Intel Corporation

Santhosh K Vanaparthy

Ravi H Motwani

Poovaiah Manavattira Palangappa

Zion S Kwok

Aman Bhatia

Rainer Strobel

Debarnab Mitra

Justin Kang

Santosh Emmadi

Techniques and mechanisms for supporting error correction of data accessed in a non-volatile (NVM) memory. In an embodiment, a memory controller performs reads each of the same codeword from the NVM, wherein a first read is based on a first reference voltage, and a second read is based on a second reference voltage. Based on the first read and the second read, confidence values are assigned each to respective bits of the codeword. Error correction comprises applying a weights to respective bits of the codeword based on the confidence values. In other embodiments, a relatively low confidence value is assigned to codeword bits that have a different value for the first read and the second read.

Soft read for NVM that is subject to destructive reads

An embodiment of an electronic apparatus may comprise one or more substrates, and a decoder coupled to the one or more substrates, the decoder including logic to perform a first decode stage with a first fixed quantization width, and perform a second decode stage with a second fixed quantization width that is different from the first fixed quantization width. Other embodiments are disclosed and claimed.

Hybrid LDPC decoder with mixed precision components

A low-density parity-check (LDPC) decoder performs check node computations as N different segments of the check nodes which have connections only to a codeword segment of length C/N bits as well as check nodes that have connections across the entire codeword of length C. The decoder can include a controller or other compute hardware to decode the codeword, including to perform computations for separate segments of C/N bits of the codeword. The system can perform computations including adjustment of the decode computations based on an expected error rate for selected segments of the codeword.

Application of low-density parity-check codes with codeword segmentation

Techniques and mechanisms to provide one or more substrates, and logic coupled to the one or more substrates. In an embodiment, the logic is to generate a pseudo-random sequence of bits, and to permute one or more bits of binary unscrambled data. In another embodiment, the logic is further to generate scrambled data based on an exclusive-or operation between the pseudo-random sequence of bits and the permuted data.

Data scrambler for persistent memory

Examples relate to a Forward Error Correction (FEC) encoder, an FEC decoder, Passive Optical Network (PON) systems, an Optical Line Terminal (OLT), an Optical Networking Unit (ONU), and to corresponding methods and computer programs. A forward-error-correction (FEC) encoder that is suitable for generating FEC data for use with hard-decision input at a receiver and for use with soft-decision input at the receiver is configured to generate the FEC data based on payload bits using a Low-Density Parity-Check (LDPC) code. The generated FEC data is generated using a single LDPC code that is suitable for use with soft-decision input and hard-decision input at the receiver or using one of two LDPC codes that are suitable for soft-decision input and hard-decision input, respectively.

LDPC encoder and decoder for multi-mode higher speed passive optical networks

Methods and apparatuses for generating optimized LDPC codes are proposed. One of the methods is a method for generating an optimized LDPC code for an asymmetric transmis&#xac; sion channel. The method includes receiving an initial LDPC code for the asymmetric transmission channel. Further, the method includes performing a density evolution threshold optimization for the initial LDPC code in order to obtain the optimized LDPC code for the asymmetric transmission channel. A uniformly mixed symmetric channel density for the asymmetric transmission channel is used in the density evolution threshold optimization.

Methods and apparatuses for generating optimized LDPC codes

Examples include techniques to improve error correction using an exclusive OR (XOR) rebuild scheme that includes two uncorrectable codewords. Examples include generation of soft XOR codewords using bits of correctable codewords to rebuild a codeword read from a memory that has uncorrectable errors and adjust bit reliability information to generate a new codeword having correctable errors. Examples also include techniques to prevent mis-correction due to read reference voltage shifts using non-linear transformations.

Techniques to improve error correction using an XOR rebuild scheme of multiple codewords and prevent miscorrection from read reference voltage shifts

Systems, apparatuses and methods may provide for technology to receive a codeword containing an SC-LDPC code and conduct a min-sum decode of the SC-LDPC code based on a plurality of scaling factors. In an embodiment, the scaling factors are non-uniform across check nodes and multiple iterations of the min-sum decode.

Non-uniform iteration-dependent min-sum scaling factors for improved performance of spatially-coupled LDPC codes

Examples include techniques to use intrinsic information when implementing a bit-flipping algorithm. An error correction control (ECC) decoder uses the intrinsic information to decode a low density parity count (LDPC) codeword. The intrinsic information including bits of a copy of a received LDPC codeword are compared to bits for variable nodes during an iteration of the bit-flipping algorithm to aid a determination as whether one or more bits for the variable nodes are to be flipped.

Techniques to use intrinsic information for a bit-flipping error correction control decoder

Techniques and mechanisms for determining information which is to be used for performing bit error correction. In an embodiment, bit reliability information is calculated based on a first likelihood function, and also on data signals which each indicate a respective bit of a codeword. Values of the bit reliability information each indicate, for a respective bit of the codeword, a confidence that the bit is at a particular logic state. A syndrome vector is calculated based on the bit reliability information, and one of the first likelihood function or a second likelihood function is selected based on one or more bit errors which are indicated by the syndrome vector. The selected one of the first likelihood function or the second likelihood function is used to correct bit errors of the codeword. In another embodiment, the first likelihood function is a default likelihood function to be used for initial syndrome vector calculations.

Device, system and method for determining bit reliability information

Examples include techniques for improving low-density parity check decoder performance for a binary asymmetric channel in a multi-die scenario. Examples include logic for execution by circuitry to decode an encoded codeword of data received from a memory having a plurality of dies, bits of the encoded codeword stored across the plurality of dies, using predetermined log-likelihood ratios (LLRs) to produce a decoded codeword, return the decoded codeword when the decoded codeword is correct, and repeat the decoding using the predetermined LLRs when the decoded codeword is not correct, up to a first number of times when the decoded codeword is not correct. When a correct decoded codeword is not produced using predetermined LLRs, further logic may be executed to estimate the LLRs for a plurality of buckets of the plurality of dies, normalize magnitudes of the estimated LLRs, decode the encoded codeword using the normalized estimated LLRs to produce a decoded codeword, return the decoded codeword when the decoded codeword is correct, and repeat the decoding using the normalized estimated LLRs when the decoded codeword is not correct, up to a second number of times when the decoded codeword is not correct.

Die-wise residual bit error rate (RBER) estimation for memories

Technology for correcting memory read errors including a preprocessing majority logic decode based on a plurality of identity structures of a parity check matrix, before ECC decoding using the parity check matrix, to estimate a set of erased or punctured bits of a codeword.

Techniques for correcting data errors in memory devices

One embodiment provides a silent data corruption (SDC) mitigation circuitry. The SDC mitigation circuitry includes a comparator circuitry and an SDC mitigation logic. The comparator circuitry is to compare a successful decoded codeword and a corresponding received codeword, the successful decoded codeword having been deemed a success by an error correction circuitry. The SDC mitigation logic is to reject the successful decoded codeword if a distance between the corresponding received codeword and the successful decoded codeword is greater than or equal to a threshold.

Mitigating silent data corruption in error control coding

Provided are an apparatus, non-volatile memory storage device and method for detecting drift in in non-volatile memory. A determination is made as to whether bits to write have more of a first value than a second value. Each of the bits are flipped to another of the first or second value when the bits have more of the first value than the second value. Indication is made whether the bits were flipped or not flipped. Parity is calculated for the bits and the bits and the parity for the bits are written to a location in the non-volatile memory. The bits at the location in the non-volatile memory are read and each of the bits having the first value are flipped to the second value and each of the bits having the second value are flipped to the first value in response to indication that the bits were flipped.

Growing community of inventors

Santa Clara, CA, United States of America

Santhosh K Vanaparthy