Research On Polar Codes And Its Application In NAND FLASH

The Polar code has a fixed coding structure,a low-complexity coding and decoding algorithm.It is the only coding method that can theoretically prove that can reach the Shannon tolerance.In the ten years Polar code has been proposed,and gradually become research hotspots in the field of error correction code.It has been widely used in digital communication systems,digital storage systems,etc.,and has been identified as a coding scheme for 5G control channel enhanced mobile broadband scenarios.Channel polarization is the key to the Polar code,which is represented by the combination and splitting of channels.The channel polarization causes the symmetric capacity of channel to change,according to the difference of capacity selects good channel to transmit free bit,and selects bad channel to transmit frozen bit.When the code length tends to infinity,the channel fully polarized codeword can reach the Shannon tolerance;when the code length is limited,the channel cannot be fully polarized,which may result in insufficient performance of the SC decoding algorithm of the Polar code.Therefore,the decoding algorithm needs to be optimized.In order to suppress the influence of channel noise and error propagation characteristics on the SC decoding algorithm,this paper proposes a retransmission bit-assisted Polar code encoding and decoding algorithm,which uses CRC check factor and retransmission bit to detect and correct error for SC decoding result.A CRC check is performed on each decoding result,and if it is not passed,retransmission is performed,thereby suppressing error propagation due to channel noise,and obtaining better decoding performance.The simulation results show that compared with the SC decoding algorithm,the RAP algorithm can obtain a performance gain of about 1.30 d B.On the other hand,with the increasing scale of data storage and data processing,the requirements for reliability,security,and power consumption of data storage devices are gradually increasing.NAND FALSH is widely used in digital storage because of its low cost,large storage capacity,and low power consumption.However,as NAND FALSH increases in storage density and increases the number of P/E cycles,its reliability will decrease.An error correction code is needed for error control to ensure the reliability of the stored data.Because the Polar code has excellent error correction capability,this paper proposes a NAND FALSH Polar code error control scheme based on metadata information.Firstly,from the perspective of joint source channel coding,the metadata information in NAND FLASH is fully used to participate in the Polar code coding,and a high-performance Polar code is constructed.Then,due to the limitation of the Polar code coding structure,the Punctured Polar code is introduced.Using Punctured Polar code to puncture the previous code,so that the code rate of the codeword after the puncturing matches the code rate of the NAND FLASH,and the NAND FLASH is error-controlled.The simulation results show that the scheme using the metadata information for encoding can achieve a performance gain of about 0.72 d B compared to the direct coding scheme.In order to further improve the performance of the Polar code,the retransmission bit-assisted Polar code encoding and decoding algorithm proposed in this paper is applied to the NAND FLASH Polar code error control scheme based on metadata information.The simulation results show that the metadata information is used.Compared with the direct codec scheme,the algorithm can achieve better error correction performance.