The Research On Encoding And Decoding Algorithms Of LDPC Codes And Application In LTE-A System

3GPP LTE get widespread concern of mainstream communications equipment manufacturers and operators and gradually began large-scale commercial.3GPP began LTE-Advanced as a smooth evolution of LTE in order to meet the higher demand for the ITU IMT-Advanced (4G) and the future of communications.1Gbps data rate of LTE-Advanced puts forward new demands on the original channel coding techniques. In addition to Turbo codes, many companies are also considering other encoding, such as low density parity check (of LDPC Low Density Parity Check) code. LDPC codes is known as the code closest to the Shannon limit. Compared to Turbo codes, LDPC has advantages of higher coding gain in the case of large amount of data, a slight decrease in decoding complexity and easier to parallel realization. The LTE-A study group submitted a proposal of the LDPC codes to3GPP, and the subject is still in the research stage.This article researches encoding and decoding technology of LDPC codes, and codes LDPC codes channel coding of different modulation schemes and channel model. The paper also simulated and compared the factors affecting the performance of LDPC codes. According to the simulation results, combined with the LTE-A system needs, the paper selects the optimal quasi-cyclic LDPC codes and the min-sum decoding algorithm to build the LTE-A system simulation down link.The major research work includes:(1) The paper codes and compares different LDPC codes including three kinds of encoding algorithm and four decoding algorithms. Three kinds of encoding algorithm are the traditional encoding, RU encoding and quasi-cyclic LDPC encoding. Four decoding algorithms are the bit flipping decoding algorithm, the probability of BP (Belief Propagation) decoding algorithm, logarithmic domain BP decoding algorithm and the min-mum decoding algorithm. And do some improvement of the min-sum decoding algorithm to make it more adapted to the needs of the LTE-A link.(2) The paper does simulation and comparison of the factors. The influencing factors include:the modulation type, channel type, encoding and decoding algorithms, code length, and decoding iterations. MATLAB simulates these factors and finds that the using of quasi-cyclic LDPC coding and the min-mum decoding algorithm can get good performance of the LDPC codes channel coding gain.(3) The article builds a LTE-A system simulation downlink using LDPC channel coding technique according to the protocol standards for LTE-A system. The downlink combined the LDPC codes and the OFDMA, MIMO technology. The article simulates the bit error rate curves without channel coding, with Turbo coding and with LDPC-coding, obtains the conclusion that the performance of the LDPC codes is better than Turbo code and can meet the1Gbps channel coding requirements of the LTE-A system.