| With the continuous development of communication technology,various emerging communication scenarios,such as the Internet of Things,industrial control systems,interactive telemedicine,continue to emerge.These applications have put forward higher requirements for communication reliability and latency.Channel coding is one of the key technologies for achieving the goal of low-latency and high-reliable communication.Low-Density Parity-Check(LDPC)codes are widely used in various communication systems due to the advantages in performance.To further improve the performance of LDPC codes,the thesis researches on LDPC codes.The main innovations and contributions are as follows:1.Aiming at the problem of poor error-correction performance of hard-decision decoding of LDPC codes,a transmission scheme based on Quasi-Cyclic Low-Density Parity-Check(QC-LDPC)codes with matrix Q is proposed.The relationship between elements of the flipping reference vector is strengthened,and then the bit error rate is decreased based on hard-decision.The proposed transmission scheme can achieve a performance gain of 1.56 d B compared to BF decoding algorithm.In addition,through introducing soft information,a reliability-based Q decoding algorithm is proposed,which can achieve a performance gain of 0.3～0.5d B compared to Q decoding algorithm in the proposed transmission scheme.2.To solve the problem of high complexity of Ordered Statistic Decoding(OSD)algorithm for LDPC codes,a Threshold-based and Set-Divided Ordered Statistic Decoding(TSDOSD)algorithm is proposed,which divides the bits in most reliable basis into several sets according to the reliability threshold,and flips bits in the most reliable basis with different probabilities.It can help to filter some Test Error Patterns(TEPs).Besides,a combined decoding algorithm of Belief Propagation(BP)and TSDOSD is proposed to improve error-correction performance while maintaining low complexity.Later,there is still room for improvement about the error-correction performance of loworder OSD algorithm.Cyclic Redundancy Check(CRC)is performed on candidate codeword estimation sequences,and a CRC-Aided Ordered Statistic Decoding(CA-OSD)algorithm is proposed.Based on CA-OSD algorithm,the maximum likelihood decoding is analyzed from the perspective of TEPs,and a CA-OSD algorithm with sorting TEPs is proposed,where the TEP with a higher probability in the CA-OSD algorithm is prioritized to participate in the recoding.It can achieve a performance gain of 0.24 d B compared to OSD algorithm.3.In response to the current situation that many emerging applications in lowlatency and high-reliable communication scenarios require the transmission of short data packets,and many decoding algorithms with superior performance for long code are not suitable for short codes,the thesis investigates Guessing Random Additive Noise Decoding(GRAND),which is commonly used for short length and high rate codewords.Besides,the principle and implementation process of Ordered Reliability Bits Guessing Random Additive Noise Decoding(ORBGRAND)algorithm for CRC codes and LDPC codes are studied.Moreover,a CRC-aided ordered reliability bits guessing random additive noise decoding algorithm is proposed for LDPC codes,which reduces the probability of undetectable errors caused by only utilizing the parity-check matrix for checking during codebook queries,thereby to improve the error-correction performance. |
