ‘3R3’ should be sent,

where ‘R’ is the repeat tone. When receiving Selcall tones the CMX882 will indicate the repeat

tone when it is received, it is up to the host to interpret this and decode tones accordingly.

FRS Signalling Processor CMX882

1.5.3.5 Receiving FFSK Signals

The CMX882 can decode incoming FFSK/MSK signals at either 1200 or 2400 baud data rates. It can

achieve this by deriving the baud rate from the received signal. Alternatively a control word may set the

baud rate, in which case the device only responds to signals operating at that rate. The form of

FFSK/MSK signals for these baud rates, excluding noise, is shown in Figure 11.

The received signal is filtered and data is extracted. A PLL is used to extract the clock from the recovered

serial data stream. The recovered data is stored in a 2 or 4 byte buffer (grouped into 16-bit words) and an

interrupt issued to indicate received data is ready. Data is transferred over the C-BUS, controlled by host

instructions. If this data is not read before the next data is decoded it will be overwritten. The MSK bit

clock is not output externally. It is up to the user to ensure that the data is transferred at an adequate rate

following data ready being flagged, see Table 9.

The extracted data is compared with the 16 bit programmed frame sync pattern. It is preset to $CB23

following a RESET command. An interrupt will be flagged when the programmed frame sync pattern is

detected or when the following Frame Head is decoded, see section 1.5.5. The host may stop the frame

sync search by disabling the MSK demodulator.

If set to decode a Frame Head before interrupting, the CMX882 will check the CRC portion of the Frame

Head Control Field. If this indicates a corrupt Frame Head a search for a new Frame Sync pattern will be

automatically restarted.

FFSK may be transmitted in conjunction with a CTCSS or DCS sub-audio component. The device will

handle the sub-audio signals as already described. If a sub-audio signal turns off during reception of

FFSK, it is up to the host µC to turn off the FFSK decoding as the device will continue receiving and

processing the incoming signal until commanded otherwise by the host µC.

The host must keep track of the message length or otherwise determine the end of reception (e.g. by

using sub-audio information or the Auxiliary ADC to check for signal level) and disable the FFSK

demodulator at the appropriate time. Note that when using packets with embedded size information the

CMX882 will indicate when the last data block has been received.

1.5.3.6 Receiving XTCSS Signals

The CMX882 can decode and monitor for XTCSS signalling. XTCSS is used to identify the start and

optionally the end of voice/data/other calls. It provides additional information and control over the basic

CTCSS method of channel coding.

XTCSS coding starts with a 4 tone sequence indicating the address and content of the following

message. Immediately after the 4 tone sequence a sub-audio maintenance tone is sent for the duration

of the call. At the end of the call the maintenance tone is removed and an optional 4 tone sequence sent

indicating the end of message (EOM). For further details on XTCSS see section 1.5.6.

By enabling XTCSS reception the host instructs the CMX882 to search for a valid 4 tone sequence, an

interrupt (if enabled) will be generated when this occurs. The 4 tone sequence will be indicated in the C-

BUS register ($C9) for the host to read out using the tone numbers in Table 6.

The sub-audio tone will be searched for after a valid 4 tone In-band sequence if CTCSS detection is also

enabled. CTCSS codes will be decoded and reported as defined in section 1.5.3.2. It is not necessary to

enable CTCSS in the Mode Control register for the device to search for the XTCSS sub audio tone.

In receive, whenever the XTCSS detect bit is set the CMX882 will search for a valid 4 tone In-band

sequence however detection of a CTCSS tone will inhibit the search for 4 tone sequences. To be valid

the 4 tones must be preceded and followed by silence in the audio band (signals below the audio detect

level - see program register P1.1) for the programmed no tone time. The presence (or absence) of the

sub-audio maintenance tone will only be indicated to the host if the CTCSS detect bit is also set. After the

4 tone sequence is detected the maintenance tone can be used by the host to detect fades and the end of

the message and hence can disable the audio path in sympathy with this tone being absent. At any time

the XTCSS enable bit is set and maintenance tone is not decoded the 4 tone set will be automatically

searched for.

FRS Signalling Processor CMX882

It is possible (although unlikely) that a fade will exactly coincide and obliterate 2 lots of 4 tone sequence

indicating an EOM and the start of a new message. In this case, the host could misinterpret the received

signal as a long fade and enable the voice when the maintenance tone reappears. It is therefore

recommended that the host operates a timer that is started on loss of maintenance tone. If this times out

the host can then assume that the fade is long enough that the original call is lost or has become so

corrupted that it is not worth continuing with. The host could then choose only to restore the audio path

on the next occurrence of a valid XTCSS tone set. Note that the XTCSS detector operates independently

and the host may enable or disable the audio path at any time.

1.5.4 Transmit Mode

The device operates in half duplex, so when the device is in transmit mode the receive path (discriminator

and audio output amplifiers) should be disabled, and can be powered down, by the host µC.

Two modulator outputs with independently programmable gains are provided to facilitate single or two-

point modulation, separate sub-audio and voice band outputs. If one of the modulator outputs is not used

it can be disabled to conserve power.

To avoid erroneous transmission of out of band frequencies when changing from Rx to Tx the MOD_1

and MOD_2 outputs are ramped to the quiescent modulator output level, VBIAS before switching.

Similarly, when starting a transmission, the transmitted signal strength is ramped up from the quiescent

VBIAS level and when ending a transmission the transmitted signal strength is ramped down to the

quiescent VBIAS level. The ramp rates are set in the Programming register P4.6. When the modulator

outputs are disabled, their outputs will be set to VBIAS. When the modulator output drivers are powered

down, their outputs will be floating (high impedance), so the RF modulator will need to be turned off.

Table 7 Concurrent Tx Modes Supported by the CMX882

Sub-Audio Voice band

CTCSS

CTCSS + Voice

CTCSS + In-band tone

CTCSS + FFSK/MSK

CTCCS^ + XTCSS

DCS

DCS + Voice

DCS + In-band tone

DCS + FFSK/MSK

Voice

In-band

FFSK/MSK

XTCSS

^ Special subaudio tone only

For all transmissions the host must only enable signals after the appropriate data and settings for those

signals are loaded into the C-BUS registers. As soon as any signalling is enabled the CMX882 will use

the settings to control the way information is transmitted.

A programmable gain stage in the microphone input path facilitates a host controlled VOGAD capability.

FRS Signalling Processor CMX882

1.5.4.1 Processing Voice Signals for Transmission over Analogue Channels

The microphone input(s), with programmable gain, can be selected as the voice input source. Pre-

emphasis is selectable with either version of the 2 analogue Tx audio filters (for 12.5kHz and 25kHz

channel spacing). These are designed for use in ETS-300-086 and/or TIA/EIA-603 compliant

applications. Both filters attenuate sub-audio frequencies below 250Hz by more than 33dB wrt the signal

level at 1kHz. These filters together with a built in limiter help ensure compliance with ETS-300-086

(25kHz and 12.5kHz channel spacing) when levels and gain settings are set up correctly in the target

system.

-60

-50

-40

-30

-20

-10

10 100 1000 10000 100000

Frequency (Hz)

Gain (dB)

r eference 0dB at 1kH z

+0.5dB/-2dB wrt ref.

-33dB

3kHz

250Hz

300Hz

-14dB/octave

Figure 8 25kHz Channel Audio Filter Response Template

The filter characteristics of the 12.5kHz channel filter fits the filter template shown in Figure 9 (solid

outline). This filter also facilitates implementation of systems compliant with TIA/EIA-603 ‘A’ and ‘B’

bands. To achieve attenuation above 3kHz of better than −100dB/decade for TIA/EIA-603 ‘C’ bands

(dashed outline), additional external circuitry is required, such as suggested in section 1.4.2.

-90

-80

-70

-60

-50

-40

-30

-20

-10

10 100 1000 10000 100000

Frequency ( Hz)

Gain (dB)

reference 0dB at 1kHz

+0.5dB/-2dB wrt ref.

-33dB

3kHz

250Hz

300Hz

- 60dB/decade wrt r ef.

2.55kHz

-100dB/d ec ade

(-50dB)

(-82.5dB)

20kH

Figure 9 12.5kHz Channel Audio Filter Response Template

FRS Signalling Processor CMX882

The CMX882 provides selectable pre-emphasis filtering of +6dB per octave from 300Hz to 3000Hz,

matching the template shown in Figure 10.

-20

-16

-12

-8

-4

100 1000 10000

Frequency (Hz)

+1dB

0dB

+12dB/octave

+6dB/octave

250

2500

-3dB

Figure 10 Audio Frequency Pre-emphasis Template

Modulator Output Routing

The sub-audio component can be combined with the voice band signal and this composite signal routed

to both MOD_1 and MOD_2 outputs, or the sub-audio and voice band signal can be output separately

(sub-audio to MOD_2 and voice band to MOD_1), in accordance with the settings of the Signal Routing

Tx Companding (Compressing)

The CMX882 incorporates an optional syllabic compandor in both transmit and receive mode. This

compresses voice band signals before transmission to enhance dynamic range. The compandor attack,

decay and 0dB point are defined in section 1.8.1. See section 1.6.9 for details of how to control this

function.

Audio Scrambling

The CMX882 incorporates an optional frequency inversion scrambler in transmit mode. This scrambles

voice band signals to be de-scrambled in the receiver. See section 1.6.9 for details of how to control this

function.

Voice Processing Combinations

Table 1 shows the valid voice processing combinations (see section 1.5)

1.5.4.2 CTCSS Tone

The sub-audio CTCSS tone generated is defined in the Tx CTCSS register ($C2). Table 3 lists the

CTCSS tones and the corresponding value for programming the TX TONE bits.

1.5.4.3 DCS Code

A 23 or 24-bit sub-audio DCS Code can be generated, as defined by the ‘DCS Code’ words (P2.4-5) of

the Programming register ($C8); the same DCS Code pattern is used for detection and transmission. The

DCS Code is NRZ encoded at 134.4±0.4 bits/s, low pass filtered and added to the voice band signal, prior

to passing the signal to the modulator output stages. Valid 23-bit DCS codes and the corresponding

settings for the DCS Code Register are shown in Table 5, this does not preclude other codes being

programmed. The least significant bit of the DCS code is transmitted first and the most significant bit is

FRS Signalling Processor CMX882

transmitted last. The CMX882 is able to encode and transmit either of the two DCS modulation modes

defined by TIA/EIA-603 and described in Table 4.

To signal the end of the DCS transmission, the host should set the special sub-audio bits in the Audio &

CTCSS Control register ($C2) to enable the DCS turn off tone for 150ms to 200ms. After this time period

has elapsed the host should then disable DCS in the Mode register ($C1). Do not enable CTCSS in the

Mode Control ($C1) register when transmitting the DCS turn off tone. To summarize, detection of DCS

turn off tone requires the CTCSS decoder to be enabled, whereas generation of the DCS turn off tone

requires the CTCSS encoder to be disabled.

1.5.4.4 Transmitting In-band Tones

The In-band tone to be generated is defined in the Tx Tone register ($C3). The tone level is set in the

Programming register (P1.4). The In-band tone must be transmitted without other signals in the voice

band, so the host µC must disable the voice path prior to initiating transmission of a In-band tone, and

restore the voice path after the In-band tone transmission is complete. Table 6 shows valid In-band

tones, together with the values for programming the In-band bits of the Tx Tone register.

Custom In-band tone frequencies are set in the program register ($C8) P1.6-9. See section 1.6.20.2 for

programming details.

1.5.4.5 Transmitting FFSK/MSK Signals

The FFSK/MSK encoding operates in accordance with the bit settings in the Modem Control register

($C7). When enabled the MSK modulator will immediately begin transmitting data using the settings and

values in the Tx Data and General Control register(s) and block 0 of the Programming Register.

Therefore, these registers should be programmed to the required value before transmission is enabled.

The CMX882 generates it’s own internal data clock and converts the binary data into the appropriately

phased frequencies, as shown in Figure 11 and Table 8. The binary data is taken from registers $CA and

$CB, most significant bit first. The following data words must be provided over the C-BUS within certain

time limits to ensure the selected baud rate is maintained. The time limits will be dependent on the data

coding being used, see Table 9.

SIGN

LI/P

LOGIC '1' LOGIC '0'

2400 B

LOGIC '1' LOGIC '0'

R x

S I G N

L I / P

1 2 0 0 B

U D

L O G I C ' 1 ' L O G I C ' 0 ' L O G I C ' 1 ' L O G I C ' 0 '

Figure 11 Modulating Waveforms for 1200 and 2400 Baud FFSK/MSK Signals

FRS Signalling Processor CMX882

The table below shows the combinations of frequencies and number of cycles to represent each bit of

data, for both baud rates.

Table 8 Data Frequencies for each Baud Rate

Baud Rate Data Frequency Number of Cycles

1200 baud 1 1200Hz one

0 1800Hz one and a half

2400 baud 1 1200Hz half

0 2400Hz one

Note: FFSK may be transmitted in conjunction with a CTCSS or DCS sub-audio component.

1.5.4.6 Transmitting XTCSS Signalling

XTCSS signals can be transmitted by loading the 4 tone pattern and the special CTCSS (XTCSS

maintenance) tone into the C-BUS registers and enabling XTCSS. The device will transmit the 4 tones in

sequence, raise an interrupt when this is complete and then automatically generate the CTCSS tone (if

enabled). At the end of the message the CTCSS tone is disabled by clearing the XTCSS maintenance

tone bit (b9) to ‘0’ in the Audio and Device Address Control register. The XTCSS 4 tone sequence must

be transmitted on its own, so if a voice or a data signal is being transmitted, this must be disabled during

the XTCSS 4 tone transmission. See section 1.5.6 for more information.

1.5.5 FFSK/MSK Data Packeting

The CMX882 has extensive data packeting features that can be controlled by the Modem Control register

($C7). The CMX882 can packet data in a variety of formats so the user can have the optimum data

throughput for various signal to noise ratios. Data is transferred in packets or frames, each frame is made

up of a Frame Head followed by any associated User Data. The Frame Head is composed of a 16 bit Bit

Sync and 16 bit Frame Sync pattern immediately followed by 4 bytes of data. The 4 bytes of data start

with an 8 bit address followed by 1 byte carrying information about the format of the following Data Block,

a further byte indicates the size of the packet and the last byte is a checksum to detect if any of the 4

Frame Head bytes has been corrupted.

1.5.5.1 Tx Hang bit

When transmitting FFSK/MSK data the user should ensure that the data is terminated with a hang bit. To

do this, the host must set the 'Last Data' bit in the Modem Control Register ($C7) after the last data word

has been loaded into the Tx Data register, as described in section 1.6.11. This will append a hang bit

onto the end of the current word and will stop modulating after the hang bit has been transmitted. It will

also generate (if enabled) an interrupt when the hang bit has left the modulator.

1.5.5.2 Frame format

←---------------------------------Frame Head---------------------------------→

←---Sync Field---→ ←------------------Control Field------------------→ ←--------Data Block--------→

Bit

Sync Frame

Sync Address

byte Format

byte Size^ or

User byte Check-

sum A Data

bytes

Check-

sum B*

16 bits 16 bits Byte 1 Byte 2 Byte 3 Byte 4

* Checksum B not applied to all Data Block types

^ Byte 3 is only reserved on sized data blocks.

The Data Block is made up from the User Data. This consists of a variable number of data bytes

optionally encoded to ensure secure delivery over a radio channel to the receiver. Checksum B is only

applied at the end of sized Data Blocks, the receiver can then detect if any of the User Data has been

corrupted. Checksum B is composed of 16 bits for messages ≤16 bytes and 32 bits for longer messages.

1.5.5.3 Frame Head

FRS Signalling Processor CMX882

The Frame Head forms an important part of a frame as it allows the receiver to detect and lock on to MSK

signals, provides basic addressing to screen out unwanted messages and indicates the format, coding

and length of any following data.

The 4 Control Field bytes have Forward Error Correction (FEC) applied to them in the transmitter, this

adds 4 bits to every byte and the receiver can correct errors in the received bytes. The 4 received bytes

are then checked for a correct CRC so corrupted Frame Heads can be rejected. If checksum A indicates

that the Control Field bytes are correct, the Address (byte 1) is compared with that stored in the Device

Address bits of register $C2. If a match occurs, or if the received address is '40' then an interrupt is

raised indicating a valid Frame Head has been received. The Frame Head is 80 bits long (16 + 16 +

{4x12}).

1.5.5.4 Data Block Coding

The Data Block follows the Frame Head and can be coded with different levels of error correction and

detection. The Data Block format is controlled by Frame Head byte 2, see also section 1.6.11. Messages

can take the following formats:

Type: Description:

0 Raw data, the CMX882 will transmit 16 bits at a time. In receive the device will search for the

programmed 16 bit frame sync pattern and then output all following data 16 bits at a time, the

host will have to perform all other data formatting. The data scrambler does not operate in this

mode. This mode can be useful when interfacing to a system using a different format to those

available in the CMX882.

1 Frame Head only, no User Data will be added. Th is format can be useful for indicating channel

or user status by using byte 3 and the User Bit of the Frame Head.

2 Frame head followed by raw data. User data is appended to the Frame Head in 2 byte units

with no formatting or CRC added by the CMX882. No size information is set in the Frame Head

and User Data may contain any even number of bytes per Frame.

3 Frame Head followed by FEC coded data only. Each byte of the User Data has 4 bits of FEC

coding added. No size information is set in the Frame Head and User Data may contain any

even number of bytes per Frame. No CRC is added to the data.

4 Frame Head followed by FEC coded data with an automatic CRC at the end of the User Data.

The number of User Data bytes in the Frame must be set in Frame Head byte 3. The CRC is

automatically checked in the receiver and the result indicated to the host. Up to 255 bytes of

User Data can be sent in each Frame using this format.

5 As per ‘4’ with the addition of all Data Block bytes being interleaved. This spreads the

transmitted information over time and helps reduce the effect of errors caused by fading.

Interleaving is performed on blocks of 4 bytes, the CMX882 automatically adds and strips out

pad bytes to ensure multiples of 4 bytes are sent over the radio channel.

Notes:

• Message types 1, 2 and 3 have no size information requirement and do not reserve Frame Head

byte 3. This byte may be freely used by the host to convey information. In message types 4 and

5 this byte must be set to the number of User Bytes in the message attached to that Frame Head

(≤255) to allow the receiver to correctly decode and calculate the CRC.

• Type 0 data transfers do not use frame heads. In Tx the host must transfer the bit and frame

sync data before sending the message data. In Rx the host must decode all data after the frame

sync.

• When using type 0 (raw data) messages it is strongly recommended that the default Frame Sync

pattern is not used. This is to reduce the loading on receivers using the formatted data types

which would otherwise try to decode the 6 bytes following a type 0 frame sync as a Frame Head.

Table comparing different message types:

FRS Signalling Processor CMX882

Air time for FFSK

message (ms)

at at

Data Block

Formatting

Type:

Total over-

air bits for

an 80 byte

message 1200b/s 2400b/s

Over air

efficienc

Burst length

protection at 1200b/s

[for 2400b/s divide

both times by 2]

Probability

of detecting

errors

2 720 600 300 89% None Zero

3 1040 867 433 62% <0.83ms in any 10ms Poor

4 1088 907 453 59% <0.83ms in any 10ms Excellent

5 1088 907 453 59% <3.33ms in any 40ms Excellent

Higher levels of error protection have the penalty of adding extra bits to the over air signal and this

reduces the effective bit rate. Less error protection increases the effective bit rate, however in typical

radio conditions the penalty is a greater risk of errors leading to repeated messages and a net reduction in

effective bit rate compared to using error correction and detection.

1.5.5.5 Data Scrambling / Privacy Coding

It is preferable for MSK over air data to be reasonably random in nature to ensure the receiver can track

timing using the bit changes and to smooth the frequency spectrum. To reduce the possibility of User

Data causing long strings of 1’s or 0’s to be transmitted, a 16 bit data scrambler is provided and operates

on all bits after the Frame Head.

The default (standard) setting for this scrambler is with a start code (seed) of $FFFF and any receivers

with the same seed may decode this data. However, if the transmitter and receiver pre-arrange a

different seed then the scrambler will start its sequence in another place and any simple receiver that

does not know the transmitted seed will not be able to successfully decode the data. This method gives

over 65,000 different starting points and the chance of others decoding data successfully is reduced.

The CMX882 provides the option of two custom 16 bit words that are programmable by the user in

Program Register P0.2 to P0.5. Bits 0 and 1 in the Frame Head Format byte indicate which setting

(standard, Seed1, Seed2 or none) the flowing Data Block has been scrambled with, see section 1.6.11.

Note that a seed of $0000 will effectively turn off the scrambler and provide no protection against long

sequences of 1’s or 0’s. Reception of scrambled data will only be successful when the receiving device

has been programmed with the correct (identical) seed to that used by the transmitter.

By using this method the CMX882 provides a privacy code that will protect against casual monitoring,

however the data is not encrypted and a sophisticated receiver can decode the data by using moderately

simple decoding techniques. If data encryption is required it must be performed by the host. The

scrambler function is controlled by bits 0,1 of the Modem Control register ($C7).

1.5.5.6 Data Buffer Timing

Data must be transferred at the rate appropriate to the signal type and data format. The CMX882 buffers

signal data in up to two 16-bit registers. The CMX882 will issue interrupts to indicate when data is

available or required. The host must respond to these interrupts within the maximum allowable latency for

the signal type. Table 9 shows the maximum latencies for transferring signal data to maintain appropriate

data throughput. Table 9 Maximum Data Transfer Latency

Max time to read from

or write to data buffer

Data encoding type 1200b/s 2400b/s

Data buffer

size

0 13.3ms 6.6ms 2 bytes

1 N/A* N/A* 4 bytes

2 13.3ms 6.6ms 2 bytes

3 20ms 10ms 2 bytes

4 40ms 20ms 4 bytes

5 40ms 20ms 4 bytes

* Type 1 message is an isolated Frame Head, there is no subsequent data to load (Tx) or read (Rx).

1.5.6 XTCSS Coding

FRS Signalling Processor CMX882

The CMX882 allows addressed calling using a 4 tone In-band tone burst followed by an optional sub-

audio ‘XTCSS maintenance tone’ (at 64.7Hz). In transmit the CMX882 handles the transmission of the 4

tone sequence and the sub-audio tone. In receive the CMX882 will search for valid In-band tone

sequences containing the previously programmed address.

The over air signalling of XTCSS is shown below:

XTCSS sub-audio

In-band tones A1 A0 S0 S1

Voice / Data

Notes:

• To reduce 'cut on' time with XTCSS voice calls, the host can enable the receive audio path at 'B'

(as soon as the 4 tone sequence is available), before the sub-audio is detected.

• XTCSS 4 tone sequences must be prefixed and suffixed with a silent 'no-tone' period of at least

the length of each tone. See also programming register P1.1.

In-band tones A1 and A0 are the BCD (binary coded decimal) representation of the Device Address bits

of $C2 register, the valid XTCSS address range is 01 to 99, A0 is the least significant digit. The XTCSS

address '40' is reserved for an all call address - regardless of the XTCSS address being searched for the

CMX882 will always indicate when a valid 4 tone set containing address '40' has been received.

In transmit mode, the CMX882 will automatically generate the sub-audio maintenance tone, if the XTCSS

maintenance tone is selected in the Audio and Device Address Control register. CTCSS must not be

enabled in the Mode Control register (CTCSS Enable bit, b11 = 0). The sub-audio maintenance tone (if

enabled) will automatically be output after the 4th XTCSS tone has been transmitted. An XTCSS interrupt

is generated (if enabled) at point 'A' - see diagram above. The host should then wait one tone period

before enabling the audio path (or transmitting data) to ensure sufficient no-tone suffix to the XTCSS 4

tone set.

To receive the XTCSS maintenance tone, CTCSS must be enabled (b11 = 1) in the Mode Control register

(no special sub-audio tone selected in the Audio and Device Address Control register). When the

maintenance tone is detected a CTCSS change is flagged in the Status register and the XTCSS sub-

audio tone is identified in the Tone Status register.

In-band tone S0 is selected from the normal tone range of $B - $D to maintain compatibility with HSC type

addressing. In-band tone S1 is selected from the normal tone range $0 - $9. The bit patterns for S0 and

S1 indicate the type of information to follow according to the following tables:

In-band tone S0

Dec Hex

0-10 0-$A Reserved, do not use for S0

11 $B Data or other silent (non-voice) call to follow, see S1a

12 $C Voice to follow - see S1b

13 $D Reserved, do not use for S0

FRS Signalling Processor CMX882

In-band tone S1a

Dec Hex

0-3 0-3 User option for S1a

4 4 MSK data to follow 1200b/s Formatted Blocks

5 5 MSK data to follow 1200b/s Raw Data

6 6 MSK data to follow 2400b/s Formatted Blocks

7 7 MSK data to follow 2400b/s Raw Data

8 8 End of XTCSS coded message (EOM)

9-13 9-$D Reserved - do not use for S1a

In-band tone S1b

Dec Hex

Voice message Compressed

0 0 No

1 1 Yes

2-13 2-$D Reserved, do not use for S1b

Note: Tone numbers in the above tables refer to the Normal tone column as defined in Table 6.

Examples:

Device

address Over air

4 tones Meaning

$22 34C0 Address 34, Un-compressed voice to follow

$03 03C1 Address 03, Compressed voice to follow

$2C 4EB0 Address 44, Non voice, user option 0, (E is repeat character)

Note: For all XTCSS coding the CMX882 will add (in Tx) and strip out (in Rx) the repeat tone as

required. The host µC need only load or read out the normal tones listed in Table 6.

FRS Signalling Processor CMX882

1.5.7 C-BUS Operation

This block provides for the transfer of data and control or status information between the CMX882’s

internal registers and the µC over the C-BUS serial interface. Each transaction consists of a single

µC to be written into one of the CMX882’s Write Only Registers, or one or more data byte(s) read out from

one of the CMX882’s Read Only Registers, as illustrated in Figure 12.

Data sent from the µC on the Command Data line is clocked into the CMX882 on the rising edge of the

Serial_Clock input. Reply Data sent from the CMX882 to the µC is valid when the Serial_Clock is high.

The CSN line must be held low during a data transfer and kept high between transfers. The C-BUS

interface is compatible with most common µC serial interfaces and may also be easily implemented with

general purpose µC I/O pins controlled by a simple software routine.

The number of data bytes following an A/C byte is dependent on the value of the A/C byte. The most

significant bit of the address or data are sent first. For detailed timings see section 1.8.1.

C-BUS Write:

See Note 1 See Note 2

CSN

Serial_Clock

CMD_DATA 7 6 5 4 3 2 1 0 7 6 … 0 7 … 0

MSB LSB MSB LSB MSB LSB

Address / Command byte Upper 8 bits Lower 8 bits

REPLY_DATA

High Z state

C-BUS Read:

See Note 2

CSN

Serial_Clock

CMD_DATA 7 6 5 4 3 2 1 0

MSB LSB

Address byte Upper 8 bits Lower 8 bits

REPLY_DATA

7 6 … 0 7 … 0

High Z state MSB LSB MSB LSB

Data value unimportant Repeated cycles Either logic level valid

Figure 12 C-BUS Transactions

Notes:

1. For Command byte transfers only the first 8 bits are transferred ($01 = Reset).

2. For single byte data transfers only the first 8 bits of the data are transferred.

3. The CMD_DATA and REPLY_DATA lines are never active at the same time. The Address byte

determines the data direction for each C-BUS transfer.

4. The Serial_Clock input can be high or low at the start and end of each C-BUS transaction.

5. The gaps shown between each byte on the CMD_DATA and REPLY_DATA lines in the above

diagram are optional, the host may insert gaps or concatenate the data as required.

FRS Signalling Processor CMX882

1.6 C-BUS Register Description

1.6.1 C-BUS Register Summary

C-BUS Write Only Registers

ADDR.

(hex) REGISTER Word Size

(bits)

$01 C-BUS RESET 0

$B0 ANALOGUE GAIN 16

$B1 SIGNAL ROUTING 16

$B2 AUXILIARY ADC THRESHOLDS 16

$B3 AUXILIARY ADC CONTROL 8

$C0 POWER DOWN CONTROL 16

$C1 MODE CONTROL 16

$C2 AUDIO & DEVICE ADDRESS CONTROL 16

$C3 TX TONE 16

$C7 MODEM CONTROL 16

$C8 PROGRAMMING REGISTER 16

$CA TX DATA 1 16

$CB XTCSS & TX DATA 2 16

$CD AUDIO TONE 16

$CE INTERRUPT MASK 16

$CF RESERVED REGISTER ADDRESS 16

The C-BUS address $CF is allocated for production testing and must not be accessed in normal

operation.

C-BUS Read Only Registers

ADDR

(hex) REGISTER Word Size

(bits)

$B4 AUXILIARY ADC DATA 8

$C5 RX DATA 1 16

$C6 STATUS 16

$C9 XTCSS & RX DATA 2 16

$CC TONE STATUS 16

Interrupt Operation

The CMX882 will issue an interrupt on the IRQN line when the IRQ bit (bit 15) of the Status register and

the IRQ Mask bit (bit 15) are both set to ‘1’. The IRQ bit is set when the state of the interrupt flag bits in

the Status register change from a '0' to '1' and the corresponding mask bit(s) in the Interrupt Mask register

is(are) set.

All interrupt flag bits in the Status register except the Programming Flag (bit 0) are cleared and the

interrupt request is cleared following the command/address phase of a C-BUS read of the flag register.

The Programming Flag bit is set to '1' only when it is permissible to write a new word to the Programming

FRS Signalling Processor CMX882

1.6.2 $01 C-BUS RESET: address only.

The reset command has no data attached to it. It sets the device registers into the states listed below.

Addr. REG. 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

$B0 ANALOGUE GAIN 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$B1 SIGNAL ROUTING 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$B2 AUXILIARY ADC THRESHOLDS 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0

$B3 AUXILIARY ADC CONTROL 0 0 0 0 0 0 0 0

$B4 AUXILIARY ADC DATA X X X X X X X X

$C0 POWER DOWN CONTROL 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$C1 MODE CONTROL 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$C2 AUDIO & DEVICE ADDRESS

CONTROL 00000000000 0 0 0 00

$C3 TX TONE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$C5 RX DATA 1 XXXXXXXXXX X X X X XX

$C6 STATUS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$C7 MODEM CONTROL 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$C8 PROGRAMMING REGISTER 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$C9 XTCSS & RX DATA 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$CA TX DATA 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$CB XTCSS & TX DATA 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$CC TONE STATUS 00000000000 0 0 0 00

$CD AUDIO TONE 00000000000 0 0 0 00

$CE INTERRUPT MASK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

$CF Reserved Register Address 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Following a general reset all of the programming registers (P0 – P4) are reset to zero except the

following:

P0.0 Frame Sync LSB - - - - 0 0 0 0 0 0 1 0 0 0 1 1

P0.1 Frame Sync MSB - - - - 0 0 0 0 1 1 0 0 1 0 1 1

P0.6 Bit Sync LSB - - - - 0 0 0 0 0 1 0 1 0 1 0 1

P0.7 Bit Sync MSB - - - - 0 0 0 0 0 1 0 1 0 1 0 1

P4.7 Transmit Limiter Control - - 1 1 1 1 1 1 1 1 1 1 1 1 1 1

This initiates the device with the MSK frame sync pattern of $CB23 and bit sync of alternate 1’s and 0’s.

The transmit limiter value is initialised to the maximum limit.

To initialise the device following power-up, or to clear the current device state, apply the following

sequence of C-Bus actions:

1. Send a C-Bus Reset command.

2. Send $2001 to the Mode Control register (C-Bus address $C1).

3. Send $0000 to the Mode Control register.

The device is now ready to be configured for its next application.

FRS Signalling Processor CMX882

1.6.3 $B0 ANALOGUE GAIN: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

Inv_1 MOD_1

Attenuation Inv_2 MOD_2

Attenuation 0 Input

Gain Audio Output

Attenuation

Bits 15 and 11 set the phase of the MOD_1 and MOD_2 outputs. When set to '0' the 'true' signal (0°

phase shift) will be produced, when set to '1' the signal will be inverted (180° phase shift). This can be

useful when interfacing with rf circuitry or when generating an inverted turn off coe for CTCSS. Any

change will take place immediately after these bits are changed.

The output paths provide user programmable attenuation stages to independently adjust the output levels

of the modulators. Finer level control of the MOD_1 and MOD_2 outputs can be achieved with the FINE

OUTPUT GAIN 1 and FINE OUTPUT GAIN 2 registers (P4.2-3).

Bit 14 Bit 13 Bit 12 MOD_1 Output

Attenuation Bit 10 Bit 9 Bit 8 MOD_2 Output

Attenuation

0 0 0 >40dB 0 0 0 >40dB

0 0 1 12dB 0 0 1 12dB

0 1 0 10dB 0 1 0 10dB

0 1 1 8dB 0 1 1 8dB

1 0 0 6dB 1 0 0 6dB

1 0 1 4dB 1 0 1 4dB

1 1 0 2dB 1 1 0 2dB

1 1 1 0dB 1 1 1 0dB

Bit 7 is reserved - set to 0.

Bits 6 to 4 control the input path programmable gain stage - useful when amplifying low power voice

signals from the microphone inputs. Finer gain control can be achieved with the ‘FINE INPUT GAIN’

control register (P4.0). In receive mode it is recommended to set the gain to 0dB.

Bit 6 Bit 5 Bit 4 Input Gain Bit 3Bit 2Bit 1Bit 0 Audio Output

Attenuation

0 0 0 0dB 0 0 0 0 >60dB

0 0 1 3.2dB 0 0 0 1 44.8dB

0 1 0 6.4dB 0 0 1 0 41.6dB

0 1 1 9.6dB 0 0 1 1 38.4dB

1 0 0 12.8dB 0 1 0 0 35.2dB

1 0 1 16.0dB 0 1 0 1 32.0dB

1 1 0 19.2dB 0 1 1 0 28.8dB

1 1 1 22.4dB 0 1 1 1 25.6dB

1 0 0 0 22.4dB

1 0 0 1 19.2dB

1 0 1 0 16.0dB

1 0 1 1 12.8dB

1 1 0 0 9.6dB

1 1 0 1 6.4dB

1 1 1 0 3.2dB

1 1 1 1 0dB

Bits 3 to 0 control the output path programmable attenuation stage to adjust the volume of the audio

output signal. Finer volume control can be achieved with the ‘FINE OUTPUT GAIN 1’ control register

(P4.2).

FRS Signalling Processor CMX882

1.6.4 $B1 SIGNAL ROUTING: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

0 0 Tx MOD_1 and

MOD_2 Routing 0 0 0 0 0 0

Analogue

i/p select AUDIO

o/p select Ramp

Up Ramp

Down

Bits 15 and 14 reserved - set to 0.

Bits 13 and 12 select the routing of the transmit signals allowing 1 or 2 point modulation and interfaces.

Bit 13 Bit 12 Tx MOD_1 and MOD_2 routing

0 0 Tx, MOD_1 and MOD_2 outputs set to bias.

0 1 Tx, In band signals to MOD_1, Subaudio signals to MOD_2

1 0 Tx, In band and Subaudio to MOD_1, Subaudio signals to

MOD_2

1 1 Tx, In band and Subaudio to both MOD_1 and MOD_2

To route In-band and Subaudio to MOD_1 and Vbias to MOD_2, select b13 = 1 (b12 = 0 or = 1) and set

MOD_2 attenuation to >40dB in the Analogue Gain register.

‘In-band’ in this context refers to any of the signals; Voice, In-band tone etc.

Bits 11 to 6 are reserved - set to 0.

Bit 5 Bit 4 Analogue Input select

0 0 No input selected (Input = VBIAS)

0 1 Input amplifier 2 (Input_2 i/p)

1 0 Microphone (MIC i/p)

1 1 Discriminator (DISC i/p)

Bit 3 Bit 2 AUDIO Output select

0 0 No output selected (Output = VBIAS)

0 1 Received Voice signal

1 0 MOD_1 signal (for Tx monitoring)

1 1 Reserved, do not use

When bits 1 or 0 are set to '1' output signals are ramped up (bit 1) or ramped down (bit 0) to reduce

transients in the transmitted signal. Time to ramp up / down is set in the 'Ramp Rate Control' section of

the Programming register (P4.6).

1.6.5 $B2 AUXILIARY ADC THRESHOLDS: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

High Threshold [Range: 0 to 255] Low Threshold [Range: 0 to 255]

If the selected signal level exceeds the High Threshold, the ‘Signal High’ bit of the Status register will be

set to 1. If the Signal level falls below the Low Threshold, the ‘Signal Low’ bit of the Status register will be

set to 1. If the corresponding interrupt bit is enabled, a C-BUS interrupt will be generated. These status

bits are cleared when the Status register is read. The behaviour of the CMX882 is not defined if the high

threshold is less than the low threshold.

Threshold resolution: VDD(A)/256 per LSB

Threshold accuracy: ±2 LSB

Differential linearity: ±1 LSB [monotonic]

The ‘Auxiliary ADC Thresholds’ register must not be updated whilst the Auxiliary ADC is enabled.

FRS Signalling Processor CMX882

1.6.6 $B3 AUXILIARY ADC CONTROL: 8-bit write-only

Bit: 7 6 5 4 3 2 1 0

Aux ADC i/p select Conversion Interval

The ‘Conversion Interval’ (bits 5 to 0) defines the time between measurements whilst the Auxiliary ADC is

enabled. This allows the user to trade-off device power consumption with response time.

Auxiliary ADC power = 0.5mW/VDD(A)/conversion (approximate)

Conversion Interval = 20.8µs per LSB. (approximate)

The user should set an interval to ensure that no part of a received signal is missed, so that the signal

type can be correctly identified. If using the Rx Auto start-up feature the recommended maximum

Conversion Interval is 125µs. The ‘Auxiliary ADC’ register must not be updated whilst the Aux ADC is

enabled.

The Aux ADC i/p select (bits 7 to 6) control the input to the Auxiliary ADC. Control is independent of the

Analogue i/p select bits and hence the Aux ADC can monitor any one of the 4 inputs independently.

Bit 7 Bit 6 Auxiliary ADC input from:

0 0 Signal monitor (Sig_Monitor i/p)

0 1 Input amplifier 2 (Input_2 i/p)

1 0 Microphone (MIC i/p)

1 1 Discriminator (DISC i/p)

FRS Signalling Processor CMX882

1.6.7 $C0 POWER DOWN CONTROL: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8

Input_2

amp MIC

amp Disc

amp Input

Gain Output Fine

Gain 1 Output Fine

Gain 2 O/P Coarse

Gain 1 O/P Coarse

Gain 2

Bit: 7 6 5 4 3 2 1 0

Audio

Output BIAS Signal

Processing Prog Reg

Save Xtal_N Clock_Out_N Enable

Aux ADC Rx Auto

start-up

Bits 15 to 5 provide the power control of the specified blocks. If a bit is '1', the corresponding block is on,

else it is powered down. A C-BUS or Power up reset clears all bits in this register to '0'.

If bit 5 is '0' the internal signal processing blocks are reset and placed into a power-save mode.

If bit 4 is clear to ‘0’, the program registers will be reset to the Power-up and C-BUS Reset default state

whenever the Signal Processing block comes out of power save (b5 0 → 1). To preserve the current

settings of the Programming register values bit 4 should be set to a '1'. Setting bit 4 to ‘1’ prevents the

Programming register values being reset when the Signal Processing block comes out of power save,

such as during Rx Auto start-up. This facility should only be used if all the Programming register values

have been initialised or programmed by the host µC prior to the signal processing block being put into

power save.

Bits 3 and 2 control the xtal clock circuit. The xtal circuit is powered down by setting bit 3 to '1'. Note: The

Clock/Xtal pin may be driven by an external clock source regardless of the setting of these bits. The

Clock_Out pin is disabled (held low) by setting bit 2 to ‘1’. After a Power-up or C-BUS reset bits 2 and 3

are cleared to ‘0’, so that both the xtal circuit and clock output are enabled.

Bit 1 controls the Auxiliary ADC. If set to '1' the Auxiliary ADC will generate interrupts in accordance with

the settings of the interrupt mask bits. If bit 1 is '0' the Auxiliary ADC is disabled and powered down.

Bit 0 controls Rx Auto start up. If bit 0 is set to '1' and the Aux ADC input rises above the ‘High Threshold’

the device will automatically enter receive mode and initiate Rx signal type identification for those signals

enabled in the Mode register. The correct Aux ADC input, Rx signal routing and power down bits must be

set for automatic receive start up to operate, the mode control bits (b1, b0) should be set to '01' (Receive)

in this case. If bit 0 is cleared to '0' the CMX882 will not automatically start-up and it is up to the host to

respond to Aux ADC interrupts in this case. Bit 0 must be set to '0' whilst writing through register $C8 -

Programming Register.

FRS Signalling Processor CMX882

1.6.8 $C1 MODE CONTROL: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8

Enable

Voice In band signalling:

In-band tone, XTCSS Generate

Audio Tone Enable

CTCSS Enable DCS Enable DCS

Inverse 0

Bit: 7 6 5 4 3 2 1 0

0 0 0

Enable

2400b/s Enable

1200b/s 0 Mode Select

Bits 1 and 0 control the overall mode of the CMX882 according to the table below:

Bit 1 Bit 0 Device Mode

0 0 Idle

0 1 Receive Mode

1 0 Transmit Mode

1 1 Reserved - do not use

During transmit, only one signal type may be enabled for each of the sub-audio and voice bands, see

Table 7. During receive the CMX882 will search for all signals enabled in this register and report those

that are successfully decoded. See also Table 2 in section 1.5.3.

In transmit mode the CMX882 begins transmission of a selected signal immediately after it has been

enabled. The host µC must ensure all associated data and control bits have been set to their required

values before enabling the signal in this register.

Bits 4 and 3 control the modem functions of the CMX882 in accordance with the following table:

Bit 4 Bit 3 Tx - Transmitted signal Rx - Monitored signal(s)

0 0 None None

0 1 MSK 1k2b/s MSK 1k2b/s

1 0 MSK 2k4b/s MSK 2k4b/s

1 1 Reserved MSK 1k2 & 2k4b/s

In transmit mode data transmission will start or finish (regardless of whether all data has been

transmitted) immediately after the modem control bits are changed. To transmit a second data message

the modem control bits must be set to '0', data bytes for the following message loaded, and the required

bits set to '1'.

Bits 2 and 5 to 8 are reserved - set to '0'.

Bits 11 to 9 determine the sub-audio transmission / reception signalling:

Bit 11 Bit 10 Bit 9 Tx - Transmitted signal: Rx - Monitored signal(s):

0 0 0 No Sub-Audio Transmitted No Sub-audio Monitoring

0 0 1 Inverted DCS* Inverted DCS*

0 1 0 DCS DCS

0 1 1 Do not use DCS + inv DCS*

1 0 0 CTCSS CTCSS

1 0 1 Do not use CTCSS + inv DCS*

1 1 0 Do not use CTCSS + DCS

1 1 1 Do not use CTCSS + DCS + inv DCS*

* See Table 4 DCS Modulation Modes.

Bit 12 enables Audio tone generation (see section 1.6.14). This operates in transmit and receive modes.

In transmit mode this bit will only enable the Audio Generator when no other voice band signals are being

transmitted i.e. bits 14, 13, 4 and 3 set to '0'.

FRS Signalling Processor CMX882

Bits 14 and 13 select the type of In band tone to transmit or receive. When transmitting In band signals

the voice path must be disabled by clearing ‘Enable Voice’ bit 15 to '0'.

Bit 14 Bit 13 Tx - Transmitted signal Rx - Monitored signal

0 0 No voice band tone transmitted No voice band tones monitored

0 1 In-band tone In-band tone

1 0 Reserved Reserved

1 1 XTCSS XTCSS

When set to '1', bit 15 enables the voice path. In transmit mode the selected audio input is routed to the

modulator outputs. In receive mode the voice processing path is enabled to the audio output. In transmit

mode bit 15, if set to '1', will be temporarily disabled (cleared to '0') whenever any of the bits 3, 4, 12, 13

and 14 are set to '1'. In receive mode bit 15, if set to '1', will be temporarily disabled (cleared to '0')

whenever bit 12 is set to '1'. It is up to the host µC to control bit 15 when voice band signals are received.

When in receive mode, if Voice is enabled and In-band tone detection is also selected, the voice

processing quality (SINAD) is reduced.

The Mode Control register ($C1) may be written to at any time (subject to C-BUS timing restrictions). If

the enable bit of the currently decoded signal is disabled whilst in phase 2 the CMX882 will return to

phase 1 for that band. If the same signal needs to be searched for again then the appropriate bit needs to

be set back to ‘1’ in $C1. However, to de-emphasise in-band tones, bit 15 must be set to ‘1’.

The CMX882 will only detect signals when their amplitude is above the threshold set for each band (sub-

audio and voice), as set in the program registers. Therefore even if valid tones or signals are present the

CMX882 will ignore them unless they exceed the detect threshold. Time and level hysteresis is applied to

reduce chattering in marginal conditions.

Detection strategies used by the CMX882 whilst in receive mode:

When in receive mode the CMX882 treats the received signal in two bands; Sub-audio (60-260Hz) and

voice band (300-3kHz). For the sub-audio the CMX882 can monitor and decode CTCSS and DCS

signals in parallel. Because certain FFSK bit patterns can mimic some In-band tones the In-band receiver

is temporarily disabled when an FFSK frame sync is detected. If using sized packets (type 1, 4 and 5) the

CMX882 will automatically restore In-band tone detection when the received message has ended. If

using unsized packets (type 0, 2 and 3) the host must monitor the received data and restore XTCSS / In-

band tones (by setting bits 14 and 13 as required) when it has detected the end of data.

Configuring the CMX882 for Automatic Receive Start-Up

Prior to setting the CMX882 into Automatic Receive Start-Up mode, the Mode Control register should be

clear (Tx and Rx modes disabled). Set up the Programming Register blocks as required. Write to the

Powerdown Control register with the appropriate functions enabled or disabled, including Signal

Processing bit clear (b5 = 0), Prog Reg Save bit set (b4 = 1) and the Rx Auto Start-Up bit set (b0 = 1).

While the Signal Processing block is in its powersave condition, the Mode Control register should be set

up for the appropriate signalling schemes to be processed and Rx mode selected (b1 = 0, b0 = 1). When

auto start-up is triggered the signal processing bit in the Powerdown Control register will automatically be

set (= 1) and the selected receiver processing will start.

FRS Signalling Processor CMX882

1.6.9 $C2 AUDIO & DEVICE ADDRESS CONTROL: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

0 Scramble Compand Voice filter mode Special

Sub-Audio Device Address

Bits 7 to 0 define the device address. This setting is used for the CTCSS and XTCSS address in both Tx

and Rx modes and as the receive MSK frame head address. The range of valid addresses is: CTCSS

tone (1-51 in decimal), MSK address (1-255 in decimal) and XTCSS (1-99 in decimal). In Tx this number

will be used to select the addressing of the enabled signal, if the address is outside the valid range no

signalling will occur. In Rx this address (along with the all-call address of '40') will be searched for each

signalling format enabled in the Mode register. The detected signal type will be reported in the Status

Status register $CC.

Bits 9 to 8 select special sub-audio tones in accordance with the following table. Selecting the ‘DCS turn

off tone’ during DCS transmit will cause the DCS turn off tone to be transmitted; this will override the DCS

data being transmitted. ‘DCS turn off tone’ must be selected in this register to enable detection of the DCS

turn off tone during receive. To transmit the 64.7Hz XTCSS maintenance tone, XTCSS transmit must be

selected in the Mode Control register and XTCSS maintenance tone must be selected in this register.

Transmission of the maintenance tone overrides any other CTCSS tone being transmitted. The XTCSS

maintenance tone decoder is enabled by selecting XTCSS receive mode, so it is not necessary to select

the XTCSS maintenance tone in this register when receiving. If the Tone CloneTM mode is selected this

allows the device in Rx to non-predictively detect any CTCSS frequency in the range of valid tones, the

received tone number will be reported in the Tone Status register $CC.

Bit 9 Bit 8 Freq (Hz) Special Sub-Audio tone

0 0 - None

0 1 134.4 DCS turn off tone

1 0 64.7 XTCSS maintenance tone (Tx Only)

1 1 Clone CTCSS Tone clone mode (Rx only)

The voice filter control bits 12 and 11 determine the Voice Band Filter mode applied to the voice signal

before it is transmitted or after it has been received. Bit 10 controls the de-emphasis (Rx) or pre-

emphasis (Tx) mode of the voice band filtering.

Bit 12 Bit 11 Bit 10 Voice filter mode

X X 0 Disable de/pre-emphasis

X X 1 Enable de/pre-emphasis

0 0 X No filtering applied

0 1 X In transmit mode:

HPF (to remove SA) + 12.5kHz channel filtering

In receive mode:

HPF (to remove SA) + Low pass filter

1 0 X In transmit mode:

HPF (to remove SA) + 25.0kHz channel filtering

In receive mode:

HPF (to remove SA) + Low pass filter

1 1 X Reserved – do not use

Bit 13 controls the audio band compandor. When set to '1' audio band signals are compressed in transmit

mode and expanded in receive mode. When set to '0' no companding is performed.

Bit 14 controls the audio band scrambler. When set to ‘1’ voice signals are scrambled, by frequency

inversion, in transmit and receive modes. When set to ‘0’ no scrambling is performed.

Bit 15 is reserved, set to '0'.

FRS Signalling Processor CMX882

1.6.10 $C3 Tx In-Band Tones: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

Tx In-band tone 0 0 0 0 0 0 0 0 0 0 0

Bits 15 to 11 define the tone transmitted when Tx In-band tone is enabled. The frequency is as defined in

Table 6 In-band Tones.

Bits 10 to 1 are reserved, set to '0'.

1.6.11 $C7 Modem Control: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

0 0 0 0 0 Type 0

format Last Tx

data 0 MSK message

format 0 0

User

bit Scramble

seed

This register configures the way the CMX882 handles MSK data in both transmit and receive.

Bits 15 to 11, 8, 4 and 3 are reserved, set to '0'.

Bit 10 sets the raw data mode when set to '1'. When set to '0' the built in data packeting is enabled. The

way this bit is set determines how the CMX882 handles new data when transmit and receive modes:

Bit 10, Receive mode:

• If Bit 10 is ‘1’ the device will look for the programmed Frame Sync. pattern, raise an interrupt (if

enabled) and decode the following data 16 bits at a time, making them available in register $C5.

• If Bit 10 is ‘0’ the device will look for a complete Frame Head before raising an interrupt (if enabled)

and then decode the following data in accordance with the received message format. The Frame

Head Control Field bytes, User Data and any CRC will be presented in registers $C5 and $C9.

Bit 10, Transmit mode:

• If Bit 10 is ‘1’ the device will transmit data 16 bits at a time from register $CA. Bit and frame sync

pattern generation and all formatting of the data have to be performed by the host in this case.

• If Bit 10 is ‘0’ the device will transmit the programmed bit and frame sync patterns followed by a

Frame Head containing the information supplied in bits 7–0 of this register and the 2 bytes in

transmission of the data packet as defined in the Frame Head bytes.

Bit 9 is only valid when transmitting data with type 0, 2 or 3 formatting and indicates to the CMX882 that it

can cease modulation. The host must set this bit to '1' immediately after the interrupt for ‘load more data’

occurs. In receive, or when transmitting other message formats, this bit must be set to '0'.

When bit 10 = ‘0’ bits 7-0 control the format and coding used for MSK messages: (See also section

1.5.5.)

Bits 7 to 5 control the data formatting used by the CMX882 in transmit and receive:

Bit 7 Bit 6 Bit 5 Type Message format:

0 0 1 1 Frame Head only, no payload

0 1 0 2 Frame Head + Unsized payload of raw 16 bit words

0 1 1 3 Frame Head + Unsized payload with FEC

1 0 0 4 Frame Head + Sized payload with FEC + CRC

1 0 1 5 Frame Head + Sized payload with FEC + CRC + interleaving

All other patterns - Reserved

FRS Signalling Processor CMX882

Bit 2 is a User bit and may be freely used by the host. This bit has no effect on the message format or

encoding and will be reported in the Rx Data register for the receiving host to use as appropriate. This bit

could be used to indicate a special message, e.g. one containing handset or channel set-up information.

Bits 1-0 select the data scrambler seed used when transmitting User Data. The receiving CMX882 will

automatically de-scramble the received User Data using the setting indicated in the received Frame Head.

The receiving host can read the scrambler setting (0-3) used by the transmitter via register $C9 and may

use this when returning messages. See also section 1.5.5.5.

Bit 1 Bit 0 MSK data scrambling setting:

0 0 Standard scrambling (seed = $FFFF)

0 1 Scramble Seed1 (see P0.2-3)

1 0 Scramble Seed2 (see P0.4-5)

1 1 No scrambling (seed = $0000)

1.6.12 $C8 PROGRAMMING REGISTER: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

First

Word Block

Number Block /

Data Programming Data

See section 1.6.20 for a description of this register.

FRS Signalling Processor CMX882

1.6.13 $CA and $CB TX DATA & XTCSS Codes: 2 x 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

$CA Tx MSK Data Byte 0

(Frame Head: Address byte) Tx MSK Data Byte 1

(Frame Head: Size / Information byte)

$CB* Tx Data Byte 2* Tx Data Byte 3*

XTCSS Tone 3 (S1) XTCSS Tone 2 (S0) 0 0 0 0 0 0 0 0

*Register $CB is only used for Type 4 and 5 data formats and Frame Heads, the Tx buffer is effectively 4 bytes long in these cases.

These 2 words hold next 2 bytes (Byte 0 and 1) or 4 bytes (Bytes 0, 1, 2 and 3) of MSK data to be

transmitted. Outgoing data is continuous, if new data is not provided before the current data has been

transmitted the current data will be re-transmitted, until new data is provided. Transmission of current

data will be completed before transmission of newly loaded data begins. See section 1.5.5.6.

$CB holds the codes to be used when transmitting an XTCSS type tone set. Each 4 bits define the In-

band tone used, see Table 6 In-band Tones. S0 and S1 are the information section of the 4 tone set.

This register must be set to the required value before XTCSS transmission is enabled. For more details

see section 1.5.6. Note: the XTCSS address used is defined in the Audio & Device Address Control

Although $CB holds both Tx data and Tx XTCSS tone information, these functions can not be used

simultaneously so no conflict will occur.

When transmitting formatted MSK data packets the host must first load the correct Address and Size /

Information bytes for the following packet into register $CA. The CMX882 will automatically add the

Control byte (based on the settings in register $C7) and calculate the Frame Head Checksum A byte.

The CMX882 will read the Size and Message formatting information and automatically format all following

data; adding error correction bytes, adding pad bytes, interleaving, scrambling and calculating and

appending Checksum B as required. The only task the host need perform during the transmission of a

Frame is to download new data when it is required.

Note: These 2 words must be written separately. i.e. Two 16 bit C-BUS transactions.

1.6.14 $CD AUDIO TONE: 16-bit write only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

0 0 0 0 Audio Tone

When the required bits of the Mode Control register ($C1) are set an audio tone will be generated with the

frequency set by bits (11-0) of this register in accordance with the formula below. If bits 11-0 are

programmed with '0' no tone (i.e. Vbias) will be generated when the Audio Tone is enabled.

frequency = Audio Tone (i.e. 1Hz per LSB)

The Audio Tone frequency must only be set to generate frequencies from 300Hz to 3000Hz.

The host must suppress other voice band signalling and set the correct audio routing before generating

an audio tone and re-enable signalling and audio routing on completion of the audio tone. The timing of

intervals between these actions is also controlled by the host µC.

This register may be written to whilst the audio tone is being generated, any change in frequency will take

place after the end of the C-BUS write to this register. This allows complex sequences (e.g. ring or alert

tunes) to be generated for the local speaker (Tx or Rx via the AUDIO pin) or transmitted signal (Tx via the

MOD1/2 pins).

FRS Signalling Processor CMX882

1.6.15 $CE INTERRUPT MASK: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8

IRQ

MASK 0 Rx In-band

detect MASK XTCSS

MASK Rx CTCSS

detect MASK Rx DCS

detect MASK Aux ADC

High MASK Aux ADC Low

MASK

Bit: 7 6 5 4 3 2 1 0

FFSK end

MASK Data transfer

MASK 0 Rx 2400b/s

detect MASK Rx 1200b/s

detect MASK 0 0

Prog Flag

MASK

Bit Value Function

15 1 Enable selected interrupts

0 Disable all interrupts (IRQN pin not activated)

14 Reserved – Set to 0

13 1 Enable interrupt when a change to a In-band tone is detected as indicated

by a '0' to '1' change of bit 13 of the Status register

0 Disabled

12 1 Enable interrupt when a valid XTCSS 4 tone set is detected or has

finished being transmitted as indicated by a '0' to '1' change on bit 12 of

the Status register

0 Disabled

11 1 Enable interrupt when a change to a programmed CTCSS tone is

detected as indicated by a '0' to '1' change of bit 11 of the Status register

0 Disabled

10 1 Enable interrupt on a change in the detect status of the DCS decoder as

indicated by a '0' to '1' change of bit 10 of the Status register

0 Disabled

9, 8 1 Enable interrupt when the corresponding Aux ADC status bit changes

0 Disabled

7 1 Enable interrupt when FFSK data transmission has ended

0 Disabled

6 1 Enable interrupt when an FFSK data transfer is required

0 Disabled

5 Reserved - Set to 0

4 1 Enable interrupt when valid 2400b/s data is detected

0 Disabled

3 1 Enable interrupt when valid 1200b/s data is detected

0 Disabled

2, 1 Reserved - Set to 0

0 1 Enable interrupt when Prog Flag bit of the Status register changes from '0'

to '1' (see Programming register $C8)

0 Disabled

FRS Signalling Processor CMX882

The following 5 registers (including 2 receive data registers) are read only

1.6.16 $B4 AUX ADC MONITOR DATA: 8-bit read-only

Bit: 7 6 5 4 3 2 1 0

Signal Monitor Data

This data holds the result of the last measurement performed by the auxiliary ADC. The signal processor

must be on to read Aux ADC data, so Power Down Control register b5 must be set to ‘1’. This is

independent of whether Tx or Rx modes are selected.

1.6.17 $C6 STATUS: 16-bit read-only

Bit: 15 14 13 12 11 10 9 8

IRQ 0 In-band

tone state

change

XTCSS 4

tone set

complete

CTCSS

state

change

DCS

state

change

Aux ADC

Monitor High Aux ADC

Monitor Low

Bit: 7 6 5 4 3 2 1 0

End of

FFSK data

transfer

required

Block

CRC error Rx 2400b/s Rx 1200b/s 0 0 Programming

Flag

This word holds the current status of the CMX882: the value read out is only valid when bit 5 of the Power

Down Control register ($C0) is set to '1'. Changes in the Status register will cause the IRQ bit (bit 15) to

be set to '1' if the corresponding interrupt mask bit is enabled. An interrupt request is issued on the IRQN

pin when the IRQ bit is '1' and the IRQ MASK bit (bit 15 of register $CE) is set to '1'.

Bits 1 to 15 of the Status register are cleared to '0' after the Status register is read. Bit 0 is only cleared by

writing to the Programming Register.

Bits 14 and 2 to 1 are reserved.

Bits 13, 11 and 10 indicate that a In-band tone, CTCSS or DCS event caused the interrupt, the host

should then read the Tones Status register ($CC) for further information. In transmit these bits will be set

to '0'. Detection of the DCS turn off tone and removal of the DCS turn off tone are both flagged as DCS

events in the Status register, not as CTCSS events. The assertion or removal of the ‘XTCSS Maintenance

Tone’ (64.7Hz) is flagged as a CTCSS event.

In receive bit 12 indicates that a valid XTCSS 4 tone set with the correct addressing (see $C2) has been

detected, the 4 received tones are indicated in $C9. In Tx mode bit 12 will be set to '1' at the end of the

4th XTCSS tone transmitted.

Aux ADC High (bit 9) and Aux ADC Low (bit 8) reflect the recent history of the Aux ADC level, with respect

to the high and low thresholds. The most recent Aux ADC reading can be read from $B4.

Aux ADC

Monitor High Aux ADC

Monitor Low Aux ADC history since last reading:

0 0 Neither threshold crossed

0 1 Signal gone below low threshold

1 0 Signal gone above high threshold

1 1 Signal gone below low threshold and above high

threshold

In Rx mode bit 7 will be set at the same time as bit 6 after receiving the last part of a sized FFSK Frame or

Type 1 (frame head only) message, bit 5 (CRC) will also be updated at this time. When the host detects

bit 7 is set it may power down the CMX882 or set the CMX882 to transmit or receive new information as

appropriate.

In Tx mode bit 7 will be set when the last bit of FFSK data has been transmitted. Note; when using type

0, 2 or 3 data formats (see section 1.5.5.4) this bit will only be set if bit 9 of the Modem Control register

FRS Signalling Processor CMX882

($C7) is set at the appropriate time. After allowing a short time delay associated with the external

components and radio circuitry, the host may power down the CMX882 and transmitter or set the

CMX882 to transmit or receive new information as appropriate.

Bit 6 indicates that new transmit data is required (in Tx mode) or received data is ready to be read (in Rx

mode). For continuous transmission or reception of information, a data transfer should be completed

within the time appropriate for that data (see Table 9).

Bit 5 will be set after receiving the CRC portion of a sized Data Block (types 4 and 5) and will be set to ‘0’

when the locally calculated CRC indicates the received data has no errors.

Bits 4 and 3 indicate the received data rate after a valid data sequence has been received. If data Type 0

formatting is enabled these bits will be set on detection of a valid frame sync pattern. If Type 0 formatting

is disabled then these bits will only be set when a Frame Head is detected with a correct CRC.

Rx 2400b/s Rx1200b/s Bit rate of detected valid data:

0 0 No data

0 1 1200b/s

1 0 2400b/s

1 1 Reserved

Programming Flag, bit 0: The Programming Register ($C8) should only be written to when bit 0 is set to

'1' (with both Mode select bits set low – See register $C8). Writing to the Programming Register ($C8)

clears bit 0 to '0'. Bit 0 is restored to '1' when the programming action is complete, normally within 250 µs,

when it is then safe to write to the Programming Register.

1.6.18 $C5 and $C9 RX DATA: 2 x 16-bit read-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

$C5 Rx Data Byte 0

(Frame Head: Address byte) Rx Data Byte 1

(Frame Head: Format byte)

$C9 Rx Data Byte 2*

(Frame Head: Size / Information byte) Rx Data Byte 3*

(Frame Head: CRC byte)

XTCSS Tone 3 (S1) XTCSS Tone 2 (S0) XTCSS received address

*$C9 is used when receiving Type 4 and 5 formats and Frame Heads, the Rx buffer is effectively 4 bytes long in these cases.

These 2 words hold the most recent 2 bytes (Byte 0 and 1) or 4 bytes (Bytes 0, 1, 2 and 3) of MSK data

decoded. Received data is continuous, if the data is not read before the next data is received the current

data will be over-written.

$C9 holds the information decoded after receiving an XTCSS type tone set. Bits 7 to 0 represent the

received address in hex based on the XTCSS tones A1 and A0. This register will only be updated if the

received address matches the one programmed in the Audio and Device Address Control register or is

the all call address of '40'. Bits 15 to 12 and 11 to 8 defines the received S1 and S0 tones, see Table 6

and section 1.5.6.

Although $C9 holds both Rx data and Rx XTCSS tone information, only one type of signal will be present

in the received signal at any one time so no conflict will occur.

After receiving a Frame Head the host can read the Address and Size / Information bytes for the following

packet from $C5 and optionally the Control byte and the Frame Head Checksum A byte from $C9. The

CMX882 will read the Size and Message formatting information and if the message is of type 3, 4 or 5

(see section 1.5.5.4) it will set the automatic decoding of the following data; descrambling, de-interleaving,

decoding error correction bytes, stripping out pad bytes, calculating and checking Checksum B as

required. The only task the host need perform during the reception of formatted Frames is to read out

data when it is ready.

Note: These 2 words must be read separately. i.e. Two 16 bit C-BUS transactions.

FRS Signalling Processor CMX882

1.6.19 $CC TONES STATUS: 16-bit read-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

Detected Selcall tone frequency Sub-Audio Status 0 0 Detected CTCSS code

This word holds the current status of the CMX882 sub-audio and In-band tone sections. This word should

be read by the host after an interrupt caused by a DCS, CTCSS or In-band tone event.

The value in bits 5 to 0, Detected CTCSS code, identifies the detected sub-audio tone by its position in

Table 3 CTCSS Tones. If bits 5 to 0 = '000000' there is no CTCSS tone currently being detected. If bits

5 to 0 = '110111' (= 55 in decimal) this indicates that an Invalid Tone has been detected. An Invalid Tone

is any tone in the subaudio band which is not the selected subaudio tone nor the all-call tone, or is a tone

not listed in Table 3. A change in the state of bits 5-0 to Invalid Tone from the no tone condition will not

cause Status register ($C6), b11 to be set to '1'. Any other change in the state of bits 5-0 will cause the

Status register ($C6), b11 to be set to '1'.

A detected In-band frequency is indicated by the value in bits 15 to 11, ‘Detected In-band tone frequency’,

identifies the frequency by its position in Table 6 In-band Tones. If bits 15 to 11 = '00000' there is no In-

band tone currently being detected. A change in the state of bits 15 to 11 will cause bit 13 of the Status

and No Tone.

Bits 10 to 8 indicate the DCS and special sub-audio tone status. The Status register ($C6) will indicate

the type of signal detected. If DCS or special CTCSS tones are detected they will be indicated in bits 10

to 8 according to the table below and bits 7 to 0 will be set to '00000000'. If a normal CTCSS tone is

detected bits 10 to 8 will be set to '000' and bits 7 to 0 will indicate the decoded tone. A change in the

state of bits 10 to 8 will cause the relevant bit (10 or 11) of the Status register to be set to '1'.

Bit 10 Bit 9 Bit 8 Sub-Audio status

0 0 0 No DCS or special CTCSS detected

0 0 1 Reserved

0 1 0 DCS sequence detected Only enabled with DCS

0 1 1 inverted DCS sequence detected Only enabled with DCS

1 0 0 Reserved

1 0 1 134.4Hz DCS turn off tone detected Only enabled with DCS

1 1 0 64.7Hz XTCSS sub-audio tone detected Only enabled with XTCSS

1 1 1 Reserved

When the relevant detection mode is not enabled, the associated bits will be set to '0'. In Tx mode this

During DCS receive, the device can flag an interrupt when the DCS code fails to be recognised. This may

be due to code dropout. The turn off tone may be flagged shortly after, if the transmission is ending.

Alternatively the DCS link may be restored and DCS detection will be flagged again.

FRS Signalling Processor CMX882

1.6.20 $C8 PROGRAMMING REGISTER: 16-bit write-only

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

First

Word Block

Num. Block Num.

or Data Programming Data

This register is used for programming various gains, levels, offset compensations, tones and codes. If the

user programs any of these values then bit 4 of $C0 (Power Down Control) must be set to '1'. Following a

C-BUS Reset or a Power Up Reset, the programmed values are initialised in accordance with the settings

described in section 1.6.2 (C-BUS Reset).

The Signal Processing function and the XTAL clock circuit must both be enabled in order to write to the

Programming Register, so Power Down Control register bit 5 must be set to '1' and bit 3 must be set to '0'.

All other interrupt sources should be disabled while loading the programming register blocks.

The Programming Register should only be written to when the Programming Flag bit (bit 0) of the Status

'0'). The Programming Flag is cleared when the Programming Register is written to. When the

corresponding programming action has been completed (normally within 250µs) the CMX882 will set the

flag back to '1' to indicate that it is now safe to write the next programming value. The Programming

sequence of 16-bit words to the Programming Register, in the order shown in the following tables. Writing

data to the Programming Register must be performed in the order shown for each of the blocks, however

the order in which the blocks are written is not critical. If later words in a block do not require updating the

user may stop programming that block when the last change has been performed. e.g. If only 'Fine output

gain 1' needs to be changed the host will need to write to P4.0, P4.1 and P4.2 only.

The user must not exceed the defined word counts for each block. The word P4.8 is allocated for

production testing and must not be accessed in normal operation.

The high order bits of each word define which block the word belongs to, and if it is the first word of that

block: Bit 15 Bit 14 Bit 13 Bit 12 Bit 11 – Bit 0

1 X X X 1st data for each block

0 X X X 2nd and following data

X 1 0 0 Write to block 0 (12 bit words)

X 1 0 1 Write to block 1 (12 bit words)

X 1 1 0 Write to block 2 (12 bit words)

X 1 1 1 Reserved - do not use

X 0 Write to block 4 (14 bit words)

Block 0 – Modem Configuration:

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P0.0 1 1 0 0 0 MSK Frame Sync LSB

P0.1 0 1 0 0 0 MSK Frame Sync MSB

P0.2 0 1 0 0 0 Scramble Seed 1 LSB

P0.3 0 1 0 0 0 Scramble Seed 1 MSB

P0.4 0 1 0 0 0 Scramble Seed 2 LSB

P0.5 0 1 0 0 0 Scramble Seed 2 MSB

P0.6 0 1 0 0 0 MSK Bit Sync LSB

P0.7 0 1 0 0 0 MSK Bit Sync MSB

FRS Signalling Processor CMX882

Block 1 – XTCSS and In-band tone Setup:

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P1.0 1 1 0 1 Audio band Tx level Emph

P1.1 0 1 0 1 XTCSS

tone length Audio band detect threshold In-band tone detect

bandwidth

P1.2 0 1 0 1 0 Programmable In-band Tone 0

P1.3 0 1 0 1 0 Programmable In-band Tone 1

P1.4 0 1 0 1 0 Programmable In-band Tone 2

P1.5 0 1 0 1 0 Programmable In-band Tone 3

Block 2 – CTCSS and DCS Setup:

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P2.0 1 1 1 0 CTCSS and DCS Tx level

P2.1 0 1 1 0

DCS

24 0 CTCSS and DCS detect threshold CTCSS detect

bandwidth

P2.2 0 1 1 0 DCS Code bits 11 – 0

P2.3 0 1 1 0 DCS Code bits 23/22 – 12

P2.4 0 1 1 0 Sub-audio drop out time 0

Block 3 – Reserved. Do not use.

Block 4 – Gain and Offset Setup:

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P4.0 1 0 Fine Input Gain

P4.1 0 0 Reserved - set to '0'

P4.2 0 0 Fine Output Gain 1

P4.3 0 0 Fine Output Gain 2

P4.4 0 0 Output 1 Offset Control

P4.5 0 0 Output 2 Offset Control

P4.6 0 0 Ramp Rate Control

P4.7 0 0 Limiter Setting (all 1's = Vbias +/- 0.5 Vdd)

P4.8 0 0 Special Programming Register (Production Test Only)

FRS Signalling Processor CMX882

1.6.20.1 PROGRAMMING REGISTER Block 0 – Modem Configuration:

$C8 (P0.0-1) MSK Frame Sync

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P0.0 1 1 0 0 0 MSK Frame Sync LSB

P0.1 0 1 0 0 0 MSK Frame Sync MSB

Bits 7 to 0 set the Frame Sync pattern used in Tx and Rx MSK data. Bit 7 of the MSB is compared to the

earliest received data. After a power on or C-BUS reset they are set to $CB23.

$C8 (P0.2-5) Scramble Seed 1 and 2

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P0.2 0 1 0 0 0 Scramble Seed 1 LSB

P0.3 0 1 0 0 0 Scramble Seed 1 MSB

P0.4 0 1 0 0 0 Scramble Seed 2 LSB

P0.5 0 1 0 0 0 Scramble Seed 2 MSB

These bits set the scramble seed used on all data bi ts following a Frame Head. If $0000 is programmed

as the seed then no scrambling will occur when selected. If either programmable scramble seeds are

selected, both the transmit and receive devices must use the same seed pattern for data to be transferred

correctly.

$C8 (P0.6-7) MSK Bit Sync

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P0.6 0 1 0 0 0 MSK Bit Sync LSB

P0.7 0 1 0 0 0 MSK Bit Sync MSB

This bit pattern is used when transmitting the bit sync portion of a Frame Head. After a power on or C-

BUS reset they are set to $5555 (‘0101…0101’).

FRS Signalling Processor CMX882

1.6.20.2 PROGRAMMING REGISTER Block 1 – XTCSS and In-band tone Setup:

$C8 (P1.0) Voice band tones Tx Level

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P1.0 1 1 0 1 Voice band tones Tx level Emph

Bits 11 (MSB) to 1 (LSB) set the transmitted In-band tone, Audio Tone and MSK signal level (pk-pk) with a

resolution of VDD(A)/2048 per LSB (1.465mV per LSB at VDD(A)=3V). Valid range for this value is 0 to

1536.

Bit 0 controls Rx In-band tone de-emphasis. When set to '0' the signal going to the In-band tone detector

is not de-emphasised. When voice processing is enabled in the Mode register, de/pre-emphasis is

enabled in the Audio & Device Address register and this bit (b0) is set to '1', signals going to the In-band

tone detector are de-emphasised in accordance with Figure 6.

$C8 (P1.1) In-band tone Detect Bandwidth and Audio Band Detect Threshold

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P1.1 0 1 0 1 XTCSS

tone length Audio band detect threshold In-band tone detect

bandwidth

XTCSS tone length (bits 11 to 10) set the transmit tone length for each of the 4 tones in a XTCSS

sequence. In receive these bits define the minimum silent prefix and suffix qualification periods for

successful reception, also the maximum receive tone length is double the time set (e.g. for 60ms setting

each received tone must be less than 120ms in length for successful XTCSS decoding). '00' = 40ms,

'01' = 60ms, '10' = 80ms and '11' = 100ms.

The ‘detect threshold’ bits (bits 9 to 4) set the minimum In-band tone and/or MSK signal level that will be

detected. The levels are set according to the formula:

Minimum Level = Detect Threshold × 3.63mV rms at VDD(A) = 3V

The In-band tone detected bandwidth is set in accordance with the following table:

BANDWIDTH

Bit 3 Bit 2 Bit 1 Bit 0 Will Decode Will Not Decode

1 0 0 0 ±1.1% ±2.4%

Recommended for EEA ⇒ 1 0 0 1 ±1.3% ±2.7%

1 0 1 0 ±1.6% ±2.9%

1 0 1 1 ±1.8% ±3.2%

$C8 (P1.2-5) Programmable In-band Tones

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P1.2-5 0 1 0 1 0 Programmable In-band Tone

N (see below) R (see below)

These words set the programmable In-band tones used in transmit and receive. The frequency is set in

bits 10-0 for each word according to the formula:

N = Integer part of (0.042666 x frequency)

R = (0.042666 x frequency - N) x 6000 / frequency (round to nearest integer)

Example: For 1010Hz, N = 43, R = 1. The programmed tones must only be set to frequencies from

400Hz to 3000Hz.

FRS Signalling Processor CMX882

1.6.20.3 PROGRAMMING REGISTER Block 2 – CTCSS and DCS Setup:

$C8 (P2.0) CTCSS and DCS TX LEVEL

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P2.0 1 1 1 0 CTCSS and DCS Level

Bits 11 (MSB) to 0 (LSB) set the transmitted CTCSS or DCS sub-audio signal level (pk-pk) with a

resolution of VDD(A)/16384 per LSB (0.183mV per LSB at VDD(A)=3V, giving a range 0 to 749.8mV pk-

pk).

$C8 (P2.1) CTCSS TONE BW AND LEVEL

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P2.1 0 1 1 0

DCS

24 0 CTCSS and DCS detect threshold CTCSS detect

bandwidth

Bit 11, DCS 24, sets the length of DCS code transmitted or searched for. When this bit is set to ‘1’ 24 bit

codes are transmitted and decoded. When this bit is set to ‘0’ 23 bit codes are used.

The ‘detect threshold’ bits (bits 9 to 4) set the minimum CTCSS or DCS signal level that will be detected.

The levels are set according to the formula:

Minimum Level = Detect Threshold × 2mV rms at VDD(A) = 3V

The CTCSS detected tone bandwidth is set in accordance with the following table:

BANDWIDTH

Bit 3 Bit 2 Bit 1 Bit 0 Will Decode Will Not Decode

Recommended for use with

split tones

±0.5%

±1.8%

0 1 1 1 ±0.8% ±2.1%

Recommended for CTCSS ⇒1 0 0 0 ±1.1% ±2.4%

1 0 0 1 ±1.3% ±2.7%

1 0 1 0 ±1.6% ±2.9%

1 0 1 1 ±1.8% ±3.2%

$C8 (P2.2-3) DCS CODE (LOWER) and DCS CODE (UPPER)

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P2.2 0 1 1 0 DCS Data (bits 11-0)

P2.3 0 1 1 0 DCS Data (bits 23/22-12)

These words set the DCS code to be transmitted or searched for. The least significant bit (bit 0) of the

DCS code is transmitted or compared first and the most significant bit is transmitted or compared last.

Note that DCS Data bit 23 is only used when bit 11 (DCS 24) of P2.1 is set to ‘1’.

$C8 (P2.4) SUBAUDIO DROP OUT TIME

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P2.4 0 1 1 0 Subaudio Drop Out Time 0

The Subaudio Drop Out Time defines the time that the sub-audio signal detection can drop out before

loss of sub-audio is asserted. The period is set according to the formula:

Time = Subaudio Drop Out Time × 8.0ms [range 0 to 120ms]

The setting of this register defines the maximum drop out time that the device can tolerate. The setting of

this register also determines the de-response time, which is typically 90ms longer than the programmed

drop out time.

FRS Signalling Processor CMX882

1.6.20.4 PROGRAMMING REGISTER Block 3 – Reserved

1.6.20.5 PROGRAMMING REGISTER Block 4 – Gain and Offset Setup

$C8 (P4.0) FINE INPUT GAIN

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P4.0 1 0 Fine Input Gain (unsigned integer)

Gain = 20 × log([32768-IG]/32768)dB IG is the unsigned integer value in the ‘Fine Input Gain’ field

Fine input gain adjustment should be kept within the range 0 to -3.5dB.

$C8 (P4.1) Reserved

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P4.1 0 0 Reserved - set to '0'

This register is reserved and should be set to '0'.

$C8 (P4.2-3) FINE OUTPUT GAIN 1 and FINE OUTPUT GAIN 2

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P4.2 0 0 Fine Output Gain 1 (unsigned integer)

P4.3 0 0 Fine Output Gain 2 (unsigned integer)

Gain = 20 × log([32768-OG]/32768)dB OG is the unsigned integer value in the ‘Fine Output Gain’ field

Fine output gain adjustment should be kept within the range 0dB to -3.5dB.

$C8 (P4.4-5) OUTPUT 1 OFFSET and OUTPUT 2 OFFSET

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P4.4 0 0 2’s complement offset for MOD_1, resolution = VDD(A)/16384 per LSB

P4.5 0 0 2’s complement offset for MOD_2, resolution = VDD(A)/16384 per LSB

Can be used to compensate for inherent offsets in the output path via MOD_1 (Output 1 Offset) and

MOD_2 (Output 2 Offset). It is recommended that the offset correction is kept within the range +/-50mV.

$C8 (P4.6) RAMP RATE CONTROL

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P4.6 0 0 Ramp Rate Up Control (RRU) Ramp Rate Down control (RRD)

The ramp-up and ramp-down rates can be independently programmed. The ramp rates apply to all the

analogue output ports. They only affect those ports being turned on (ramp-up) or turned off (ramp down).

The ramp rates should be programmed before ramping any outputs.

Time to ramp-up to full gain = (1 + RRU) × 1.333ms

Time to ramp down to zero gain = (1 + RRD) × 1.333ms

FRS Signalling Processor CMX882

$C8 (P4.7) TRANSMIT LIMITER CONTROL

Bit: 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

P4.7 0 0 Limiter Setting, resolution = VDD(A)/32768 per LSB

This unsigned number sets the clipping point (maximum deviation from the centre value) for the MOD_1

and MOD_2 pins. The maximum setting ($3FFF) is +/- VDD(A)/2 i.e. output limited from 0 to VDD(A).

The limiter is set to maximum following a C-BUS Reset or a Power Up Reset. The limiter is only applied to

voice signals, not internally generated audio band signals. The levels of internally generated signals must

be limited by setting appropriate transmit levels.

$C8 (P4.8) Special Programming Register – do not access.

FRS Signalling Processor CMX882

1.7 Application Notes

Receive

FFSK

Voice+SubAudio

CMX882

Transmit

FFSK

Voice+SubAudio

Dem odulator

2-point Tx

Modulator

Host

MicroController

GPS

Radio Section

BUS

MIC 1

MIC 2

Audio output

Audio input 1

Audio input 2

Discriminator

input

M odulator

output 1

M odulator

output 2

Dis

Figure 13 Possible FRS + GPS Configuration

1.7.1 CRC and FEC encoding inform ation

For messages with FEC coding the following matrix is used to calculate and decode bytes:

Data bits FEC bits

7 6 5 4 3 2 1 0 3 2 1 0

1 1 1 0 1 1 0 0 1 0 0 0

1 1 0 1 0 0 1 1 0 1 0 0

1 0 1 1 1 0 1 0 0 0 1 0

0 1 1 1 0 1 0 1 0 0 0 1

8 bit CRC is used in all frame heads with the following generator polynomial (GP):

8 + x7 + x4 + x3 + x 1 + x0

16 bit CRC is used at the end of sized data messages of up to 16 bytes with the following GP:

16 + x12 + x5 + x0

32 bit CRC is used at the end of sized data messages of over 16 bytes with the following GP:

32 + x31 + x30 + x28 + x27 + x25 + x24 + x22 + x21 + x 20 + x16 + x10 + x9 + x6 + x 0

The scrambler uses a 16 bit pseudo random number generator with the following feed back taps:

16 + x14 + x1 + x0

FRS Signalling Processor CMX882

1.7.2 Data Interleaving

The built in FFSK packeting includes the option of interleaving the data in each block (Type 5). This,

together with Forward Error Correction (FEC), reduces the effects of one of the commonest sources of

data errors which is burst noise. Interleaving does not add any bits to the message, the packet is

assembled in 'rows' and then transmitting in 'columns'.

Data (8 bits) FEC (4 bits)

0 1 2 3 4 5 6 7 8 9 10 11

12 13 14 15 16 17 18 19 20 21 22 23

24 25 26 27 28 29 30 31 32 33 34 35

36 37 38 39 40 41 42 43 44 45 46 47

In the above example the packet is assembled as 4 rows with 12 bits of information per row. When this

packet is transmitted interleaved the bits are sent over the communication channel in the following order:

0, 12, 24, 36, 1, 13, 25, 37, 2, 14, ... , 33, 45, 10, 22, 34, 46, 11, 23, 35, 47.

In the receiving modem the packet is re-assembled (de-interleaved) before error correction. The 882's

built in packet receive modem is able to recognise (by using the frame head bytes) when the data has

been interleaved by the transmitter and will decode the data using the correct method.

1.7.3 Scrambler Seed Transfer

The CMX882 data scrambler provides some security for information transfer by allowing the use of

different seed patterns. This requires that the transmitting and receiving devices have identical scrambler

seeds. Should two radios wish to transfer data using the scrambler the seed must be transferred between

them before data transfer takes place. This is best done 'off air' by prior arrangement. Obviously sending

the seed over air by voice or by using FFSK (with either no scrambling or the standard scrambler) will

enable an eavesdropper to grab the seed as it is transferred and then decode any following messages

using that seed.

The following method allows the transfer of a seed between two unrelated radios (from A to B) without the

actual 16 bit seed pattern appearing over the air:

Example (4 bit), X=3, Y=9

Radio Calculates Sends over air Calculates Sends over air

A Seed + X Seed + X 8 + 3 11

B Rx + Y Seed + X + Y 11 + 9 4

A Rx - X Seed + Y 4 - 3 1

B Rx - Y 1 - 9 = 8 -

The values used for the Seed, X and Y can be any number up to 65535 and all 3 numbers should be

different for each seed transfer.

This and other simple methods allow an eavesdropper to use basic mathematical techniques (11-4+1 in

the above example) to use the over air information to derive the original seed (they must however have

detected all three over air messages to do this). To reduce this possibility the host can employ more

sophisticated techniques for seed transfer (e.g. functions using large prime numbers) as this will greatly

increase the mathematical complexity required to find the original seed. However such methods are;

beyond the scope of this document, may not be permitted in certain territories and probably not

appropriate for the level of security provided by the inbuilt scrambler.

FRS Signalling Processor CMX882

1.8 Performance Specification

1.8.1 Electrical Performance

The performance data are target figures, that may change subject to the outcome of device evaluation.

Absolute Maximum Ratings

Exceeding these maximum ratings can result in damage to the device.

Min. Max. Unit

Supply: VDD(D)- VSS(D) −0.3 7.0 V

VDD(A)- VSS(A) −0.3 7.0 V

Voltage on any pin to VSS(D) −0.3 VDD(D) + 0.3 V

Voltage on any pin to VSS(A) −0.3 VDD(A) + 0.3 V

Current into or out of VDD(A), VSS(A), VDD(D) and VSS(D) −30 +30 mA

Current into or out of any other pin −20 +20 mA

Voltage differential between power supplies:

VDD(D) and VDD(A) 0 0.3 V

VSS(D) and VSS(A) 0 50 mV

D6 Package (SSOP) Min. Max. Unit

Total Allowable Power Dissipation at Tamb = 25°C 550 mW

... Derating 9 mW/°C

Storage Temperature −55 +125 °C

Operating Temperature −40 +85 °C

E1 Package (TSSOP) Min. Max. Unit

Total Allowable Power Dissipation at Tamb = 25°C 400 mW

... Derating 5.3 mW/°C

Storage Temperature −55 +125 °C

Operating Temperature −40 +85 °C

Operating Limits

Correct operation of the device outside these limits is not implied.

Notes Min. Max. Unit

Supply (VDD - VSS) 2.7 3.6 V

Operating Temperature −40 +85 °C

Clock/Xtal Frequency 11 18.3 18.6 MHz

Notes: 11 Nominal clock frequency is 18.432MHz.

FRS Signalling Processor CMX882

Operating Characteristics

For the following conditions unless otherwise specified:

External components as recommended in Figure 2.

Maximum load on digital outputs = 30pF.

Xtal Frequency = 18.432MHz ±0.01% (100ppm).

VDD = 2.7V to 3.6V; Tamb = −40°C to +85°C.

Signal levels are defined for Vdd = 3V.

Signal levels track with supply voltage, so scale accordingly

Reference Signal Level = 308mV rms at 1kHz with VDD = 3V.

Signal to Noise Ratio (SNR) in bit rate bandwidth.

Input stage gain = 0dB.

Output stage attenuation = 0dB.

DC Parameters Notes Min. Typ. Max. Unit

Supply Current

IDD(D) (VDD = 3.0V) 21 4.5 8.0 mA

IDD(A) (VDD = 3.0V) 21 1.0 2.0 mA

IDD(D) (All Power-saved) (VDD = 3.0V) 21 2.0 10 µA

IDD(A) (All Power-saved) (VDD = 3.0V) 21 2.0 10 µA

C-BUS Interface

Input Logic ‘1’ 70% VDD

Input Logic ‘0’ 30% VDD

Input Leakage Current (Logic ‘1’ or ‘0’) −1.0 1.0 µA

Input Capacitance - 7.5 pF

Output Logic ‘1’ (IOH = 120µA) 90% VDD

Output Logic ‘0’ (IOL = 360µA) 10% VDD

“Off” State Leakage Current 10 µA

IRQN (Vout = VDD(D))

−1.0 1.0 µA

REPLY_DATA (output HiZ) −1.0 1.0 µA

CLOCK_OUT

Output Logic ‘1’ (IOH = 120µA) 90% VDD

(IOH = 1mA) 80% VDD

Output Logic ‘0’ (IOL = 360µA) 10% VDD

(IOL = -1.5mA) 15% VDD

CLOCK/XTAL 22

Input Logic ‘1’ 70% VDD

Input Logic ‘0’ 30% VDD

Input current (Vin = VDD) 40 µA

Input current (Vin = VSS)

−40 µA

VBIAS 23

Output voltage offset wrt VDD/2 (IOL < 1µA) -2% +2% VDD

Output impedance 22

kΩ

Notes: 21 Not including any current drawn from the device pins by external circuitry.

22 Characteristics when driving the CLOCK/XTAL pin with an external clock source.

23 Applies when utilising VBIAS to provide a reference voltage to other parts of the

system. When using VBIAS as a reference, VBIAS must be buffered. VBIAS must

always be decoupled with a capacitor as shown in Figure 2.

FRS Signalling Processor CMX882

AC Parameters Notes Min. Typ. Max. Unit

CLOCK/XTAL Input

'High' pulse width 31 21 ns

'Low' pulse width 31 21 ns

Input impedance (at 18.432MHz)

Powered-up Resistance 150 kΩ

Capacitance 20 pF

Powered-down Resistance 300 kΩ

Capacitance 20 pF

Clock frequency 18.432 MHz

Clock stability/accuracy ±100 ppm

Clock start up (from power-save) 400 ms

CLOCK_OUT Output

CLOCK/XTAL input to CLOCK_OUT timing:

(in high to out high) 32 15 ns

(in low to out low) 32 15 ns

'High' pulse width 33 22 27.13 33 ns

'Low' pulse width 33 22 27.13 33 ns

VBIAS

Start up time (from power-save) 30 ms

Microphone, Input_2 and Disc Inputs

(MIC, INPUT_2, DISC)

Input impedance 34 1 MΩ

Input signal range 35 10 90 %VDD

Load resistance (pin 12, 14 and 16) 80 kΩ

Amplifier open loop voltage gain ⎫

(I/P = 1mV rms at 100Hz) ⎭ 60 dB

Unity gain bandwidth 1.0 MHz

Programmable Input Gain Stage 36

Gain (at 0dB) −0.5 0 0.5 dB

Cumulative Gain Error ⎫

(wrt attenuation at 0dB) ⎭ −1.0 1.0 dB

Notes: 31 Timing for an external input to the CLOCK/XTAL pin.

32 CLOCK/XTAL input driven by external source.

33 18.432MHz XTAL fitted.

34 With no external components connected

35 After multiplying by gain of input circuit, with external components connected.

36 Gain applied to signal at output of buffer amplifier, pin 12, 14 or 16

FRS Signalling Processor CMX882

AC Parameters Notes Min. Typ. Max. Unit

Modulator Outputs 1 and 2 and Audio Output

(MOD_1, MOD_2, AUDIO)

Power-up to output stable 37 50 100 µs

Modulator Attenuators

Attenuation (at 0dB) −0.2 0 0.2 dB

Cumulative Attenuation Error ⎫

(wrt attenuation at 0dB) ⎭ −0.6 0.6 dB

Output Impedance ⎫ Enabled 38 600 Ω

⎭ Disabled 38 500 kΩ

Audio Attenuator

Attenuation (at 0dB) −0.5 0 0.5 dB

Cumulative Attenuation Error ⎫

(wrt attenuation at 0dB) ⎭ −1.0 1.0 dB

Output Impedance ⎫ Enabled 38 600 Ω

⎭ Disabled 38 500 kΩ

Feedback load resistance 80 kΩ

Amplifier open loop voltage gain ⎫

(I/P = 1mV rms at 100Hz) ⎭ 60 dB

Unity gain bandwidth 1.0 MHz

Notes: 37 Power-up refers to issuing a C-BUS command to turn on an output. These limits

apply only if VBIAS is on and stable. At power supply switch-on, the default state

is for all blocks, except the XTAL and C-BUS interface, to be in placed in power-

save mode.

38 Small signal impedance, at VDD = 3.0V and Tamb = 25°C.

FRS Signalling Processor CMX882

AC Parameters (cont.) Notes Min. Typ. Max. Unit

Auxiliary ADC (Signal Monitor)

8 Bit ADC Mode

Resolution 8 Bits

Input Range 10% 90% VDD

Conversion time 41 20.8 µs

Input impedance

Resistance 10

MΩ

Capacitance 5 pF

Zero error ⎫

(input offset to give ADC output = 0) ⎭ −20 +20 mV

Integral Non-linearity ⎫ 42 2 LSB

⎭ 43 4 LSB

Differential Non-linearity ⎫ 42 1 LSB

⎭ 43 3 LSB

Source output impedance 44 24 kΩ

Level Threshold Detect Mode

Threshold Resolution 8 Bits

Upper threshold range (VTH) 45 VTL VDD(A) V

Lower threshold range (VTL) 45 VSS(A) VTH V

Signal Monitor change to IRQ 46 120 µs

Signal Monitor change to Receiver-Turn-On 47 60 µs

Audio Compandor

Attack time 4.0 ms

Decay time 13 ms

0dB point 48 100 mVrms

Compression / Expansion ratio 2:1

Notes: 41 With clock frequency of 18.432MHz.

VDD(A) ≥ 3.0V

43 VDD(A) < 3.0V

44 Denotes output impedance of the driver of the Signal Monitor input, to ensure < 1

bit additional error under nominal conditions.

45 Upper threshold > Lower threshold

46 Time from Signal Monitor input rising above Upper Threshold or falling below

Lower Threshold, to IRQN being asserted.

47 Time from Signal Monitor input rising above Upper Threshold to receiver path

powering up, settling and starting automatic signal type identification.

48 VDD(A) = 3.0V

FRS Signalling Processor CMX882

AC Parameters (cont.) Notes Min. Typ. Max. Unit

Receiver Signal Ty pe Identification

Probability of correctly identifying signal type

(SNR = 12dB) >>99.9 %

CTCSS Detector

Sensitivity (Pure Tone) 51 -26 dB

Response Time (Composite Signal) 52 140 250 ms

De-response Time (Composite Signal) 52, 55 210 ms

Dropout immunity 55 160 ms

Frequency Range 60 260 Hz

IN-BAND TONE Detector

Sensitivity (Pure Tone) 53 -26 dB

Response Time (Good Signal) 35 ms

De-response Time (Good Signal) 45 ms

Dropout immunity 20 ms

Frequency Range (In-band tone) 400 3000 Hz

DCS Decoder

Sensitivity 51 58 mVp-p

Bit-Rate Sync Time 2 edges

FFSK/MSK Decoder

Signal Input Dynamic Range 54 100 800 mVrms

Bit Error Rate (SNR = 20dB) 54 <1 10-8

Receiver Synchronisation (SNR = 12dB)

Probability of bit 16 being correct >99.9 %

Notes: 51 Sub-Audio Detection Level threshold set to 16mV.

52 Composite signal = 308mV rms at 1kHz + 75mV rms Noise + 31mV rms Sub-

Audio signal. Noise bandwidth = 5kHz Band Limited Gaussian.

53 In-band Tone Detection Level threshold set to 16mV.

54 VDD(A) = 3.0V, for a “101010101 ... 01” pattern measured at the input amplifier

feedback pin (12). Signal level scales with VDD(A). See

Figure 15 for variation of BER with SNR.

With sub-audio dropout time (P2.4) set to ≥ 120ms. The typical dropout immunity

is approximately 40ms more than the programmed dropout immunity. The typical

de-response time is approximately 90ms longer than the programmed dropout

immunity. See section 1.6.20.3, P2.4.

FRS Signalling Processor CMX882

AC Parameters (cont.) Notes Min. Typ. Max. Unit

CTCSS Encoder

Frequency Range 60.0 260 Hz

Tone Frequency Accuracy ±0.3 %

Tone Amplitude Tolerance 61 -1.0 +1.0 dB

Total Harmonic Distortion 62 2.0 4.0 %

In-band tone Encoder

Frequency Range 400 3000 Hz

Tone Frequency Accuracy ±0.3 %

Tone Amplitude Tolerance 63 -1.0 +1.0 dB

Total Harmonic Distortion 62 2.0 4.0 %

DCS Encoder

Bit Rate 134.4 bits/s

Amplitude Tolerance 61 -1.0 +1.0 dB

FFSK/MSK Encoder

Output signal level 775 mVrms

Output level variation -1.0 +1.0 dB

Output distortion 5 %

3rd harmonic distortion 3 %

Logic 1 freq 1200baud and

2400baud 1198 1200 1202 Hz

Logic 0 freq 1200baud 1798 1800 1802 Hz

2400baud 2398 2400 2402 Hz

Isochronous distortion (0 to 1 and 1 to 0) 40 µs

Notes: 61 VDD(A) = 3.0V and TX Sub-Audio Level set to 88mV p-p (31mV rms).

62 Measured at MOD_1 or MOD_2 output.

63 VDD(A) = 3.0V and Tx Audio Level set to 871mV p-p (308mV rms).

FRS Signalling Processor CMX882

AC Parameters (cont.) Notes Min. Typ. Max. Unit

Analogue Channel Audio Filtering

Pass-band (nominal bandwidth):

Received voice 71 300 3000 Hz

12.5kHz channel transmitted voice 72 300 2550 Hz

25kHz channel transmitted voice 73 300 3000 Hz

Pass-band Gain (at 1.0kHz) 0 dB

Pass-band Ripple (wrt gain at 1.0kHz) -2 +0.5 dB

Stop-band Attenuation 33.0 dB

Residual Hum and Noise 76 -50 dBp

Pre-emphasis 74 6 dB/oct

De-emphasis 75

−6 dB/oct

Audio Scrambler

Inversion frequency 3300 Hz

Pass band 300 3000 Hz

Notes: 71 The receiver voice filter complies with the characteristic shown in Figure 5. The

high pass filtering removes sub-audio components from the audio signal.

72 The 12.5kHz channel filter complies with the characteristic shown in Figure 9.

73 The 25kHz channel filter complies with the characteristic shown in Figure 8.

74 The pre-emphasis filter complies with the characteristic shown in Figure 10.

75 The de-emphasis filter complies with the characteristic shown in Figure 6.

76 dBp represents a psophometrically weighted measurement.

FRS Signalling Processor CMX882

C-BUS Timing

Figure 14 C-BUS Timing

C-BUS Timing Notes Min. Typ. Max. Unit

tCSE CSN Enable to SClk high time 100 ns

tCSH Last SClk high to CSN high time 100 ns

tLOZ SClk low to ReplyData Output Enable

Time 0.0 ns

tHIZ CSN high to ReplyData high impedance 1.0 µs

tCSOFF CSN high time between transactions 1.0 µs

tNXT Inter-byte time 200 ns

tCK SClk cycle time 200 ns

tCH SClk high time 100 ns

tCL SClk low time 100 ns

tCDS Command Data setup time 75 ns

tCDH Command Data hold time 25 ns

tRDS Reply Data setup time 50 ns

tRDH Reply Data hold time 0 ns

Notes: 1. Depending on the command, 1 or 2 bytes of COMMAND DATA are transmitted to the

peripheral MSB (Bit 7) first, LSB (Bit 0) last. REPLY DATA is read from the peripheral MSB

(Bit 7) first, LSB (Bit 0) last.

2. Data is clocked into the peripheral on the rising SERIAL_CLOCK edge.

3. Commands are acted upon at the end of each command (rising edge of CSN).

4. To allow for differing µC serial interface formats C-BUS compatible ICs are able to work

with SERIAL_CLOCK pulses starting and ending at either polarity.

5. Maximum 30pF load on IRQN pin and each C-BUS interface line.

These timings are for the latest version of C-BUS and allow faster transfers than the original C-BUS timing

specification. The CMX882 can be used in conjunction with devices that comply with the slower timings,

subject to system throughput constraints.

FRS Signalling Processor CMX882

1.0E-05

1.0E-04

1.0E-03

1.0E-02

1.0E-01

4567891011121314

SNR (dB) - Noise in b i t rate b a n d width

1200 b/s

2400 b/s

Figure 15 Typical FFSK/MSK Bit Error Rate Graph

FRS Signalling Processor CMX882

1.8.2 Packaging

Figure 16 Mechanical Outline of 28-pin SSOP (D6): Order as part no. CMX882D6

Figure 17 Mechanical Outline of 28-pin TSSOP (E1): Order as part no. CMX882E1

FRS Signalling Processor CMX882

Handling precautions: This product includes input protection, however, precautions should be taken to prevent device

damage from electro-static discharge. CML does not assume any responsibility for the use of any circuitry described. No

IPR or circuit patent licences are implied. CML reserves the right at any time without notice to change the said circuitry and

this product specification. CML has a policy of testing every product shipped using calibrated test equipment to ensure

compliance with this product specification. Specific testing of all circuit parameters is not necessarily performed.

www.cmlmicro.com

For FAQs see: www.cmlmicro.com/products/faqs/

For a full data sheet listing see: www.cmlmicro.com/products/datasheets/download.htm

For detailed application notes: www.cmlmicro.com/products/applications/

Oval Park, Langford,

Maldon, Essex,

CM9 6WG - England.

Tel: +44 (0)1621 875500

Fax: +44 (0)1621 875600

Sales:

sales@cmlmicro.com

Technical Support:

techsupport@cmlmicro.com

4800 Bethania Station Road,

Winston-Salem,

NC 27105 - USA.

Tel: +1 336 744 5050,

800 638 5577

Fax: +1 336 744 5054

Sales:

us.sales@cmlmicro.com

Technical Support:

us.techsupport@cmlmicro.com

No 2 Kallang Pudding

Road, #09 - 05/06 Mactech

Industrial Building,

Singapore 349307

Tel: +65 6745 0426

Fax: +65 6745 2917

Sales:

sg.sales@cmlmicro.com

Technical Support:

sg.techsupport@cmlmicro.com

No. 218, Tian Mu Road

West, Tower 1, Unit 1008,

Shanghai Kerry Everbright

City, Zhabei,

Shanghai 200070,

China.

Tel: +86 21 6317 4107

+86 21 6317 8916

Fax: +86 21 6317 0243

Sales:

cn.sales@cmlmicro.com.cn

Technical Support:

sg.techsupport@cmlmicro.com