Research On Decoding Algorithm Of Polar Code

Polar code is the channel code scheme which can achieve channel capacity in theory.It has been one of coding standards in 5G.However,the finite polarization code decoding performance is unsatisfactory because the channel polarization is not sufficient.For example,Successive Cancellation(SC)decoding algorithm has a high error rate.Belief Propagation(BP)decoding algorithm has high complexity because of more iterations.Successive Cancellation List(SCL)and SCL-Cyclic Redundancy Check(SCL-CRC)decoding algorithms are not optimal in path selection.Therefore,an improved decoding algorithm of polar code is proposed in this thesis.Furthermore,it is combined with cooperative communication to improve performance of the communication systems.The main contents of this thesis are as follows:(1)A three-dimensional figure is drawn by analyzing the influence of the code length,the delete probability and the critical value that affect channel polarization in the binary erasure channel(BEC)channel.(2)Compare the polar code and RM code performance in different channels.The results show that the information transfer rate of the polar code is approach to the channel capacity and the bit error rate is lower than the RM code when the code length is constant.(3)An improved decoding scheme is proposed in this thesis.That is to say,an adaptive search path method is introduced to solve the path selection problem in SCL and SCL-CRC decoding algorithms.Meanwhile,construct the inner code for the bit that is prone to error to improve the performance.Dividing the code length into two parts,one part uses the SCL-CRC decoding algorithm for adaptive search path,and the other part uses the SC decoding algorithm.The result shows that decoding performance is improved by 0.3dB compared with the SCL and improved by 0.8dB compared with the SC method in the AWGN channel and this decoding scheme can reduce the computational complexity.(4)Apply the polar codes into the cooperative communications and research the amplify-forwarding scheme and the decode-forwarding scheme,respectively.The simulation results show that the amplify-forwarding scheme achieves a coding gain of 5dB and the decode-forwarding scheme achieves coding gain of 5.9dB compared with the traditional cooperative system when the bit error rate is 10-3.