Encoding And Decoding Optimization Of Polar Codes In Optical Fiber Communication Systems

As a wired optical communication technology,optical fiber communica-tion has become a key technology that supports modern network communi-cation and carries the development of the Internet,which can realize large-capacity,long-distance,and high-reliability communication transmission.With the development of the Internet,the amount of data transmitted and exchanged through the Internet has exploded.Improving the data transmission speed and communication reliability has become a major demand for optical fiber com-munication technology.Therefore,the application of channel coding in opti-cal fiber communication systems to obtain coding gains to support a greater transmission speed has become a research focus.In 2009,the polar code was proposed as a channel coding method that can be strictly proved to reach the channel capacity.It has clear coding structure,low decoding complexity,and robust error correction performance,arouses widespread attention and research in a short time.The polar code is closely related to the channel condition and has the characteristic of non-universality.The reliability of the polarization channel will change under different channel conditions,which affects the construction process of the polar code and its transmission performance.In optical fiber communication systems,research on optimizing the polar code encoding and decoding process according to the characteristics of optical fiber communica-tion systems to improve the transmission performance has important theoretical significance and application value.This paper takes the optical fiber communication system using polar codes encoding as the main research background,focuses on the encoding,decoding process as well as code rate selection method of the polar code,and proposes an optimization scheme of polar code encoding based on the neural network channel detection module which obtains the optical fiber channel information,so as to improve the accuracy of polarization channels" reliability estimation;an optimization scheme of the polar code decoding neural networks based on the acquisition of channel SNR information to expand the input information of networks to improve the decoding performance;an optimization scheme of the code rate of the polar code based on the adaptive code rate adjustment algo-rithm to improve the communication efficiency of the polar code.The specific research contents are as follows:(1)An optimization scheme of the polar code encoding by obtaining chan-nel information is proposed.In the optical communication system,the channel detection module constructed by the neural network is introduced to estimate the channel SNR value,and the estimated SNR value is used as the design-SNR when polar codes are encoded at the sending side to improve the accuracy of polarization channels' reliability estimation,so that the communication perfor-mance can be improved.The optimization scheme is applied to the CO-OFDM system,and the steps of modulation,demodulation and serial-to-parallel con-version of the CO-OFDM system are adjusted accordingly to make it more com-patible with the optimization scheme.Simulation results show that by obtaining the channel SNR value of each subcarrier and encoding polar codes differently,it can bring performance advantages.For polar codes with code length N=512 and code rate R=0.5,the coding gain is about 1.4dB when the BER is 10-4.(2)An optimization scheme of the polar code decoding neural networks is proposed.The scheme uses the neural network channel detection module to es-timate the electrical SNR of the optical fiber channel,calculates the LLR value of the received codeword,and uses the SNR information and the LLR value as the input information of the neural network decoder to make the network opti-mize the decoding process under different channel conditions.Constructed and trained neural network decoders based on different neural network models,and verified the performance of the optimized decoders in the optical fiber com-munication system.Simulation results show that the neural network decoders adopting the optimization scheme have performance advantages over the unop-timized neural network decoders.(3)An optimization scheme of the code rate of the polar code based on the adaptive code rate adjustment algorithm is proposed.The adaptive code rate adjustment algorithm recommends the code rate of the polar code according to the code length,the SNR of the channel and the importance of the transmission information,and uses the reception error probability to approximate the com-munication performance of the polar code to reduce the difficulty of algorithm application.A neural network is trained on the basis of the algorithm to re-duce the computational complexity of each code rate search.Simulation results show that the algorithm can recommend an appropriate code rate of polar codes according to the requirements of system communication reliability to improve the communication efficiency.