Overlap In The Linear Block Code Encoding Iterative Decoding Technology

Linear block code is one of the most important class of codes for channel error control coding and the basic principle is based on the foundation of algebra group which makes linear block code have strict algebra structure. Comparing to convolution code, the parity-check bits of a linear block code are only related to the information bits of contained in the same block, and the relation between information bits and parity-check bits is described by linear superpositions, or by the the parity-check matrix and the generation matrix. Both the two matrices drop into the critical properties of linear block code, sucha as the structure, the error control ability and the decoding architecture.This thesis presents a new coding-decoding scheme called the overlap coding and iterative decoding for hard decision decoding of linear block code. The new resulted code can be regard as a complex concatenated code. Two component codewords have a fraction of same information bits as overlapping. Decoding uses the same information bits as iteration media to pass the correct decoding massage to the individual decoder for the two received sequences. And this procedures would increase the overall decoding performance.RS(Reed-Solomon) code is a strong kind of M-ary linear block code for correcting both burst and random errors and has been widely applied in disk array,wireless communication,DVB and deep-space communication fields after Irving Reed and Gus Solomon proposed the code in 1960. In the third chapter of this thesis, we at first briefly introduce the basic operation rules in finite field and the typical decoding algorithm then we in detailed analyse the BM decoding algorithm and the Chase decoding algorithm for soft-decision. In chapter four, we apply the overlap coding and iterative decoding method to the coding and decoding of RS code and present comprehensively our simulation results for the BM hard-decision decoding and the Chase soft-decision decoding. The simulation results show that the overlap coding and iterative decoding method increase the performance of RS code.LDPC (Low Density Parity Check) code is another notable linear block code. It has the same performance of closing to the Shannon limit as Turbo code. There is a linear relationship between the LDPC decoder's complexity and the code length, and LDPC code can get ten times the performance of Turbo code. In the fourth chapter of this thesis, we at first introduce the coding mode of LDPC code based on the WiMAX standard, which is used in simulation, then we in detailed introduce the BP(Belief Propagation) decoding algorithm, which is the most widely used nowadays; In the sixth chapter, we apply the overlap coding and iterative decoding method to the coding and decoding of LDPC code, the simulation results show that the overlap coding and iterative decoding methed can also increase the performance of LDPC code.