Silicon Motion, Inc.

Shiuan-Hao Kuo

Zhen-U Liu

Hung-Jen Huang

A memory controller for use in a data storage device is provided. The memory controller includes a variable-node circuit and a check-node circuit. The check-node circuit obtains a codeword difference from the variable-node circuit, and calculates a syndrome according to the codeword difference. The variable-node circuit includes a threshold-tracking circuit which is configured to track a threshold used by the variable-node circuit during a low-density parity check (LDPC) decoding process to determine whether the variable-node circuit has entered a trapping status. In response to determining that the variable-node circuit has entered the trapping status during the LDPC decoding process, the variable-node circuit switches a bit-flipping algorithm used by the variable-node circuit during the LDPC decoding process from a first flipping strategy to a post-processing flipping strategy to bring the variable-node circuit out of the trapping status. The first flipping strategy is different from the post-processing flipping strategy.

Memory controller and method for bit flipping of low-density parity-check codes

A method and apparatus for decoding a Low-Density Parity-Check (LDPC) code whereby the apparatus comprises an LDPC decoder comprising variable-node calculation circuitry and check-node calculation circuitry: the check-node calculation circuitry is arranged operably to perform a modulo 2 multiplication on a codeword and a parity check matrix to calculate a plurality of first syndromes in a first-stage state. The variable-node calculation circuitry is arranged operably to perform a bit flipping algorithm to generate variable nodes, and calculate soft bits for the variable nodes in a second-stage state. The check-node calculation circuitry is arranged to perform the modulo 2 multiplication on the variable nodes and the parity check matrix to calculate second syndromes in the second-stage. When second syndromes indicate that the previously generated variable nodes are incorrect, a third stage state is repeated until decoding succeeds or a total number of iterations exceeds a threshold.

Method and apparatus for decoding low-density parity-check (LDPC) code

The invention relates to an apparatus and a method for generating a Low-Density Parity-Check (LDPC) code. The apparatus includes: a LDPC encoder, a look-ahead circuitry and an exclusive-OR (XOR) calculation circuitry. The LDPC encoder is arranged operably to encode a front part of a user data using a 2-stage encoding algorithm with a parity check matrix to generate a first calculation result. The look-ahead circuitry is arranged operably to perform a dot product operation on a rear part of the user data and one of a plurality of feature rows corresponding to the parity check matrix to generate a second calculation result in each iteration. The XOR calculation circuitry is arranged operably to perform an XOR operation on the first calculation result and the second calculation result to generate a front part of the LDPC code.

Apparatus and method for generating low-density parity-check (LDPC) code

A memory controller for use in a data storage device is provided. A low-density parity check (LDPC) process performed by the memory controller includes an initial phase, a decoding phase, and an output phase. The memory controller includes a variable-node circuit and a check-node circuit. During the initial phase, the variable-node circuit performs the following steps: obtaining a channel value, that is read from a flash memory, from a channel-value memory; transmitting the channel value to the check-node circuit to calculate a syndrome; and in response to the syndrome not being 0, setting a value of a register corresponding to each entry of a plurality of entries in a variable-node memory, and entering the decoding phase.

Memory controller and method of accessing flash memory

A memory controller, for use in a data storage device, is provided. A low-density parity-check (LDPC) decoding procedure performed by the memory controller includes an initial phase, a decoding phase, and an output phase in sequence. The memory controller includes a memory-index control circuit and a decoder. The decoder includes a decoding pipeline to perform the decoding phase of the LDPC decoding procedure. After the data storage device is booted up, the decoder reads a plurality of first codewords from a variable-node memory using a first order via the memory-index control circuit for LDPC decoding. In response to the decoder determining that a specific codeword among the first codewords has decoding failure, the decoder is reset to read a plurality of second codewords from the variable-node memory using a second order via the memory-index control circuit for LDPC decoding. The first order is different from the second order.

Memory controller and method for controlling data in decoding pipeline

A method employed in a low-density parity-check code decoder includes: receiving a specific data portion of a first codeword; calculating a flipping function value of the specific data portion of the first codeword according to the specific data portion by using checking equations of a parity check matrix to calculate checking values of the specific data portion; and determining whether to flip the specific data portion of the first codeword by comparing the flipping function value with a flipping threshold which has been calculated based on a plurality of flipping function values of a plurality of previous data portions earlier than the specific data portion.

Method employed in LDPC decoder and the decoder

A method for generating an LDPC (low-density parity check) code with a required error floor, comprising: using a parity generation circuit to generate an LDPC code; using a detection circuit to detect the LDPC code according to a plurality of trapping set cores in a database and to generate at least one piece of trapping-set-core information; using a verification circuit to perform an important sampling simulation according to the LDPC code and each trapping-set-core information separately to obtain an estimated error floor for each trapping-set-core information; using the verification circuit to separately compare each of the estimated error floors with an expected error floor; and when all of the estimated error floors are lower than or equal to the expected error floor, using the verification circuit to output the LDPC code.

Method and apparatus for generating an LDPC code with a required error floor

A non-volatile (NV) memory accessing method using data protection with aid of look-ahead processing, and associated apparatus such as memory device, controller and encoding circuit thereof are provided. The NV memory accessing method may include: receiving a write command and data from a host device; obtaining at least one portion of data to be a plurality of messages, to generate a plurality of parity codes through look-ahead type encoding, wherein regarding a message: starting encoding a first partial message to generate a first encoded result; applying predetermined input response information to a second partial message to generate a second encoded result, and combining the first and the second encoded results to generate a first partial parity code; and starting encoding the message to generate a second partial parity code, and outputting the first and the second partial parity codes to generate a parity code; and writing into the NV memory.

Non-volatile memory accessing method using data protection with aid of look-ahead processing, and associated apparatus

A flash memory controller is configured to decode a codeword. During the decoding process, the flash memory can check the decoding status of each codeword segment in the codeword and skip the decoding of a codeword segment whose decoding status is passed, thereby saving time decoding and also improving decoding efficiency. Even though only a part of the codeword segments in the codeword have been successfully decoded in the decoding process at the previous time, the flash memory controller can replace the part of the codeword segments in the codeword with the correct results obtained previously, and then decoding the re-formed codeword again. Accordingly, the decoding accuracy can be increased and the burden on the subsequent decoding process or data recovery can be reduced.

Flash memory controller, storage device and reading method

A flash memory controller used to access a flash memory includes a read-only memory, a processor, and an error correction code unit. The read-only memory is used to store a code. The processor executes the code to control access to the flash memory. The error correction code unit includes a control module and a decoder. The control module respectively calculates a first correlation between innate bad-column information which records the location of innate bad columns that become damaged after the read-only memory being manufactured and a plurality of trapping sets of a plurality of preset LDPC (low-density parity check) codes and uses the preset LDPC code which has the lowest first correlation as a selected LDPC code. The decoder decodes read information obtained from the flash memory according to the selected LDPC code.

Flash memory controller, storage device and reading method thereof

A memory controller for use in a data storage device is provided. The memory controller includes a variable-node circuit and a check-node circuit. The check-node circuit is configured to obtain a codeword difference from the variable-node circuit, and calculate a syndrome according to the codeword difference. During each LDPC (low-density parity check) decoding iterative operation, the variable-node circuit executes the following steps: determining syndrome weight according to a channel value and the syndrome from the check-node circuit; obtaining a previous codeword generated by a previous LDPC decoding iterative operation; determining a flipping strategy of a bit-flipping algorithm for each codeword bit in the previous codeword according to the syndrome weight and a predetermined threshold, and flipping one or more codeword bits in the previous codeword according to the flipping strategy to generate an updated codeword; and subtracting the previous codeword from the updated codeword to generate the codeword difference.

An encoding method includes: processing a plurality of data blocks to generate a plurality of partial parity blocks, wherein the partial parity blocks includes a first portion and a second portion; using a first computing circuit to generate a first calculating result according to the second portion of the partial parity blocks; using the first calculating result to adjust the first portion of the partial parity blocks; performing circulant convolution operations upon the adjusted first portion to generate a first portion of parity blocks; and using a second computing circuit to generate a second portion of the parity blocks according to at least the first portion of parity blocks; wherein the first portion of the parity blocks and the second portion of the parity blocks serve as a plurality of parity blocks generated in response to encoding the data blocks.

Encoder, associated encoding method and flash memory controller utilizing divided partial parity blocks for circulant convolution calculations

The present invention provides an encoding circuit of a flash memory controller, wherein the encoding circuit includes an auxiliary data generating circuit and an encoder. In the operations of the encoding circuit, the auxiliary data generating circuit is configured to receive a plurality of data chunks to generate auxiliary data corresponding to the data chunks. The encoder is configured to encode the data blocks to generate parity codes according to a parity check matrix, and to use the auxiliary data to replace a portion of the parity codes to generate adjusted parity codes, wherein the data chunks and the adjusted parity codes are written into a flash.

Flash memory controller and encoding circuit and decoding circuit within flash memory controller

A memory controller for use in a data storage device is provided. A low-density parity check (LDPC) process performed by the memory controller includes an initial phase, a decoding phase, and an output phase. The memory controller includes a variable-node circuit and a check-node circuit. During each LDPC decoding iterative operation in the decoding phase: the check-node circuit obtains a codeword difference from the variable-node circuit, and calculates a syndrome according to the codeword difference. The variable-node circuit is configured to: determine a syndrome weight according to the syndrome from the check-node circuit; obtain a previous codeword from a variable-node memory without obtaining a channel value from a channel-value memory; perform bit-flipping on one or more codeword bits in the previous codeword according to the calculated syndrome weight to generate an updated codeword; and subtract the previous codeword from the updated codeword to obtain the codeword difference.

Growing community of inventors

Taipei, Taiwan

Shiuan-Hao Kuo