Research On LDPC Decoding Algorithm Based On Trapping Set Ontology

LDPC (Low Density Parity Check Code, LDPC) is a class of sparse check matrixof a linear block code proposed by Dr. Robert G. Gallager in1963. Because of its nearShannon limit performance, low decoding complexity and flexible structure, LDPC hasbecome a research hotspot in the field of channel coding in recent years. It has beenwidely used in deep space communication, optical fiber communication, satellite digitalvideo and audio broadcasting.This paper conducts a study of trapping set ontology. On the basis of the trappingset ontology theory, this paper gives a simpler and more effective new decoding methodthan BP decoding algorithm. The main work of this paper is summarized as follows:1. The course of development of modern coding theory has been systematicallysummarized. LDPC code Tanner graph model and its theoretical basis are brieflyilluminated. Message passing mechanism and belief propagation decoding algorithm areanalyzed in detail.2. The concept of trapping set and trapping set ontology are thoroughly analyzed.The application of trapping set ontology in the encoding and decoding is introduced.Based on trapping set ontology, a trapping set database generation algorithm is given.The database can be used to evaluate the size of the danger of a trapping set.3. A multi-bit decoding algorithm is design and realized. This algorithm adds somebits to the information passed through check nodes and variable nodes. These added bitscarry the local neighbor information of nodes, which can make decoder decodesuccessfully escape the trap. Simulation results show that the algorithm is simple andeffective.