Micron Technology Incorporated

Storage Technology Corporation

Robert B Eisenhuth

Stephen P Van Aken

Sivagnanam Parthasarathy

Mustafa Nazmi Kaynak

Patrick Robert Khayat

Bruce A Liikanen

Gerald L Cadloni

John L Seabury

Jonathan Scott Parry

Larry Joseph Koudele

Nicholas Julian Richardson

Ka Leung Ling

Bruce A Liikenanen

Methods, systems, and devices for read threshold adjustment techniques for error recovery are described. A memory system may read a codeword from a memory array using one or more read thresholds. The memory system may increment one or more counters of the memory device based on reading the codeword. The one or more counters may indicate information related to how many bits of the codeword correspond to a particular logic value. The memory system may detect an error, such as an uncorrectable error, in the codeword based on reading the codeword. The memory system may adjust the one or more read thresholds based on the information indicated by the one or more counters and read the codeword using the adjusted read thresholds.

Read threshold adjustment techniques for non-binary memory cells

Methods, systems, and devices for read threshold adjustment techniques for memory are described. A memory device may read a codeword from a memory array of the memory device using a read threshold having a first value. The memory device may increment one or more counters of the memory device based on reading the codeword. The counter may indicate a quantity of bits of the codeword that correspond to a first logic value. The memory device may detect an error, such as an uncorrectable error, in the codeword based on reading the codeword. The memory device may adjust the read threshold from the first value to the second value based on the quantity of bits indicated by the counter. The memory device may read the codeword using the read threshold having the second value.

Read threshold adjustment techniques for memory

A digital system, components and method are configured with nonvolatile memory for storing digital data using codewords. The data is stored in the memory using multiple bits per memory cell of the memory. A code efficiency, for purposes of write operations and read operations relating to the memory, can be changed on a codeword to codeword basis based on input parameters. The code efficiency can change based on changing any one of the input parameters including bit density that is stored by the memory. Storing and reading fractional bit densities is described.

Advanced bitwise operations and apparatus in a multi-level system with nonvolatile memory

Methods, systems, and devices for error correction for internal read operations are described. In some memory systems, a memory device may perform an internal read operation, in which the memory device reads data internal to the memory device (e.g., without sending the data to a memory system controller). To detect and correct errors during an internal read operation, the memory device may use an error control circuit on a memory die. The error control circuit on the memory die may operate on the same codeword, including the same data and same parity bits, as an error control circuit at the memory system controller, effectively reusing the stored parity bits for host read operations and internal read operations. To reduce the decoding overhead at the memory device, the error control circuit on the memory die may support detecting fewer errors than the error control circuit at the memory system controller.

Error correction for internal read operations

A memory cell can have a state in a valley between adjacent data states. A determination can be made whether a state of a memory cell is in a valley between adjacent distributions of states associated with respective data states. A signal indicative of a data state of the memory cell and whether the state of the memory cell is in the valley can be transmitted.

Memory cell state in a valley between adjacent data states

A method for performing a copyback procedure is described. The method includes determining to move first encoded data from a first location in a memory die to a second location. In response to determining to move the first encoded data from the first location to the second location, a starting seed, which is associated with the first location, is combined with a destination seed, which is associated with the second location, to produce a combined seed. Based on the combined seed, the method determines a pseudorandom sequence based on the combined seed and a pseudorandom sequence table, wherein the pseudorandom sequence table maps seed values to pseudorandom sequences and the determined pseudorandom sequence maps to the combined seed in the pseudorandom sequence table. The method further combines the first encoded data with the pseudorandom sequence to produce second encoded data for storage in the second location.

Efficient scrambling and encoding for copyback procedures using precomputed values

A method for performing a copyback procedure is described. The method includes determining to move first encoded data from a first location in a memory die to a second location. In response to determining to move the first encoded data from the first location to the second location, a starting seed, which is associated with the first location, is combined with a destination seed, which is associated with the second location, to produce a combined seed. Based on the combined seed, the method generates a pseudorandom sequence and combines the first encoded data with the pseudorandom sequence to produce second encoded data. In this configuration, the first encoded data is scrambled based on the starting seed while, based on combining the first encoded data with the pseudorandom sequence, the second encoded data is scrambled based on the destination seed. Thereafter, the second encoded data is stored in the second location.

Efficient scrambling and encoding for copyback procedures in a memory subsystem

A digital system includes nonvolatile memory for storing digital data using codewords. The data is stored in the memory using multiple bits per memory cell of the memory. A code efficiency, for purposes of transferring read data and write data relating to the memory, can be changed on a codeword-to-codeword basis based on input parameters.

A memory device may include memory components to store data. The memory device may also include a processor that may decode a codeword associated with the data. The processor may receive the codeword and determine whether the codeword is independently decodable using a BCH decoder. The processor may then decode the codeword using the BCH decoder when the codeword is determined to be independently decodable using the BCH decoder. Otherwise, the processor may decode the codeword using a second decoder and the BCH decoder when the codeword is not determined to be independently decodable using the BCH decoder.

Progressive effort decoder architecture

A system uses multi-level encoding where each symbol of a plurality of symbols represents more than one bit of information in a user data symbol stream for transfer using a multilevel transmission channel. The user data symbols are represented in a digital bitwise form such that each symbol is presented as a plurality of bits and each bit is subject to a different probability of error. An error correction procedure is applied based on the different error probability that is associated with each bit in the plurality. The channel can be configured to support a mosaic tile structure, each tile containing a channel symbol such that a selected tile has a collective error probability that is different from other tiles. Customized coding can be applied to the tile structure to allocate a selected amount of error correction power to the selected tile based on an overall available correction power.

Growing community of inventors

Boulder, CO, United States of America

Robert B Eisenhuth