分块并行Turbo码译码算法的研究

A study of sub-block parallel turbo decoding algorithm

Turbo码译码采用迭代译码思想,译码时延较大是其应用于实时性要求较高的通信系统中的一大障碍.为了减少译码计算的时延,利用递推迭代的思想,给出一种分块并行译码的方法,即将接收的整个码字分成若干子块,各子块进行并行处理,其中各子块的前后向递推公式的初始值由相邻子块的前一次迭代译码的边界计算值传递.实验仿真结果表明这种并行译码方法可以取得较好的译码性能,在硬件实现方面可以大大降低译码计算复杂度和时延,从而降低整个Turbo码译码时延.

Turbo decoding adopts the idea of iterative decoding in which the decoding delay is an obstacle to the application of a real-time communication system. For the purpose of cutting down the delay of the decoding calculation, this paper presents an algorithm of sub-block parallel decoding on the basis of the theory of the handover and iteration. Here,the whole received codeword is divided into several sub-blocks and every sub-block performs parallel processing in which the initial value of the forward and backward variables is passed by the boundary calculation value of the previous iteration of the contiguous sub-block. The emulation result indicates that the parallel decoding algorithm can achieve preferable BER performance and decrease the complexity and the delay of decoding calculation greatly in the hardware. Therefore, it can reduce the delay time of the whole Turbo decoding.