The iterative channel decoder of joint source-channel decode has the transform of log-likelihood and probability, also the multiplication and accumulation operation, which improves the computation complexity of channel decoder. In order to reduce the complexity, this paper proposes the simplified iterative channel decoding algorithm. The algorithm directly utilizes the hard decided value of channel decoder and parameter prior probability to estimate the extrinsic information for iterative channel decoder. Based on Gaussian-Markov source, simulation results show that the simplified iterative channel decoder significantly improves SNR of received parameters and reduces the computation compared with separation source channel decoder. Also, the simplified algorithm does not degrade the performance of joint source channel decoding algorithm obviously.
Shannon C E. A mathematical theory of communication [J].Bell System Technical Journal, 1948, 27: 379-423, 623-656.[2]Fingscheidt T, Hindelang T, and Cox R V, et al.. Jointsource-channel (de-) coding for mobile communications [J].IEEE Transactions on Communications.2002, 50(2):200-211[3]Fingscheidt T and Vary P. Soft bit speech decoding: A newapproach to error concealment[J].IEEE Transactions onSpeech and Audio Processing.2001, 9(3):240-251[4]Subramaniam A D, Gardner W R, and Rao B D. Iterativejoint source-channel decoding of speech spectrum parametersover an additive white Gaussian noise channel [J].IEEETransactions on Audio Speech and Language Processing.2006,14(1):152-162[5]Liu Xiao-bei and Koh Soo Ngee. Iterative jointsource-channel decoding with combined prior information ofsource and channel [J].IEEE Communication Letters.2005,9(4):355-357[6]周琳, 吴镇扬. 基于迭代结构的联合信源信道GSM 语音解码[J].电子与信息学报.2005, 27(6):884-887浏览[7]周琳, 吴镇扬. 基于参数冗余量分配方案和可变参数估计准则的GSM EFR 信源信道联合解码算法[J].电子与信息学报.2008, 30(11):2617-2621浏览[8]Burr A. Modulation and Coding for WirelessCommunications [M]. United Kingdom: Pearson EducationLimited, 2001: 191-197.[9]Forney G D. Convolutional codes I: algebraic structure [J].IEEE Transactions on Information Theory.1970, 16(6):720-738