Research On LDPC Code Encoding And Decoding For NAND Flash Memory

With the rapid development of electronic technology,storage technology is constantly updated.In recent years,NAND flash memory has been widely used in electronic products due to its characteristics of high performance and low power consumption.However,with the progress of NAND flash technology and the application of MLC and TLC storage technology,the internal storage density of flash memory increases greatly,leading to a sharp increase in bit error rate.Therefore,how to ensure the correct read storage data,improve the reliability of flash memory,reduce bit error rate has important significance.Traditional error correcting codes have not been able to meet the demand of high-density flash memory.Low-density Parity Check(LDPC)codes are gradually applied in NAND flash memory because of their excellent performance,strong error correcting ability and fast decoding speed.This paper constructs(18432,16640)high bit rate quasi-cyclic LDPC code according to the demand of large flash memory capacity and low cost,constructs its check matrix with PEG construction method,and generates a quasi-cyclic check matrix,which size of each cycle sub-matrix is 256*256.In terms of coding,this paper proposes an improved fast coding algorithm to encode the code word and reduce the amount of computation to improve the coding speed.This coding algorithm can effectively reduce the consumption of hardware resources with being applied to NAND flash memory.In terms of decoding,this paper combines the normalized minimum sum algorithm with the hierarchical information transmission structure,and proposes a hierarchical row minimum sum algorithm for decoding.Simulation results show that the proposed algorithm has better performance of decoding error correction,faster convergence speed and significantly reduced average iteration times of decoding,thus improving the throughput of decoder circuit.According to the above mentioned encoding and decoding algorithms,this paper designs the corresponding encoder and decoder to implement the encoding algorithm and decoding algorithm.Matlab software and Model Sim software are adopted to constitute the software simulation environment of this paper,and the LDPC code encoder and decoder are simulated.The simulation results prove that the functions of the encoder and decoder are correct and have excellent performance.