一种多元极化码速率分配的低复杂度方法

Methods for Enhancing Successive Cancellation Decoding of Polar Codes

研究多元离散无记忆信道下速率分配问题。在码长有限时,极化现象不完全,子信道并没有完全极化到它的理想信道容量阶次。设计子信道速率方案使得其恰好契合不同子信道多样化的信道容量得以实现。当子信道传输信息比特溢出于子信道容量时,错误将会发生。对于有限码长下多元极化信道下的块错误概率(block error rate,BLER)进行线性建模,并提出了一种线性优化算法来实现多元极化码速率分配。对比穷举算法码长指数级的复杂度,线性优化算法复杂度随码长线性增长。仿真结果表明低码长情况下线性优化算法具有和穷举算法一致的最优BLER。对比二元速率分配方案,线性优化算法对BLER优化有显著增益。四元删除信道下的四元极化码用作例子,给出详细说明。

In this paper, we study methods to enhance successive cancellation decoding of polar codes. We propose two im- proved algorithms, which are two-path decision delay decoding and variable path decision delay decoding with threshold. Con ventional successive cancellation decoding is greedy algorithm in code tree, which means that successive cancellation decoding can＇t correct errors from the former nodes. In this paper, the decision delay decoding is used to provide opportunity to correct previous error for the current node. The proposed two-path and variable path decision delay decoding can improve the decoding performance by increasing the computing nodes and increasing the storage space. The simulation results show that compared to successive cancellation decoding, the two-path decision delay decoding has 1.1dB gain on decoding l-erformanee. In addition, variable path decision delay decoding has more gain.