Implementation of the Maximum-Likelihood Soft-Decision Sequential Decoding Algorithm for Convolutional Codes

• Main Authors: Tsung-Ju Wu, 吳宗儒
• Other Authors: Yunghsiang S. Han
• Format: Others
• Language: en_US
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/09740979031861094765

碩士 === 國立暨南國際大學 === 資訊工程學系 === 90 === Maximum-likelihood soft-decision sequential decoding algorithm (MLSDA) was presented by Han and Chen. In this thesis, first of all, we present the MLSDA, then study and simulate the MLSDA imposed with stack size limitation, and time limitation. All simulations are based on 8 levels quantization receiver. In order to compensate the loss on signal-to-noise ratio due to reduction on the effective code rate of a convolutional code, a block decoding type MLSDA is designed. Simulation results of (2, 1, 6), (2, 1, 8) and (2, 1, 10) convolutional codes antipodally transmitted over the AWGN channel show that the average of computational effort required by MLSDA is several orders of magnitude less than the Viterbi algorithm when signal-to-noise ratio is greater than 5 dB. Furthermore, investigation on the block decoding type MLSDA restricted on time limitation indicates that proposed decoding algorithm has about ten times advantage on average complexity of the Viterbi algorithm at similar bit error rate performance.