Title :
A memory efficient MAP algorithm for linear block codes
Author :
Ghayour, Farzad ; Takawira, Fambirai ; Xu, Hongjun
Author_Institution :
Sch. of Electr., Electron. & Comput. Eng., Univ. of KwaZulu-Natal, Durban, South Africa
Abstract :
It is usually assumed that for implementing the maximum a posteriori (MAP) decoding algorithm, saving all the state metrics is required. This memory requirement makes the MAP an impractical decoding algorithm especially for the non-binary block codes. In this paper, we present a modification of the MAP algorithm, in order to make it a more feasible algorithm. The computational complexity analysis and the memory requirements as well as the simulation results for the proposed algorithm are presented.
Keywords :
block codes; computational complexity; decoding; linear codes; maximum likelihood estimation; computational complexity analysis; linear block codes; maximum a posteriori decoding algorithm; memory efficient MAP algorithm; memory requirements; nonbinary block codes; state metrics; Algorithm design and analysis; Block codes; Complexity theory; Decoding; Measurement; Memory management; Vectors; Dual MAP; MAP algorithm; Memory saving; Non binary MAP;
Conference_Titel :
AFRICON, 2011
Conference_Location :
Livingstone
Print_ISBN :
978-1-61284-992-8
DOI :
10.1109/AFRCON.2011.6072085
