Research On Efficient LDPC Channel Coding And Decoding Technology For Air-ground Communication

Air-ground communication plays an important role in both military and civilian fields.Channel coding is one of the key technologies in ground communication system,it can effectively improve the anti-interference ability of the system.Further,in order to meet the requirements of reliability and timeliness of ground communication.Based on the theory of LDPC encoding and decoding,this thesis studies the efficient encoding and decoding technology of LDPC channel for ground communication.In coding,for the traditional LDPC coding algorithm complexity and code length code rate inflexibility problems,this thesis selects the 5G NR standard under the LDPC code type to complete the construction of the check matrix.LDPC code check matrix under 5G NR standard has a quasi-cyclic structure,which only needs to store the basic map in practical applications,reducing the coding complexity,and making the design of code length and rate more flexible due to more options of expansion factors.Furthermore,the LDPC coding algorithm based on 5G NR standard only performs block processing on the parity matrix without performing elementary transformation and other operations,thus preserving the sparse characteristics of the parity matrix and ensuring low coding complexity.In decoding,to address the problems of the traditional LDPC decoding algorithm with constant node update order and single normalization factor,this thesis proposes a layered normalized min-sum decoding algorithm based on the joint optimization of residuals and degree distribution.The algorithm combines the dynamic ordering strategy with the LNMS algorithm,and adjusts the layer order of the check matrix by calculating the residual value of each edge of the check node,so that the layers with high contribution to the decoding can be decoded first,which improves the convergence speed of the decoding algorithm.Furthermore,the check nodes are grouped according to the degree distribution of LDPC codes under the 5G NR standard,and the decoding error is corrected using multiple normalization factors to make the performance more close to the confidence propagation value.It can be verified through simulation that the coding gain of the DR-LNMS algorithm improves about 0.3 d B over the LNMS algorithm at a BER of 10^-5,at the same time,has a faster convergence speed.