Relevant Research On Realization Of LDPC Codes And Polar Codes

With the further development of modern digital communication, LDPC codes have been adopted by many communication standards because of their excellent error correction capability. In2006, China Mobile Multimedia Broadcasting (CMMB) standard was developed and LDPC codes were adopted by it. In order to reduce the power consumption of the decoder, this dissertation presents a non-uniform quantization strategy based on layered min-sum decoding algorithm. This dissertation also presents the design of the decoder and the structure of all the modules in application for CMMB standard. The simulation result shows that the design scheme can not only ensure the necessary throughput, but also reduce about35%computation, so it can effectively reduce the power consumption of the decoder.Currently, there are two major focuses on the research of LDPC codes. One is the design of the decoder, the other is the code performance, especially the research on the error floor at high SNR region. In order to overcome the error floor, this dissertation presents a corresponding iterative decoding algorithm in aim at the LDPC-SPC product codes. The experiment results show that the product codes can effectively reduce the error floor in theory, and only have a slight increase in decoding complexity, so they provide an alternative for the research on reducing the error floor of LDPC codes.Polar codes are a new coding technique which were proposed by E.Arikan in2007. On the binary discrete memoryless channel conditions, polar codes are proved to reach Shannon limit in theory. Recently, polar codes have become one of the most favorable error-correcting codes because of their capacity to reach the Shannon limit and their low encoding and decoding complexity. This dissertation presents a simple hardware implementation of polar encoder based on partially-parallel input. This design can effectively solve the problem that the polar codes have high encoding complexity due to the large code length.At last, this dissertation discusses the decoding algorithm of the polar codes. The large code length required by practical applications leads to high decoding latency because the conventional successive cancellation (SC) decoding algorithm decodes bits serially. This dissertation presents an improved SC decoding algorithm which possesses almost two times reduction of decoding latency compared with the traditional decoding algorithm. The simulation results of this design show that the error correcting performance degradation is negligible.