Reduced Complexity Decoding Algorithms for Repeat Accumulate Codes
XU Zong-ze · Journal of Sichuan University · 2004
Repeat-accumulate (RA) codes have a natural linear-time encoding algorithm. Under BP decoding algorithm, RA codes can be decoded reliably in linear time at rates arbitrarily closed to channel capacity.However,BP algorithm has high complexities.Minimum sum decoding algorithm is a sub-optimum algorithm and is easy for hardware implementation,but the decoding performance is degraded. Normalized algorithm and offset algorithm are proposed as a good tradeoff between decoding performance and complexities. Simulation results show that with little complexity increased, normalized algorithm and offset algorithm have better performance than that of minimum sum algorithm.