极化码自适应信道译码算法

暨南大学信息科学技术学院/网络空间安全学院,广东广州510632

通信传输技术;极化码;快速译码;自适应译码;联合译码;时间性能增益

Adaptive channel decoding method for polar codes
YE Maolin,TAN Xiaoqing,XU Liqing,and LÜShanxiang

College of Information Science and Technology/College of Cyber Security, Jinan University, Guangzhou 510632, Guangdong Province, P. R. China

communication transmission technology; polar codes; Fast-SC decoding; adaptive decoding; joint decoding; time performance gain

DOI: 10.3724/SP.J.1249.2022.05521

备注

极化码是一种理论上可以达到香农极限的信道编码方法,且有编译码复杂度低的优点,是5G通讯中信道编码解决方案之一.串行抵消列表(successivecancellationlist,SCL)译码是极化码目前最常用的译码方法,但该方法的空间和时间复杂度都很高;快速串行抵消(fastsuccessivecancellation,Fast-SC)译码可有效降低译码复杂度,但存在可靠性不高的问题.为兼顾极化码译码效率和可靠性,提出一种自适应信道的预快速SCL(preFast-SCL)译码算法.该算法通过联合Fast-SC和SCL译码算法,在译码开始时先使用Fast-SC算法快速得到一组译码结果,并对其进行校验,校验通过则作为结果输出,不通过则再使用SCL译码保证可靠性.仿真结果表明,随着信道条件变好,preFast-SCL译码算法复杂度逐步降低.在加性高斯白噪声信道下,preFast-SCL与SCL译码算法可靠性基本相同,在信噪比为2.0dB时,preFast-SCL的译码复杂度比SCL降低了45%,且在更高的信噪比条件下,时间性能增益更佳.
The polar code is a channel coding technology which can reach the Shannon limit in theory and has the advantage of low coding and decoding complexity. It is now one of the channel coding solutions in 5G communication. The successive cancellation list (SCL) decoding algorithm is the most commonly used decoding method for polar codes, but it has high memory and time complexity. Fast succesive cancellation (Fast-SC) decoding algorithm can effectively reduce the decoding complexity, but it has the problem of low reliability. In order to take into account the decoding efficiency and reliability of polar codes, a preFast-SCL decoding algorithm for adaptive channels is proposed. The algorithm combines the advantages of Fast-SC and SCL decoding algorithms. At the beginning of decoding, the Fast-SC algorithm is used to quickly obtain a group of decoding results and verify them. If the verification is passed, it will be outputed as the result, but if not, SCL decoding will be used to ensure the reliability. Simulation results show that the complexity of preFast-SCL decoding algorithm decreases gradually with the improvement of channel conditions. In additive white Gaussian noise channel, the reliability of preFast-SCL is basically the same as that of SCL decoding algorithm. When the signal to noise ratio (SNR) is 2. 0 dB, the decoding complexity of preFast-SCL is reduced by 45% compared with SCL, and the time performance gain is better under the condition of higher SNR.
·