# US 2019/0379405 A1 · Data storage device

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0379405 A1
- Document ID: US20190379405A1
- Status: Published application
- Kind: A1
- Application number: 16/549,930
- Filed: 2019-08-23
- Published: 2019-12-12
- Earliest claimed: 2015-09-25
- Assignees: SK hynix Inc.
- Applicants: SK hynix Inc.
- Inventors: Kyoung Lae CHO; Naveen KUMAR; Aman BHATIA; Yi-Min LIN; Chenrong XIONG; Fan ZHANG; Yu CAI; Abhiram PRABAHKAR
- Web page: https://patentgrep.com/patent/US20190379405A1

## CPC

- H03M 13/3707: {Adaptive decoding and hybrid decoding, e.g. decoding methods or techniques providing more than one decoding algorithm for one code}
- G06F 3/061: {Improving I/O performance}
- G06F 3/0619: {in relation to data integrity, e.g. data losses, bit errors}
- G06F 3/064: {Management of blocks}
- G06F 3/0659: {Command handling arrangements, e.g. command buffers, queues, command scheduling}
- G06F 3/0661: {Format or protocol conversion arrangements}
- G06F 3/0679: {Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]}
- G06F 11/1012: {using codes or arrangements adapted for a specific type of error (G06F11/1048 takes precedence)}
- G06F 11/1076: {Parity data used in redundant arrays of independent storages, e.g. in RAID systems}
- H03M 13/1108: {Hard decision decoding, e.g. bit flipping, modified or weighted bit flipping}
- H03M 13/1111: {Soft-decision decoding, e.g. by means of message passing or belief propagation algorithms}
- H03M: CODING; DECODING; CODE CONVERSION IN GENERAL (using fluidic means F15C4/00; optical analogue/digital converters G02F7/00; coding, decoding or code conversion, specially adapted for particular applications, see the relevant subclasses, e.g. G01D, G01R, G06F, G06T, G09G, G10L, G11B, G11C, H04B, H04L, H04M, H04N; ciphering or deciphering for cryptography or other purposes involving the need for secrecy G09C)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A data processing system includes a storage medium, and a controller including a data processing block, configured to receive data from a host, transmit the received data to the storage medium, read data from the storage medium in response to a read request from the host, and decode the read data by the data processing block according to multiple decoding modes. The data processing block includes a first decoder and a second decoder, and is configured to manage the first decoder and the second decoder to run the decoding for the read data, and activate a fast decoding having shorter latency than a normal decoding after a fast decoding condition is satisfied.

## Claims

### Claim 1

A data processing system comprising:
a storage medium, and
a controller including a data processing block, configured to receive data from a host, transmit the received data to the storage medium, read data from the storage medium in response to a read request from the host, and decode the read data by the data processing block according to multiple decoding modes,
wherein the data processing block includes a first decoder and a second decoder, and is configured to manage the first decoder and the second decoder to run the decoding for the read data, and activate a fast decoding having shorter latency than a normal decoding after a fast decoding condition is satisfied.

### Claim 2 (depends on 1)

The data processing system according to claim 1,
wherein the controller is coupled to the host through a network, and the network is a wireless network.

### Claim 3 (depends on 1)

The data processing system according to claim 1, wherein the fast decoding condition is satisfied when a result of a previous decoding includes at least one among whether the previous decoding is a failure, whether the number of error bits corrected in the previous decoding exceeds a threshold number and whether a decoding iteration count in the previous decoding exceeds a threshold iteration count.

### Claim 4 (depends on 3)

The data processing system according to claim 3, wherein the previous decoding is the normal decoding.

### Claim 5 (depends on 4)

The data processing system according to claim 4, wherein the read data includes data chunks, and
wherein the data processing block performs the previous decoding on a data chunk among the data chunks, and performs the fast decoding on at least one data chunk on which the normal decoding is not yet performed among the data chunks.

### Claim 6 (depends on 1)

The data processing system according to claim 1, further comprising a memory configured to store the decoded data and a main memory device configured to store the decoded data from the memory and to be transmitted to the host.

### Claim 7 (depends on 6)

The data processing system according to claim 6, wherein the memory is included in the controller and has a smaller capacity than the main memory device.

### Claim 8 (depends on 1)

The data processing system according to claim 1, wherein the multiple decoding modes includes at least three different decoding modes including the normal decoding, the fast decoding, and an additional decoding.

### Claim 9 (depends on 1)

The data processing system according to claim 1, wherein the first decoder performs the normal decoding on the read data before the fast decoding condition is satisfied.

### Claim 10 (depends on 1)

The data processing system according to claim 1, wherein the first decoder performs the decoding in a different decoding mode with the second decoder.

### Claim 11 (depends on 1)

The data processing system according to claim 1, wherein a capability of correcting error of the second decoder performing an additional decoding is greater than a capability of correcting error of the first decoder performing the normal decoding.

### Claim 12 (depends on 1)

The data processing system according to claim 1, wherein the first decoder performs the fast decoding on a second part of the read data in response to decoding information from a sequencer, after performing the normal decoding on a first part of the read data.

### Claim 13 (depends on 12)

The data processing system according to claim 12, wherein the sequencer is included in the data processing system and generates the decoding information according to a result of the normal decoding on the first part of the read data.

### Claim 14 (depends on 1)

The data processing system according to claim 1,
wherein the data processing block includes a core circuit for performing the decoding and an input/output circuit, and
wherein the read data is transmitted directly to the memory through the input/output circuit without passing the core circuit when the fast decoding condition is satisfied.

### Claim 15 (depends on 1)

The data processing system according to claim 1, wherein the decoded data is stored in the memory after being decoded by the first decoder and is transmitted to the second decoder when the second decoder performs an additional decoding.

### Claim 16 (depends on 1)

The data processing system according to claim 1, wherein a decoding mode of the first decoder is changed from the normal decoding to the fast decoding in response to a result of a previous decoding.

### Claim 17 (depends on 1)

The data processing system according to claim 1, wherein a decoding mode of the first decoder is changed from the normal decoding to the fast decoding when an inefficient decoding is expected.

### Claim 18 (depends on 17)

The data processing system according to claim 17, wherein the inefficient decoding includes at least one of a decoding having long execution time and a decoding ending in failure.

### Claim 19 (depends on 1)

The data processing system according to claim 1, wherein a decoding mode of the first decoder is changed from the normal decoding to the fast decoding in response to decoding information stored in a decoding information memory.

### Claim 20 (depends on 1)

The data processing system according to claim 1, wherein the storage medium includes a memory device, and the controller includes a decoder and a bit error rate (BER) predictor, coupled with the memory device, and
wherein the controller is configured to:
read data from a page in the memory device to generate a set of codewords;
start to decode the set of codewords using the decoder and generate decoder information therefrom, the decoder information including information generated before completing the decode operation in a current iteration including a number of codewords not successfully decoded in each past iteration and during a portion of the current iteration;
predict a bit error rate (BER) of the page before completing the decode operation in the current iteration in accordance with at least the decode information, using the BER predictor; and
evaluate the predicted BER and generate an evaluation result to determine whether to complete the decode operation in the current iteration, using the BER predictor, and when it is determined to complete the decode operation in the current iteration, further determine a type of decoding to be used in completing the decode operation.

### Claim 21 (depends on 20)

The data processing system according to claim 20, wherein the read of the data from the page in the memory device is performed according to select read settings, and the set of codewords are generated using an error control coding scheme.

### Claim 22 (depends on 20)

The data processing system according to claim 20, wherein the type of decoding is determined from among at least hard decoding and soft decoding.

### Claim 23 (depends on 20)

The data processing system according to claim 20, wherein the decoder information includes a product code.

### Claim 24 (depends on 20)

The data processing system according to claim 20, wherein the controller is further configured to predict the BER again, using the BER predictor, after completing the decode operation in the current iteration.

### Claim 25 (depends on 20)

The data processing system according to claim 20, wherein the evaluation result includes an indication of whether another complete or partial iteration is to be performed, the partial iteration not including the read operation.

### Claim 26 (depends on 20)

The data processing system according to claim 20, wherein the evaluation result includes an indication that another iteration is not to be performed.

### Claim 27 (depends on 20)

The data processing system according to claim 20, wherein the controller is further configured to adjust a read setting according to at least the predicted BER, and repeat the iteration when the evaluation result indicates that another complete iteration is to be performed.

### Claim 28 (depends on 25)

The data processing system according to claim 25, wherein the controller is further configured to perform another partial iteration, based on the evaluation result, to generate updated decoder information, update the BER prediction, and re-evaluate the updated BER prediction.

### Claim 29 (depends on 1)

The data processing system according to claim 1, wherein the controller includes:
a processor; and
the memory comprising computer-readable instructions associated with a low-density parity-check (LDPC) decoder, the instructions upon execution by the processor cause the system to at least:
initiate an iterative decoding of an LDPC codeword, wherein the iterative decoding is bounded by a maximum number of iterations;
determine that a current iteration of the iterative decoding corresponds to a current number of iterations that is equal to or greater than an iteration number threshold and is smaller than the maximum number of iterations;
compare, when the current number of iterations is equal to or greater than the iteration number threshold, a syndrome associated with the LDPC codeword at the current number of iterations and a checksum threshold, wherein the syndrome is non-zero; and
terminate the iterative decoding when the comparing indicates that the syndrome is equal to or greater than the checksum threshold, wherein the iterative decoding is terminated prior to reaching the maximum number of iterations and improves a latency performance of the memory based on the terminating prior to reaching the maximum number of iterations.

### Claim 30 (depends on 29)

The data processing system according to claim 29, wherein the comparison indicates that a weight of the syndrome associated with the LDPC codeword is equal to or greater than the checksum threshold, and wherein the iterative decoding is terminated at the current iteration.

### Claim 31 (depends on 29)

The data processing system according to claim 29, wherein comparing the syndrome and the checksum threshold comprises comparing a checksum of the codeword to the checksum threshold, wherein the checksum and the checksum threshold are compared only after the current number of iterations reaches the iteration number threshold.

### Claim 32 (depends on 29)

The data processing system according to claim 29, wherein the checksum threshold is associated with a change rate threshold, wherein comparing the syndrome and the checksum threshold comprises:
computing a rate of change to a weight of the syndrome from one or more previous iterations to the current iteration;
determining that the rate of change to the weight of the syndrome is lower than the change rate threshold; and
generating a decision to terminate the iterative decoding based on the rate of change being lower than the change rate threshold.

### Claim 33 (depends on 29)

The data processing system according to claim 29, wherein the iteration number threshold is predefined from the range of sixty to eighty percent of the maximum number of iterations, and wherein the rate of change to the syndrome is computed only after the current number of iterations reaches the iteration number threshold.

### Claim 34 (depends on 29)

The data processing system according to claim 29, wherein comparing the syndrome to the checksum threshold comprises:
computing a rate of change to weight of the syndrome from one or more previous iterations to the current iteration;
determining that the rate of change to the weight of the syndrome is constant or has not improved from the one or more previous iterations; and
generating a decision to terminate the iterative decoding based on the rate of change being lower than the change rate threshold.
