There are provided an error correction decoder and a memory system having the same. The error correction decoder includes a node processor for performing at least one iteration of an error correction decoding based on at least one parameter used for an iterative decoding, a reliability information generator for generating reliability information corresponding to a current iteration upon a determination that the error correction decoding corresponding to the current iteration has been unsuccessful, and a parameter adjuster for adjusting the at least one parameter upon a determination that the reliability information satisfies a predetermined condition, and controlling the node processor to perform a next iteration based on the adjusted.

Disclosed are devices, systems and methods for improving a quality of service of an adaptive soft decoder in a non-volatile memory device. An example method includes selecting, based on current operating conditions of the non-volatile memory device, a first decoder parameter set from an ordered plurality of decoder parameter sets, each decoder parameter set corresponding to a distinct operating condition of the non-volatile memory device and comprising parameters related to a soft decoding operation; performing, based on the first decoder parameter set, the soft decoding operation; upon a determination that the soft decoding operation has succeeded, reordering the ordered plurality of decoder parameter sets to place the first decoder parameter set at a start of the ordered plurality of decoder parameter sets, and otherwise, performing the soft decoding operation based on a second decoder parameter set selected from the ordered plurality of decoder parameter sets.

A method of generating soft decision detection parameters for a plurality of received signals. The method comprises defining a hard decision boundary and a plurality of quantisation intervals wherein each quantisation interval extends from the hard decision boundary by an interval distance, selecting a log likelihood value from a set of log likelihood values for each received signal based on the quantisation interval in which the received signal is detected, performing a soft decoding using a plurality of log likelihood values, adjusting the set of log likelihood values based on a result of the soft decoding, determining an error probability for a quantisation interval, comparing the error probability against a target error probability and adjusting the interval distance in order to obtain the target error probability.

A memory system includes a memory medium, a loop-buffer configured to store read data outputted from the memory medium in a first operation mode, a fake-command generator configured to generate a fake-command in a second operation mode, and an error correction code (ECC) decoder configured to perform an ECC decoding operation of the read data stored in the loop-buffer in response to the fake-command.

An example method for determining likelihood of erroneous data bits stored in memory cells may include sensing a first plurality of memory cells based on a first sense thresholds. Responsive to sensing the first plurality of cells, a first set of probabilistic information may be associated with the first plurality of memory cells. A second plurality of memory cells may be sensed based on a second sense threshold. Responsive to sensing the second plurality of memory cells, a second set of probabilistic information may be associated with the second plurality of memory cells. An error correction operation may be performed on the first and second pluralities of memory cells based, at least in part, on the first and second values.

Disclosed herein are memory devices, systems, and methods of encoding and decoding data. In one aspect, an encoded data chunk is received and segmented into data segments with similar features. Each segment can be decoded based on its features. Data can also be rearranged and partitioned so as to minimize an entropy score that is based on the size and entropy of the data partitions. The approach is capable of enhancing performance, reducing decoding latency, and reducing power consumption.

A storage system includes memory cells arranged in an array and a memory controller coupled to the memory cells for controlling operations of the memory cells. The memory controller is configured to perform a read operation in response to a read command from a host, perform a first soft decoding of data from the read operation using existing LLR (log likelihood ratio) values stored in the memory controller, update existing LLR values using LLR values from neighboring memory cells and existing weight coefficients that account for influence from the neighboring memory cells. The memory controller is also configured to perform a second soft decoding using the updated LLR values. If the second soft decoding is successful, the memory controller performs a recursive update of weight coefficients to reflect updated influence from neighboring memory cells and stores the updated weight coefficient in the memory controller for use in further decoding.

Techniques related to a QoS-aware decoder architecture for data storage are described. In an example, QoS specifications include a QOS latency specification indicative of an acceptable latency for completing the processing of a data read command. The decoder may store this QOS latency specification. In operation, the decoder generates a latency measurement indicative of the actual latency for the processing. If a comparison of the latency measurement and QOS latency specification indicates a violation of the QOS latency specification, the decoder can terminate the decoding and generate a decoding failure.

Techniques related to improving a performance related to at least data reads from a memory are described. In an example, data is stored in a block of the memory as codewords. A data read includes a determination of whether each bit from a portion of the block is a zero or a one based on voltage measurements. Prior to decoding the codewords by performing a decoding procedure by an ECC decoder of the memory, a first number of errors E.sub.01 and a second number of errors E.sub.10 are estimated, where the first number of errors E.sub.01 is associated with bits each being a true zero and erroneously determined as a one, and where the second number of errors E.sub.10 associated with bits each being a true one and erroneously determined as a zero. Thereafter, the decoding of the codewords based on the decoding procedure is performed.

There is provided a method of providing adjusted LLR values of a plurality of bits in a codeword to an LDPC decoder, the plurality of bits representing a plurality of charge states of a plurality of memory cells of a non-volatile memory. The method comprises storing in a non-volatile memory controller associated with the non-volatile memory LLR values of the plurality of bits. The controller then determines a plurality of levels of the charge states represented by the plurality of bits. The controller then generates, by a distribution processor, distributions of a population of the plurality of bits in the codeword at each of the plurality of levels at a first and a second time after the first time. The controller then generates the adjusted LLR values based on a comparison between the first and second distributions, and then decodes the codeword according to the adjusted LLR values.