Research On Polar Code Rate Compatibility And Its Compilation Code Algorithm

Polar code is an effective error correction code that can achieve channel capacity in discrete memoryless symmetric channels,has received extensive attention and research at home and abroad since it was proposed in 2009.Compared with traditional coding schemes,polar code has low compile code complexity,simple principles and excellent performance.Therefore,it has become the channel coding scheme in 5G e MBB scenarios,and it also has a wide range of application scenarios in optical fiber communication.However,the performance of polar code is not good in the case of short and medium code lengths;moreover,as the code length of the source code construction is limited to an integer power of 2,the polar code need to be redundant by selective deletion techniques such as puncturing algorithm or shortening algorithms to change the code length to an unrestricted arbitrary length in order to achieve the purpose of adaptive code rate code length,that is,code rate compatibility.Therefore,this paper starts the research from two aspects of code rate compatibility and decoding of polar code,and the main innovations are as follows:（1）To address the problem of limited code length when encoding polar code,a system polar code shortening scheme based on polarization weight assistance is proposed.which combines the polarization weight construction algorithm with the shortening algorithm to satisfy the polar code,and further improves the performance of the polar code by systematising it.The results of the simulation analysis show that the performance of the shortening method proposed in this paper is significantly improved compared with the Quasi Uniform Puncturing algorithm for polar code,the Quasi Uniform Puncturing algorithm algorithm for system polar code,the shortened polar code based on Gaussian approximation-aided and the shortening algorithm for polar code under the Addictive White Gaussian Noise（AWGN）channel.When the bit error rate is 10^-4,the proposed shortening algorithm achieves a gain of about 0.48dB,0.35dB,0.28dB and 0.05dB over the Quasi Uniform Puncturing algorithm for polar code,the Quasi Uniform Puncturing algorithm for system polar code,the shortened polar code based on Gaussian approximation-aided and the shortening algorithm for polar code,respectively.（2）At short and medium code lengths,the decoding performance of polar code is easily affected by channel noise and incomplete channel polarisation,leading to false decoding.To improve the decoding performance,this paper proposes a multi-bit flipping-based Successive Cancellation List decoding scheme to improve the correctness of decoding under the same code length by flipping the incorrect decoding bits,thus reducing the error block rate of decoding.The Simulation results show that the proposed algorithm significantly outperforms the SC（Successive Cancellation,SC）decoding algorithm,SCL（Successive Cancellation List,SCL）decoding algorithm,SCF（Successive Cancellation Flip,SCF）decoding algorithm,CA-SCL（CRC-Aid SCL）decoding algorithm and SCLF（Successive Cancellation List Flip,SCLF）decoding algorithm in terms of error block rate performance.The proposed algorithm has the smallest error rate,with a gain of about 1.50 dB,0.90 dB,0.70 dB,0.65 dB and 0.60dB compared with SC,SCL,SCF,CA-SCL and SCLF decoding algorithms,respectively,when the error rate is 10^-3,the code length N is 512,the list width L is 4 and the number of flipped set elements T is 64.In addition,a multi-bit flipped Successive Cancellation decoding of polar code with precise metric values is proposed,which takes into account the log-likelihood ratio and the reliability of the channel.It uses a new metric to sort CS（Critical Set,CS）,and flips the elements in CS when the decoding needs to be flipped,so as to better improve the decoding performance of polar code.The simulation results show that when BLER is 10^-3,code length N is 1024,code rate R is 0.5,and T is 16,the proposed algorithm has about 0.47dB and 0.08dB gain improvement compared with SC decoding algorithm and SCF decoding algorithm,respectively.