Skhynix Inc.

Korea Advanced Institute of Science and Technology

Soon Young Kang

Dae Sung Kim

Myung Jin Jo

Wan Je Sung

Bo Seok Jeong

Dong Hyun Kim

Kwan Su Shon

Yo Han Jeong

Jang Seob Kim

Eun Ji Choi

Sang Ho Yun

Tae Jin Hwang

Hyun Bae Lee

Jae Young Han

Woo Jin Kang

Jae Kyun Moon

The present technology provides a controller for controlling a memory device comprising: a hard syndrome calculator configured to generate a hard syndrome of a hard data chunk received from the memory device; a delta syndrome calculator configured to generate a delta syndrome of a delta bit data received from the memory device, the delta bit data indicating a reliability of the hard data chunk; a soft syndrome generator configured to generate a soft syndrome of the hard syndrome and the delta syndrome; a data determinator configured to select, as hard decision data, one of the hard data chunk and a soft data chunk, the selected data chunk corresponding to a syndrome having a lowest syndrome weight; and an error corrector configured to perform an ECC decoding operation on the hard decision data.

Electronic device having ECC decoder

A data processing device includes a plurality of variable nodes configured to receive and store a plurality of target bits; a plurality of check nodes each configured to receive stored target bits from one or more corresponding variable nodes of the plurality of variable nodes, check whether received target bits have an error bit, and transmit a check result to the corresponding variable nodes; and a group state value manager configured to determine group state values of variable node groups into which the plurality of variable nodes are grouped.

Data processing device

Provided herein may be an error correction decoder, an error correction circuit having the error correction decoder, and a method of operating the error correction decoder. The error correction decoder may include a calculator configured to output an error correction message by performing an iterative decoding operation on a first codeword, a syndrome generator configured to generate a syndrome by calculating the error correction message and a parity check matrix and to output a number of iterations representing the number of times the iterative decoding operation has been performed, and an unsatisfied check node (UCN) value representing the number of unsatisfied check nodes in the syndrome, and a speed selector configured to output a speed code for controlling a speed of the iterative decoding operation depending on the number of iterations and the UCN value.

Error correction decoder, error correction circuit having the same, and method of operating the same

Devices for using a neural network to choose an optimal error correction algorithm are disclosed. An example device includes a decoding controller inputting at least one of the number of primary unsatisfied check nodes (UCNs), the number of UCNs respectively corresponding to at least one iteration, and the number of correction bits respectively corresponding to the at least one iteration to a trained artificial neural network, and selecting any one of a first error correction decoding algorithm and a second error correction decoding algorithm based on an output of the trained artificial neural network corresponding to the input, and an error correction decoder performing error correction decoding on a read vector using the selected error correction decoding algorithm. The output of the trained artificial neural network may include a first predicted value indicating a possibility that a first error correction decoding using the first error correction decoding algorithm is successful.

Electronic device and method of operating the same

The controller that controls a memory device includes: a processor suitable for controlling the memory device to perform a first soft read operation by using first soft read voltages; and an error correction code (ECC) codec suitable for performing a first soft decision decoding operation based on first soft read data obtained through the first soft read operation, wherein the processor controls the memory device to perform a second soft read operation with an additional read voltage, of second soft read voltages, that is different than any of the first soft read voltages and which is determined based on the first soft read data, according to whether the first soft decision decoding operation failed, and wherein the ECC codec performs a second soft decision decoding operation based on the first soft read data and second soft read data obtained through the second soft read operation.

Controller and operating method thereof

The present technology includes an electronic device and a method of operating the same using an artificial neural network. The electronic device according to the present technology includes a decoding controller inputting a primary syndrome vector generated based on a read vector and a parity check matrix to a trained artificial neural, and selecting any one of a first error correction decoding algorithm and a second error correction decoding algorithm based on an output of the trained artificial neural network corresponding to the input, and an error correction decoder performing error correction decoding on a read vector using the selected error correction decoding algorithm. The output of the trained artificial neural network includes a first predicted value indicating a probability that a first error correction decoding using the first error correction decoding algorithm is successful.

An operating method of a low density parity check (LDPC) decoder, the operating method includes: initially updating codewords to variable nodes; determining an update order in which a plurality of variable node groups are updated, which is determined based on reliability of each of the variable node groups; executing local iterations including update of check nodes associated with a select variable node group among the variable node groups and update of the select variable node group based on the updated check nodes until all the variable node groups are updated based on the update order; performing syndrome check to determine whether LDPC decoding is successful, based on an operation of the updated variable nodes and a parity check matrix.

LDPC decoder and operating method thereof

Provided herein is an error correction decoder and a memory system having the same. The error correction decoder includes a memory configured to store a hard decision value of a variable node. The decoder further includes a flipping function value generator configured to generate, in an i-th iteration, a first value based on a number of unsatisfied check nodes (UCNs) corresponding to the variable node, and to generate a flipping function value as (i) a difference between the first value and an offset value or (ii) a set value, wherein i is a non-negative integer. The decoder also includes a comparator configured to output, in the i-th iteration, a first signal indicating whether to flip or not flip the hard decision value of the variable node in the memory based on comparing the flipping function value to a flipping threshold value.

Error correction decoder and memory system having the same

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.

Error correction decoder and memory controller having the same

Provided herein may be an error correction circuit. An error correction circuit for performing error correction decoding based on an iterative decoding scheme using a NB-LDPC code may include a symbol configuration circuit for configuring an initial symbol to be assigned as a variable node value to a variable node, a reliability value initialization circuit for initializing first reliability values of candidate symbols corresponding to the variable node based on the initial symbol assigned to the variable node, and a symbol correction circuit updating the first reliability values of the candidate symbols based on communications received from a check node coupled to the variable node, the candidate symbols having updated first reliability values, respectively, and adjusting the variable node value to one of the candidate symbols based on a comparison with the updated first reliability value of one of the candidate symbols with a first threshold value.

Error correction circuit and operating method thereof

An error correction device includes: a plurality of variable node units each configured to: receive a hard decision bit and a channel reliability value having a first bit-precision; and perform an iteration of a decoding operation on the hard decision bit based on the channel reliability value; a plurality of check node units each configured to: receive one or more reference reliability values having a second bit-precision from one or more variable node units coupled thereto among the plurality of variable node units during the iteration; and transmit, based on the one or more reference reliability values, one or more check reliability values having the second bit-precision to the one or more variable node units coupled thereto, wherein, during the iteration, each of the plurality of variable node units further: receives one or more first check reliability values from one or more check node units coupled thereto among the plurality of check node units; and updates the hard decision bit with reference to the channel reliability value and the one or more first check reliability values by upsizing the first bit-precision of the channel reliability value and the second bit-precision of the one or more first check reliability values.

Error correction device and electronic device including the same

An amplifier may include a differential pair circuit configured to generate an output signal according to a first input signal and a second input signal, a plurality of current sinks coupled between a ground terminal and the differential pair circuit, and a feedback circuit configured to sense a level of the output signal and generate a feedback signal. At least one of the plurality of current sinks is controlled according to the feedback signal.

Amplifier and semiconductor apparatus using the same

An operation method for a low density parity check (LDPC) decoder which includes performing an initial update operation to variable nodes by updating a codeword to the variable nodes, performing a decoding operation to the codeword based on an original parity check matrix, generating a modified parity check matrix by changing data of rows of the original parity check matrix corresponding to check nodes with dummy data, performing a node update operation based on the modified parity check matrix and performing a predetermined number of single iterations of the above processes until the decoding operation is successful.

LDPC decoder, semiconductor memory system and operating method thereof

An analog to digital converter includes a first DAC unit configured to vary a level of a reference voltage output through a first node according to a first code, a second DAC unit coupled in parallel to the first DAC unit on the basis of the first node and configured to vary the level of the reference voltage according to a second code, a comparator configured to generate a comparison result signal by comparing an input voltage and the reference voltage, and at least one register array configured to store the first code and the second code with initial values and store the first code and the second code by varying values of the first code and the second code according to the comparison result signal.

Growing community of inventors

Gyeonggi-do, South Korea

Soon Young Kang