RALCWI Vocoder
CMX608/CMX618/CMX638
This timed method may also be used where the host clock and CODEC clock are unrelated, in which case
the host must supply a synchronising signal. Further details of slip management are given in section 6.5.
5.2.2.
Multiple Frame Packet with FEC, but without STD or DTMF
This is the most suitable configuration for use in a wireless system where the channel is prone to bit errors.
Two modes are available:
3 frame packet, which represents 60ms of voice. This mode is most suitable for TDMA/DMR.
4 frame packet, which represents 80ms of voice. This mode is most suitable for FDMA/DMR.
Figure 9 shows the 3 frame mode, but the principal applies equally to the 4 frame mode. There are two
basic strategies that can be adopted for servicing the encoder:
Event-driven
This is exactly the same as for the single frame example. Bit 0 (VDA) in the STATUS ($40) register will be
set to '1' when an FEC protected frame becomes available. If interrupts are enabled then IRQN will also
be pulled low.
Timed
It is assumed that the host has an accurate method of timing, as discussed in the single frame example.
Instruct the device to encode. Wait for 80ms, then read the first Vocoder packet. Wait for 60ms before
reading the next and subsequent packets.
Figure 9 Multiple Frame Packet Encoding
Figure 9 shows the sequence of events. Adding the FEC is shown twice, though it is only executed once.
This is to show that it will start immediately after the 3rd vocoded frame is available - which depends on the
algorithmic jitter for that frame.
In the case of 4 frame packets, the initial wait should be 100ms and the subsequent waits should be 80ms.
5.3.
Decoder
Like the encoder, the decoder deals with a basic frame size of 20ms, 160 samples, of audio. 1, 2, 3, or 4
frames may be collected together and supplied to the decoder as a single packet. In the case of the 3 and
4 frame packets, error protection may also be included with the FEC option. FEC protected packets may
contain either soft bits or hard bits. A soft-bit packet is 4 times the size of a hard-bit packet, as each soft-
bit is represented by a nibble. The decoder can also detect special STD and DTMF frames and re-
produce these as tones in the audio stream.
2014 CML Microsystems Plc
16
D/608_18_38/11
CMLMICRO [ CML MICROCIRCUITS ]
