A soft input decoding method and a decoder for generalized concatenated (GC) codes. The GC codes are constructed from inner nested block codes, such as binary Bose-Chaudhuri-Hocquenghem, BCH, codes and outer codes, such as Reed-Solomon, RS, codes. In order to enable soft input decoding for the inner block codes, a sequential stack decoding algorithm is used. Ordinary stack decoding of binary block codes requires the complete trellis of the code. In one aspect, the present invention applies instead a representation of the block codes based on the trellises of supercodes in order to reduce the memory requirements for the representation of the inner codes. This enables an efficient hardware implementation. In another aspect, there is provided a soft input decoding method and device employing a sequential stack decoding algorithm in combination with list-of-two decoding which is particularly well suited for applications that require very low residual error rates.

An error correction method includes performing a first error correction operation, the first error correction operation including performing a syndrome check operation by calculating a syndrome matrix corresponding to a codeword based on a parity check matrix, performing a decoding operation for the codeword according to a result of the syndrome check operation, and iterating the decoding operation until the syndrome check operation is passed for a codeword acquired as the decoding operation is performed or an iteration count of the decoding operation reaches a threshold count; accumulating syndrome matrixes, which are calculated as the decoding operation is iterated, to an accumulation matrix; and performing a second error correction operation for a last codeword acquired through the iterating of the decoding operation for the codeword, based on the accumulation matrix, when the iteration count reaches the threshold count.

A decoding method, a memory storage device and a memory control circuit unit are provided. The decoding method includes: programming a first memory cell in a rewritable non-volatile memory module; reading the first memory cell based on a first hard-decision voltage level to obtain first hard-bit information and perform a hard-decoding process accordingly; if the hard-decoding process fails and the first memory cell belongs to a first type memory cell, reading the first memory cell based on a second hard-decision voltage level to obtain second hard-bit information and perform another hard-decoding process accordingly; if the hard-decoding process fails and the first memory cell belongs to a second type memory cell, reading the first memory cell based on multiple second soft-decision voltage level to obtain soft-bit information and perform soft-decoding process accordingly. Therefore, a balance can be maintained between a decoding speed and a decoding success rate.

Multi-dimensional parity checker (MDPC) systems and related methods are disclosed to check parity of data regions within external memories. In one embodiment, the MDPC system includes a control register and a parity checker. The parity checker receives data segments accessed from the data region. The parity checker generates and accumulates multi-dimensional parity bits for the data segments and subsequently compares accumulated bits to expected multi-dimensional parity bits to generate multi-dimensional error syndrome bits representing identified comparison errors. The parity checker also determines a syndrome state based upon the multi-dimensional syndrome bits and stores the syndrome state within the control register. The parity checker operates in different modes based upon different values stored in an operational mode field of the control register. The parity checker can operate in a real-time error correction mode to correct errors within the data region from subsequent random accesses by components within the integrated circuit.

Techniques are described for decoding a codeword. In one example, the techniques include obtaining a first message comprising reliability information corresponding to each bit in the first codeword, determining a plurality of least reliable bits in the first codeword, and generating a plurality of flipped messages by flipping one or more of the plurality of least reliable bits in the first codeword. A number of the plurality of least reliable bits is equal to a first parameter and a number of flipped bits in each of the plurality of flipped messages is less than or equal to a second parameter. The method further includes decoding one or more of the plurality of flipped messages using a hard decoder to generate one or more candidate codewords.

A method is proposed for decoding read bits including information bits from memory cells of a solid state drive. The method comprises determining a reliability indication indicative of a reliability of the read bits, and iterating the following sequence of steps: soft decoding the read bits based on said reliability indication in order to obtain said information bits, determining at least one among a time indication indicative of a time elapsed since a last writing of the memory cells and a temperature indication indicative of a temperature of the memory cells, and applying at least one among said time indication and said temperature indication to said reliability indication.

A corresponding solid state drive is also proposed.

A method is proposed for decoding read bits including information bits from memory cells of a solid state drive. The method comprises providing an indication of reliability of the read bits, and, based on said indication of reliability, iteratively soft decoding the read bits in order to obtain said information bits. Said soft decoding comprises, at each iteration of the soft decoding, if the current number of iterations has reached a predetermined number of iterations indicative of an admitted latency of the solid state drive, and if no information bits having an error rate below a predetermined error rate have been obtained, providing a further indication of reliability of the read bits, and iteratively soft decoding the read bits based on said further indication of reliability.

A corresponding solid state drive is also proposed.

A method is proposed for decoding bits stored in memory cells of a solid state drive. Each memory cell comprises a floating gate transistor adapted to store a bit pattern, among a plurality of possible bit patterns, when programmed at a threshold voltage associated with that bit pattern, each threshold voltage being variable over the memory cells thereby defining, for each bit pattern, a corresponding threshold voltage distribution. The bit pattern of each memory cell comprises first and second bits, and the solid state drive is suitable for reading the bit patterns based on fixed reference voltages, each one designed to discern between two respective adjacent threshold voltage distributions, and on additional reference voltages different from the fixed reference voltages. The solid state drive is capable of soft decoding the read bit patterns based on soft information. The method comprises:

reading the first and second bits of the memory cells based on the fixed reference voltages, to obtain read first bits and read second bits, and

soft decoding the read first bits, wherein the soft information exploited for soft decoding the read first bits are based on the read second bits.

A corresponding solid state drive is also proposed.

According to one embodiment, a decoding device comprises a converter configured to convert read data to first likelihood information by using a first conversion table, a decoder which decodes the first likelihood information, a controller which outputs a decoding result of the decoder when the decoder succeeds decoding, and a creator module which creates a second conversion table based on the decoding result when the decoder fails decoding. When the second conversion table is created, at least a part of the decoding result is converted to second likelihood information by using the second conversion table the second likelihood information is decoded.

A method of decoding polar codes based on belief propagation includes conventional belief propagation to decode the polar codes first; when a number of iterations exceeds a predefined upper limit and a cyclic redundancy check fails, the method selects log-likelihood ratio vectors of a plurality of R or L messages from a plurality of log-likelihood ratio vectors generated in each of the iterations and generates another set of log-likelihood ratio vectors (referred to as candidate vector group) to be used as initial values of the R or L messages for a subsequent belief propagation to perform belief propagation decoding iterations and cyclic redundancy check again. Whenever a decoding result passes the cyclic redundancy check, the method exits; otherwise, the method iterates the above procedure until a maximum number of candidate vector groups has been reached.