Soft decision Viterbi decoding
(step 1)


Fig. 1 Viterbi decoding. The labels on the branches indicate branch metrics; cumulative metrics of surviving paths are given in frames. The decoding may continue.

In contrast to the situation in lesson 5 we assume now that the demodulator outputs a graded multilevel confidence measure concerning the probability of a binary one and a binary zero. This kind of demodulator operation is referred to as soft decision demodulation.

Figure 1 shows the the Viterbi decoding proces. It is soft-decision Viterbi decoding. An encoded data sequence 1 0 1 1 is transmitted. The state sequence starts in state zero. We assume BPSK modulation. The received sequence is
r = (1, -0.5), (1, 0.1), (-1, -1,2), (-0.8, -0,8), (-0.8,+0.8), (+0.5,+0.5)
where the demodulator outputs +1 and -1 correspond to logical 1 and 0, respectively.


Please, verify the example with our Viterbi Algorithm Applet