一种简化的对数最大后验概率译码算法

提出了一种简化的对数最大后验概率译码算法．该简化算法基于逼近理论，用最佳平方逼近多项式近似计算校正函数，近似多项式的系数根据特征定理确定．与原算法相比，简化算法具有低复杂度、译码延时少的优点．仿真结果表明：计算复杂度比原算法降低约30％；在加性白高斯噪声信道和平坦慢衰落瑞利信道上，其编码增益比求最大值对数最大后验概率译码算法高出0．3～0．5dB，与原算法相近．因此，简化算法可替代对数最大后验概率译码算法在Turbo码译码器中使用．

Simplified Log Maximum A Posteriori Algorithm

A simplified log maximum a posteriori (Log-MAP) algorithm is presented based on approximation theory. In the simplified algorithm, the optimal square approximation polynomial, whose coefficients are derived from the characteristic equations, is adopted to calculate the correction function. Compared with the original algorithm, the proposed algorithm has advantages of low complexity and lower decoding delay. It is shown that the complexity is reduced by about 30%. The performances of the simplified Log-MAP algorithm are simulated over additive white Gaussian noise channel and flat fading Rayleigh channel. It is demonstrated that the coding gain of the simplified algorithm is 0.30.5 dB higher than that of Max-Log-MAP one, and is similar to that of Log-MAP Turbo decoder. It is concluded that the simplified algorithm is a better compromise between the complexity and the performance, and can be a substitute for Log-MAP algorithm in Turbo decoders.