Research On Polar Codes Encoding And Decoding Scheme In Free Space Optical Communication

Free-space optical communication technology has become a research hotspot in the field of communication because of its high transmission rate,large communication capacity,no spectrum limitation and strong security and confidentiality.However,due to the influence of atmospheric attenuation and atmospheric turbulence,the transmitted optical signal will have light intensity scintillation,beam drift and beam expansion,etc.,which will reduce the reliability and stability of the communication system.Channel coding can effectively suppress the influence of atmospheric turbulence without changing the structure of the transmitter and receiver.Polar code,as a new type of channel coding technology,has been proved to be the only forward direction that can reach the Shannon limit.Error-correcting codes,which have great research value.The application of polar codes in free-space optical communication can effectively improve the performance of free-space optical communication systems.Based on this,this paper analyzes and studies the decoding technology of polar codes in free-space optical communication.In this paper,the influence of atmospheric attenuation and atmospheric turbulence on the atmospheric channel in free-space optical communication is firstly analyzed,and the statistical characteristics of atmospheric turbulence effects are introduced,and then a dual-gamma distribution channel model that approximates the atmospheric turbulence channel is presented,and its optical intensity is explored.Probability Density Distribution.Then the polarization principle,coding structure and commonly used decoding algorithms of polar codes are researched and analyzed in detail.On this basis,the two common decoding algorithms of polar codes,which have the advantages of high bit error rate performance and low delay,are further improved and optimized,and are applied to the research of free space optical communication to improve the free space.Bit error rate performance of optical communications.Therefore,the main research work of this paper is as follows:(1)Aiming at the problem that it is difficult to locate the error bits in polar code decoding caused by atmospheric turbulence under free space optical communication,an LSTM-SCFlips decoding method under free space optical communication is proposed.Firstly,one-hot precoding is performed on the log-likelihood ratio(LLR)information sequence of polar code serial cancellation(SC)decoding.Under different training steps,the log-likelihood ratio(LLR)of polar code is analyzed and learned.The characteristics of the information sequence,comprehensively consider the root mean square error and computational complexity of the neural prediction model,and select an appropriate training step size to further eliminate the phenomenon of over-fitting of the prediction results on the basis of improving the accuracy of the prediction results.The LSTM neural network model is used to locate the first error bit of SC decoding,and the single-bit or multi-bit inversion of the serial offset decoding algorithm is performed in sequence according to the error probability.(2)Aiming at the problem of poor confidence propagation decoding performance of polar codes caused by atmospheric turbulence in free space optical communication,a polar code DNN-NOMS(Deep Neural Networks-Normalized and Offset Min)in free space optical communication is proposed.-Sum)decoding method.First,the traditional polar code belief propagation decoding algorithm factor graph is converted into a tanner graph similar to the low density parity check(LDPC)code,and the tanner graph is expanded and converted into a deep neural network(DNN)graphic representation on the basis of,adding a normalization factor and an offset factor to the MS(Min-Sum)decoding method to give weights to the edges of the tanner graph,simplifying the calculation method of the log-likelihood ratio of polar codes,and by limiting the number of training parameters,Select the factor parameters under the condition of the minimum loss function,and train the decoding model to obtain the optimal normalization factor and offset factor.