Research On LDPC Code Scheme For Ultra-reliability And Low Latency Communication

The 5G(5th Generation of Mobile Communication System)Ultra-reliable and Low Latency Communication(URLLC)covers a series of future-oriented applications such as automated driving,telemedicine,tactile internet,smart factory,etc,and thus has broad application prospects.The ultra-high reliability requirement of URLLC makes channel coding technology one of its key technologies.Low Density Parity Check Codes(LDPC)have certain advantages in performance and delay compared with other channel coding scheme,therefore it is considered to be competitive channel coding schemes in URLLC.However,the problem of error floor of LDPC code will cause the decrease of system reliability,and the low delay requirement must ensure that the decoding algorithm has lower complexity.Therefore,based on the application of the LDPC code scheme to the URLLC scenario,this thesis mainly solves two major problems: one is that the LDPC code has the problem of error floor;the other is the increased decoding complexity caused by the measure of eliminating the LDPC code error floor.The specific research contents are as follows:1.Aiming at the problem of error floor of LDPC code,firstly,a polar code which can be commonly used for different channel environment is constructed,which achieves the purpose of reducing the coding complexity of the polar code in this thesis.Then,a concatenated coding scheme based on polar and LDPC code is proposed.The scheme utilizes the characteristics that the polar code has no error floor,eliminating the error floor of the LDPC code,and further improves the overall decoding performance.2.Aiming at the problem of increased decoding complexity due to the introduction of polar code.Firstly,in this thesis,an adaptive decoding algorithm is proposed for the LDPC code.The algorithm solves the problem that the correction factor does not exactly match the too large part of the check node information in the existing decoding algorithm,which reduces the number of decoding iterations and has better performance,and thus reduces the decoding complexity.Secondly,an improved belief propagation(BP)algorithm is improved for polar code.This algorithm effectively reduces the complexity of decoding of polar code by omitting the update process of a large number of frozen nodes in the polar code factor graph.Finally,combining the LDPC code and the polar code improved decoding algorithm,a Low Complexity Merged Decoding(LCMD)algorithm is proposed based on the Merged Decoding(MD)structure.The LCMD algorithm can make full use of the decoding information and avoid the meaningless iterative process,which reduces the decoding complexity while ensuring the decoding performance.3.To solve the problem of increased complexity which caused by the inner and outer code rates are incompatible in the LCMD algorithm,the overall rate design scheme is given in this thesis and the performance simulation is carried out.The simulation results show that the proposed concatenated code scheme has lower overall complexity,and can achieve 0.2~0.4dB performance improvement without error floor compared with candidate LDPC codes.Therefore,the proposed scheme is more suitable for the URLLC scenario.