Maximum-Likelihood Priority-First Search Decodable Codes for Combined Channel Estimation and Error Protection - Computer Science > Information TheoryReportar como inadecuado




Maximum-Likelihood Priority-First Search Decodable Codes for Combined Channel Estimation and Error Protection - Computer Science > Information Theory - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: The code that combines channel estimation and error protection has receivedgeneral attention recently, and has been considered a promising methodology tocompensate multi-path fading effect. It has been shown by simulations that suchcode design can considerably improve the system performance over theconventional design with separate channel estimation and error protectionmodules under the same code rate. Nevertheless, the major obstacle thatprevents from the practice of the codes is that the existing codes are mostlysearched by computers, and hence exhibit no good structure for efficientdecoding. Hence, the time-consuming exhaustive search becomes the only decodingchoice, and the decoding complexity increases dramatically with the codewordlength. In this paper, by optimizing the signal-tonoise ratio, we found asystematic construction for the codes for combined channel estimation and errorprotection, and confirmed its equivalence in performance to thecomputer-searched codes by simulations. Moreover, the structural codes that weconstruct by rules can now be maximum-likelihoodly decodable in terms of anewly derived recursive metric for use of the priority-first search decodingalgorithm. Thus,the decoding complexity reduces significantly when comparedwith that of the exhaustive decoder. The extension code design for fast-fadingchannels is also presented. Simulations conclude that our constructed extensioncode is robust in performance even if the coherent period is shorter than thecodeword length.



Autor: Chia-Lung Wu, Po-Ning Chen, Yunghsiang S. Han, Ming-Hsin Kuo

Fuente: https://arxiv.org/







Documentos relacionados