FRS Signalling Processor CMX882
2004 CML Microsystems Plc 24 D/882/7
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.