Performance optimization and parallelization of turbo decoding for software-defined radio

This paper describes the optimization, parallelization, and simulated performance of a software double-binary turbo decoder implementation supporting the WiMAX standard suitable for software-defined radio (SDR). Turbo codes offer excellent error-correcting performance, but present high computational requirements, hence a parallel approach is desirable when seeking to exploit the flexibility of SDR. The development of a flexible parallel maximum a postiori (MAP) decoding algorithm is detailed, with simulation speedup results demonstrating good parallel efficiency (above 80%). For the same number of threads, a linear-log-MAP decoder implementation using 4 iterations was shown to be have nearly twice the throughput with comparable BER performance of a max-log-MAP decoder implementation using 8 iterations. In addition to parallel execution, other performance enhancements in software and through customized instructions provide a combined per-thread speedup of up to 57%.