In one implementation, the disclosure provides a system including a detector configured to generate an output of a first log-likelihood ratio for each bit in an input data stream. The system also includes at least one look-up table providing a mapping of the first log-likelihood ratio to a second log-likelihood ratio. The mapping between the first log-likelihood ratio and the second log-likelihood ratio is non-linear. The system also includes a decoder configured to generate an output data stream using the second log-likelihood ratio to generate a value for each bit in the input data stream.

Described herein is an error correction circuit that includes a syndrome check history manager configured to maintain a history of syndrome checks corresponding to one or more iterations of the iterative decoding scheme. The error correction circuit also includes a trapping set detector configured to compare a trapping set determination policy with the history of syndrome checks to determine whether the history of syndrome checks meets criteria of the trapping set determination policy, while error correction decoding is performed, and determine that a trapping set exists when the history of syndrome checks satisfies the trapping set determination policy. The trapping set determination policy is related to at least one of a change in a syndrome vector, a number of UCNs, and a change in the number of UCNs.

Devices and techniques for variable read throughput control in a storage device are described herein. Bits from can be received for a read that is one of several types assigned to reads. A low-density parity-check (LDPC) iteration maximum can be set based on the type. LDPC iterations can be performed up to the LDPC iteration maximum and a read failure signaled in response to the LDPC iterations reaching the LDPC iteration maximum.

The present disclosure relates to a method and an apparatus for decoding data packets in communication network. The method comprises receiving one or more data packets related to each of one or more data types; and decoding the one or more data packets using a parity check matrix associated with the corresponding data type, wherein the parity check matrix comprises a plurality of layers, arranged according to a combination of layers which is determined using a reinforcement model.

An improved layered decoding method for a low density parity check (LDPC) code and a device therefor are disclosed. Disclosed is the layered decoding method for an LDPC code, capable of determining whether decoding is successful by performing a syndrome check on each check node at every variable node update. In addition, the syndrome check can be performed by using reduced variable nodes, thereby reducing decoding power consumption and decoding time consumption.

A decoder circuit includes first and second decoders. The first decoder is a first type of decoder configured to receive data encoded with an error correction code and decode and eliminate errors from a first subset of codewords of the data. The second decoder is a second type of decoder configured receive the data encoded with the error correction code and decode and eliminate errors from a second subset of codewords of the data, different from the first subset of the codewords, without attempting to decode and eliminate errors from the first subset of the codewords.

A method and apparatus for decoding are disclosed. The method includes receiving a first Forward Error Correction (FEC) block of read values, starting a hard-decode process in which a number of check node failures is identified and, during the hard-decode process comparing the identified number of check node failures to a decode threshold. When the identified number of check node failures is not greater than the decode threshold the hard-decode process is continued. When the identified number of check node failures is greater than the decode threshold, the method includes: stopping the hard-decode process prior to completion of the hard-decode process; generating output indicating that additional reads are required; receiving one or more additional FEC blocks of read values, mapping the first FEC block of read values and the additional FEC blocks of read values into soft-input values; and performing a soft-decode process on the soft-input values.

Method and apparatus for decoding error correction code (ECC) code words. Reference voltages are used to extract a selected code word from a communication channel. The selected code word is processed by an ECC decoder, and an initial syndrome weight is determined indicative of unresolved parity errors. A coarse search operates to concurrently adjust, over a first succession of iterations, each of the reference voltages. A subsequent fine search operates, over a second succession of iterations, to individually adjust the reference voltages. Decoding and syndrome weight determination continues over each iteration until a minimum syndrome weight is obtained, after which a user data content of the code word is decoded. The coarse search may transition the decoder from a saturated operational region to a linear operational region. The decoder may be a low density parity check (LDPC) decoder.

A semiconductor memory apparatus including a data memory array, a parity memory array, a data read/write and correction part, a parity read/write part and a syndrome generating and decoding part is provided. The data read/write and correction part reads the data memory array and outputs a first application reading data. The parity read/write part reads the parity memory array and outputs a parity reading data. During a read cycle of an application data, the syndrome generating and decoding part generates a syndrome writing data according to the first application reading data, compares and decodes the syndrome writing data with the parity reading data to generate a verifying comparison data. In the same read cycle, the data read/write and correction part corrects the application data according to the verifying comparison data, and writes the corrected application data back to the data memory array and outputs a corresponding output data.

A decoding method for a low density parity check (LDPC) code includes: updating a first check node, among a plurality of check nodes, by receiving, by the first check node, a bit decision and an associated first reliability value from each of a subset of variable nodes including a first variable node among a plurality of variable nodes, calculating a syndrome value and a second reliability value of the first check node based on the received bit decisions and first reliability values, and outputting the calculated syndrome value and second reliability value of the first check node to a variable node of the plurality of variable nodes but not of the subset of variable nodes; and updating the first variable node by receiving, by the first variable node, a syndrome value and a second reliability value of a second check node among the plurality of check nodes, and updating the first reliability value of the first variable node based on the syndrome value and the second reliability value of the second check node.