Encoding And Decoding Algorithms Of Ldpc Codes

LDPC (Low Density parity Check) code, a linear block error correcting code, which has a sparse check matrix, has a excellent performance of closing to Shannon Limit and a good practical foreground in the stable information transmission. This makes this correction code widely noticed by academies and IT industry all over the world. But the traditional approaches of block encoding will destroy the sparse characteristics of the check matrix severely and will cause the increase of the complexity of encoding procedure. Furthermore, the limitation of decoding implement presents the requirement of low complexity encoding algorithm. In this paper, we focus on the research of the hardware implement of linear time LDPC encoding method, and how to reduce the complexity of decoding algorithm.Research in this paper is divided into two parts:Firstly, discussed an important problem in the application of LDPC codes, that is how to overcome the comparatively high decoding complexity, and proposed RU Algorithm, which performs encoding using the sparse property of LDPC codes check matrix. Based on this algorithm, designed a kind of hardware implementation of encoder, and deduced the corresponding structure and operation unit and the parameter that demonstrates the performance of encoder. Through the final encoder parameter, it can be concluded that LDPC encoder that performs pretreatment using RU algorithm is more flexible, has a short latency, need less resource and can finish encoding in linear time.Secondarily, LDPC decoding algorithm is introduced, which needs to approximate the tanh function used in this algorithm by looking up the database. In order to avoid the reduction of decoding throughput of this traditional algorithm, a complexity simplified decoding algorithm which based on the BCJR algorithm is elaborated, and some approximated methods of correcting function are presented, analyzed and compared with each other. In addition, a threshold-decoding algorithm with simplified iterative procedure is brought forward. In the end, there are computer based simulation and analysis of LDPC encoding algorithm. With the simulation results, we can draw the conclusion that the presented algorithm can reduce the decoding complexity efficiently.