Research On The Application Of Polar Codes In Underwater Acoustic Communication

From the"Jiaolong","Deep Sea Warrior"to"Fighter"and from the tortuous development of the marine economy before Chinese Economic Reform and open up to the construction of the"21st-Century Maritime Silk Road",they all demonstrate that the significantly progress of China from a marine country step by step to a marine potestatem.The underwater communication technology on this road provides powerful technical support for its information industry construction,in which channel coding is considered as one of the most critical parts.With the development of Chinese marine economy and military industries,the demand for using underwater acoustic channels for data transmission continues to increase.However,the realization of technical indicators such as high speed and low bit error rate of underwater acoustic communication is limited by the harsh underwater channel environment,and new technological breakthroughs are urgently needed for this field.At the same time,with the formal commercial use of 5th-Generation(5G)technology,as the polar codes for the control channel coding scheme in the 3rd Generation Partnership Project(3GPP)standard for enhanced mobile broadband scenarios,with its unique advantages that it can reach Shannon's limit,it has gradually emerged in the field of channel coding.In this thesis,polar codes are used as the channel coding scheme to research its application in underwater acoustic communication.The main research work is as follows:1.Analyze the characteristics of the underwater acoustic channel and establish two types of channel model.This thesis comprehensively analyzes the propagation characteristics of sound waves in the ocean from the aspects of sound velocity in the ocean,noise environment,propagation loss,multipath effects,bandwidth and power,and uses it as a basis to establish two types of channel models:Based on the traditional ray acoustic theory,the first type of channel models with the three typical channel transfer functions are established according to the different distributions of sound velocity gradients.The second type is established based on random statistical distribution models,and Rayleigh fading channel models are selected to describe underwater the receivd envelope distribution of the multipath signal.Moreover,the necessary analysis of the established channel models is carried out to facilitate the follow-up simulation experiment.2.Use the improved Monte Carlo construction method to construct the polar codes of the two types(the number of 4)of underwater acoustic channel models.Aiming at the characteristic that the coding structure of the polar codes is closely related to the channel,which is different from other channel coding schemes,this thesis uses the Monte Carlo construction method to calculate the Bhattacharyya parameters of the polarized sub-channel with a large number of simulation times to complete the selection of the indexes of the information bits.Moreover,three improvements are proposed to the traditional Monte Carlo construction method,which effectively reduces the computational complexity and improves the accuracy of the construction.For different types of channels,this thesis makes a classification based on the prerequisites for calculating the Bhattacharyya parameters:For theN_p determined model based on the ray acoustic theory,the idea of calculating the Bhattacharyya parameters when the sender and the receiver have known the channel side information(CSI)is proposed;The underwater acoustic channel model established by random statistical distribution,the idea of calculating the Bhattacharyya parameters based on the distribution function of the fading coefficient when the sender and the receiver have known the channel distribution information(CDI)is proposed.Finally,the construction of the polar codes with higher accuracy in each channel is completed.In the follow-up simulation experiment,it is shown that the polar codes coding scheme constructed in this thesis has achieved well bit error rate performance.3.The receiver structure of polar codes combined with adaptive decision feedback equalization(DFE)for decoding is designed.Aiming at the interference caused by the obviously multipath effect of the underwater acoustic channel,this thesis adds an adaptive decision feedback equalizer on the receiver based on the polar codes as the channel coding,and selects the sigmoid variable step-size least mean square(SVSLMS)algorithm as the equalization algorithm.By comparing the symbol distribution before and after the equalization and the decoding performance,it can be found that the receiver structure provides a significant performance gain,which proves the effectiveness of the polar codes combined adaptive DFE receiver structure designed in this thesis.4.In each channel model,simulate the underwater acoustic communication system and analyze the experimental results.This thesis comprehensively integrates the error correction performance and complexity of the decoder,finally uses the cyclic redundancy check aided successive cancelation list(CA-SCL)decoder as the decoding scheme.In each channel model,the performances of the polar codes with different code lengths and decoding widths are simulated.At the same time,the performance of the polar codes is compared with low density parity check code(LDPC)with similar code length.According to the analysis of simulation results,this thesis draws the following conclusions:Polar codes with longer code length have a better error correction ability.Moreover,the CA-SCL decoder which has a larger decoding list performances better.The simulation of symbol distribution with and without decision feedback equalizer shows that it significantly improves the bit error rate performance.When the code length is similar,the architecture which combines the decision feedback equalizer and polar codes have a higher coding gain when compare with the scheme of the LDPC codes.This makes our architecture more competitive for underwater acoustic communication.