基于预译码的极化码最大似然简化连续消除译码算法

针对极化码译码串行输出造成较大译码时延的问题，该文提出一种基于预译码的最大似然简化连续消除译码算法。首先对译码树节点存储的似然值进行符号提取并分组处理，得到符号向量组；然后比较符号向量组与该节点的某些信息位的取值情况，发现向量组中储存的正负符号分布规律与该节点的中间信息位的取值具有一一对应的关系；在此基础上对组合码中间的1～2 bit进行预译码；最后结合最大似然译码方法估计组合码中的剩余信息位，从而得到最终的译码结果。仿真结果表明：在不影响误码性能的情况下，所提算法与已有的算法相比可有效降低译码时延。

Pre-decoding Based Maximum-likelihood Simplified Successive-cancellation Decoding of Polar Codes

To solve the long decoding latency caused by the serial nature of the decoding of polar codes, a pre-decoding based maximum-likelihood simplified successive-cancellation decoding algorithm is proposed. First, the signs of the likelihood values stored in the decoding tree nodes are extracted and grouped to obtain symbol vectors. Then comparing the symbol vectors and the values of some information bits, the distribution rules are found that positive and negative values stored in the vectors are one-to-one corresponding to the value of middle information bits of the node. Based on the above analysis, one or two bits in the middle of the constituent code are pre-decoded. Finally, the maximum likelihood decoding method is used to estimate the remaining information bits in the constituent code, and the final decoding results are obtained. Simulation results show that the proposed algorithm can effectively reduce the decoding delay compared with the existing algorithms without affecting the error performance.