Title :
A memory reduced decoding scheme for double binary convolutional turbo code based on forward recalculation
Author :
Zhan, Ming ; Zhou, Liang
Author_Institution :
Nat. Key Lab. of Sci. & Technol. on Commun., Univ. of Electron. Sci. & Technol. of China, Chengdu, China
Abstract :
In the implementation of iterative decoder for double binary convolutional turbo code (DB CTC), memory accessing accounts for a large part of the overall power consumption. In this paper, an iterative decoding scheme with small memory size is proposed. The new method is based on an improved maximum a posterior probability (MAP) algorithm, and stores part of the backward metrics in the state metrics cache (SMC). While at the corresponding time that the not stored metrics are used, they can be recalculated by a Compare-Select-Recalculate Processing (CSRP) unit in the forward direction. Since the memory size for SMC is 25% decreased as compared with conventional scheme, less memory accessing is needed. Moreover, complexity analysis and numerical simulation are presented to demonstrate the effectiveness of our proposed scheme.
Keywords :
binary codes; convolutional codes; iterative decoding; maximum likelihood estimation; turbo codes; CSRP unit; DB CTC; MAP algorithm; SMC; compare-select-recalculate processing; double binary convolutional code; forward recalculation; iterative decoder; iterative decoding scheme; maximum a posterior probability; memory reduced decoding scheme; state metrics cache; turbo code; Algorithm design and analysis; Convolutional codes; Decoding; Iterative decoding; Measurement; Memory management; Turbo codes; MAP algorithm; backward metrics; memory reduction; recalculation;
Conference_Titel :
Turbo Codes and Iterative Information Processing (ISTC), 2012 7th International Symposium on
Conference_Location :
Gothenburg
Print_ISBN :
978-1-4577-2114-4
Electronic_ISBN :
2165-4700
DOI :
10.1109/ISTC.2012.6325203