Research On Rate-compatible LDPC Code In Underwater Acoustic Communications

Underwater acoustic communication is one of the important supporting technologies for marine environmental monitoring,ocean resource exploitation,and maritime security.Channel coding techniques that improve communication reliability are indispensable for underwater acoustic communication systems.Low Density Parity Check Code(LDPC)is a commonly used error correction code in underwater acoustic communication systems,and its decoding performance and computational complexity are important considerations.Additionally,the time-varying characteristics of the underwater acoustic communication channel require extra communication overhead for a code words of single rate.Therefore,this dissertation focuses on the design of rate compatible LDPC codes and low-complexity LDPC decoding methods for underwater acoustic communication.To address the impact of pulse noise in the underwater acoustic communication channel on decoding performance,the dissertation proposes a mathematical model based on puncturing to approximate the Log Likelihood Ratio(LLR)information.Simulation results demonstrate that the approximated LLR information can effectively suppress the impact of pulse noise on decoding and is also applicable to other types of mixed noise.In order to reduce the computational complexity of LDPC decoding,the dissertation combines serial decoding and early termination iteration methods to improve the Belief Propagation(BP)algorithm and designs a fast-converging LDPC decoding method with early termination iteration.In this method,serial decoding is combined with static and dynamic scheduling to select check nodes for updating,and a criterion for checking the iteration reliability is proposed based on the absolute mean of the total likelihood information during the iteration process.Simulation results show that the proposed decoding method combining serial decoding and early termination iteration can effectively reduce the number of iterations and the decoding latency.The underwater acoustic communication channel has obvious time-varying characteristics,and encoding with single rate can result in additional communication overhead when the channel conditions are good.The dissertation analyzes the effectiveness of implementing rate compatible LDPC codes by puncturing in underwater acoustic communication and proposes a puncturing algorithm based on recovery trees to achieve higher rates of LDPC codes.The method considers both the number of iterations required for node recovery and the probability of correct node recovery,which improves the performance of the punctured codes.Simulation results show that the proposed method effectively reduces the number of iterations required for punctured codeword recovery and improves the codeword performance under the condition of limited iterations.The effectiveness of the proposed early termination iterative serial decoding method,BP algorithm,and puncturing algorithm is verified through data analysis results from offshore experiments and shallow sea experiments.The dissertation studies the design of rate compatible LDPC codes and low-complexity LDPC decoding methods in underwater acoustic communication,which can support the application of rate compatible LDPC codes in underwater acoustic communication systems.