电子设计工程
電子設計工程
전자설계공정
ELECTRONIC DESIGN ENGINEERING
2012年
13期
57-60
,共4页
戚楠%李立欣%范捷%蒋少豪%张会生
慼楠%李立訢%範捷%蔣少豪%張會生
척남%리립흔%범첩%장소호%장회생
Turbo码%Max-Log-Map译码优化算法%分块并行%DSP
Turbo碼%Max-Log-Map譯碼優化算法%分塊併行%DSP
Turbo마%Max-Log-Map역마우화산법%분괴병행%DSP
Turbo%Max-Log-Map decoding algorithm%block and parallel manipulation%DSP
针对传统的Max-Log-Map译码算法时效性差、存储空间开销大的特点，本文对传统的Max-Log-Map译码算法进行了改进。改进的算法对前、后向度量使用了蝶形结构图，便于DSP实现；将原始帧均分为多个子块，设计子块闻的并行运算以减小系统延迟；子块内采取进一步地优化措施，以减小数据存储量并提高译码速率。在DSPC6416平台上的仿真结果表明了算法的可实现性与可靠性。
針對傳統的Max-Log-Map譯碼算法時效性差、存儲空間開銷大的特點，本文對傳統的Max-Log-Map譯碼算法進行瞭改進。改進的算法對前、後嚮度量使用瞭蝶形結構圖，便于DSP實現；將原始幀均分為多箇子塊，設計子塊聞的併行運算以減小繫統延遲；子塊內採取進一步地優化措施，以減小數據存儲量併提高譯碼速率。在DSPC6416平檯上的倣真結果錶明瞭算法的可實現性與可靠性。
침대전통적Max-Log-Map역마산법시효성차、존저공간개소대적특점，본문대전통적Max-Log-Map역마산법진행료개진。개진적산법대전、후향도량사용료접형결구도，편우DSP실현；장원시정균분위다개자괴，설계자괴문적병행운산이감소계통연지；자괴내채취진일보지우화조시，이감소수거존저량병제고역마속솔。재DSPC6416평태상적방진결과표명료산법적가실현성여가고성。
With the traditional Max-Log-Map decoding algorithm being inefficiency and consuming too much storage space, Turbo code cannot be applied unrestrainedly. This paper presents a modified Max-Log-Map decoding algorithm. Firstly,we simplify the forward metrics and the backward one, so that they can be easily implemented on DSP. Besides, the operation of the data block is used before the parallel manipulation. Then, each block is further optimized to decrease the storage cost and enhance the transmission rate. The resulting performances of BER and transmission rate manifest that the modified algorithm is realizable and reliable.
