Comparative analysis of convolutional codes based on ML decoding

Abstract

Viterbi algorithm is based on Maximum Likelihood Decoding Method. It finds a path from 2k total path through trellis that maximizes the path conditional probability. During decoding it compares the metric of all partial path entering each state. this paper involves brief information about Convolutional code in terms of strategies of encoding and decoding of convolutional codes and also, Recursive Systematic Convolutional Encoder is designed with BPSK Modulation and decoding is done by use of ML Decoding Method, compares the BER vs SNR performance at different code rates and find that as SNR increases the BER improves and also as number of bits are increases the performance of communication system also improves. At 3 db SNR the BER is 10⁁-2, when the number of bits are 2048 and 4096 at 4db SNR the BER is close to 10⁁-3.