Performance improvement of Viterbi and Sova decoding Algorithms with finite decoding length

This thesis proposes a new modified version of Viterbi Algorithm and modification of that algorithm into Soft-Output version of Viterbi Algorithm is also presented. This modification take care of the memory requirement of the SMU (Survivor Memory Unit) into account and using the concept of decoding length(depth), by introducing the finite decoding depth it has been observed that there is improvement in error performance of Viterbi and Soft-Output Viterbi Algorithm. Besides the improvement in Bit Error Rate Vs Signal to Noise curves, some hardware improvements i.e. reduction in size and power consumption in SMU and TBU (Trace Back Unit) is also achieved.