G11C16/3486

PRECISE DATA TUNING METHOD AND APPARATUS FOR ANALOG NEURAL MEMORY IN AN ARTIFICIAL NEURAL NETWORK
20220374699 · 2022-11-24 ·

Numerous examples of a precision programming apparatus are disclosed for precisely and quickly depositing the correct amount of charge on the floating gate of a non-volatile memory cell within a vector-by-matrix multiplication (VMM) array in an artificial neural network. In one example, a neuron output circuit for providing a current to program as a weight value in a selected memory cell in a vector-by-matrix multiplication array is disclosed, the neuron output circuit comprising a first adjustable current source to generate a scaled current in response to a neuron current to implement a positive weight, and a second adjustable current source to generate a scaled current in response to a neuron current to implement a negative weight.

Nonvolatile memory device and method for operating with varying programming time

A nonvolatile memory device is provided. A nonvolatile memory device comprises a word line, a bit line, a memory cell array including a first memory cell at an intersection region between the word line and the bit line, a word line voltage generating circuitry configured to generate a program voltage, the program voltage to be provided to the word line, a row decoder circuitry configured to receive the program voltage from the word line voltage generating circuitry and configured to provide the program voltage to the word line, a verification circuitry configured to generate a verification signal in response to verifying a success or a failure of programming of the first memory cell, and a control circuitry configured to apply the program voltage to the first memory cell in response to the verification signal, and configured to cut off the program voltage in response to the verification signal.

Precise data tuning method and apparatus for analog neural memory in an artificial neural network

Numerous embodiments of a precision programming algorithm and apparatus are disclosed for precisely and quickly depositing the correct amount of charge on the floating gate of a non-volatile memory cell within a vector-by-matrix multiplication (VMM) array in an artificial neural network. Selected cells thereby can be programmed with extreme precision to hold one of N different values.

Read level tracking and optimization

Systems and methods for read level tracking and optimization are described. Pages from a wordline of a flash memory device read and the raw page data read from the wordline may be buffered in a first set of buffers. The raw page data for each of the pages may be provided to a decoder for decoding and the decoded page data for each of the pages buffered in a second set of buffers. First bin identifiers may be identified for memory cells of the wordline based on the raw page data and second bin identifiers may be identified for the memory cells of the wordline based on the decoded page data. Cell-level statistics may be accumulated based on the first bin identifiers and the second bin identifiers, and a gradient may be determined for respective read levels based on decoding results for each of the pages and the cell-level statistics. Settings for the read levels may be configured in the flash memory device based on the determined gradients.

Semiconductor storage device configured to perform a detection operation after performing a program loop
11676672 · 2023-06-13 · ·

A semiconductor storage device includes memory cells a controller performing a write operation on the memory cells. The write operation includes program loops with a program operation and a verification operation. In a first loop the controller applies a first program voltage and a first verification voltage. Next, a detection operation counts the memory cells with a threshold voltage above a first threshold value. In a second program loop, after the detection operation, the controller applies a second program voltage and a second verification voltage. The values of used for second program voltage and the second verification voltage are set dependent on the counted number of memory cells with a threshold voltage above the first threshold value.

Programming of memory devices

Memory devices might include a controller configured to cause the memory device to apply a first plurality of incrementally increasing programming pulses to control gates of a particular plurality of memory cells selected for programming to respective intended data states, determine a first occurrence of a criterion being met, store a representation of a voltage level corresponding to a particular programming pulse in response to the first occurrence of the criterion being met, set a starting programming voltage for a second plurality of incrementally increasing programming pulses in response to the stored representation of the voltage level corresponding to the particular programming pulse, and apply the second plurality of incrementally increasing programming pulses to control gates of a different plurality of memory cells selected for programming to respective intended data states.

STORAGE DEVICE AND OPERATING METHOD THEREOF
20220180953 · 2022-06-09 ·

A storage device includes: a memory device including a plurality of memory cells configured to store data and a plurality of word lines connected to the plurality of memory cells and a memory controller in communication with the memory device and configured to control the memory device, including controlling the memory device to perform a read operation perform, upon a failure of the read operation on the memory cell, a read retry operation by changing the read voltage based on a history read table, and wherein the memory controller is further configured to update the history read table, upon a success of the read retry operation, based on whether the word line connected to the memory cell is a last programmed word line according to a program order among word lines connected to the group of memory cells.

Wear leveling in solid state drives
11733873 · 2023-08-22 · ·

A computer storage device having: a host interface; a controller; non-volatile storage media having memory units of different types and having different program erase budgets; and firmware. The firmware instructs the controller to: generate an address map mapping logical addresses to physical addresses of the memory units the different types; and adjust the address map based at least in part on the program erase budgets to level wear across the memory units of the different types.

MEMORY CONTROLLER, MEMORY SYSTEM WITH IMPROVED THRESHOLD VOLTAGE DISTRIBUTION CHARACTERISTICS, AND OPERATION METHOD
20220139474 · 2022-05-05 ·

A memory controller includes an over-program controller that preprograms and then erases the memory cells such that each of the memory cells has a first threshold voltage level, wherein fast cells are detected among the memory cells according to a threshold voltage less than or equal to a second threshold voltage less than the first threshold voltage, and a processor that generates fast cell information identifying the fast cells among the memory cells and stores the fast cell information in a buffer. The over-program controller controls the over-programming of the fast cells and normal programming of normal cells among the memory cells based on the fast cell information stored in the buffer.

Non-volatile memory device and program method of a non-volatile memory device

A method of programming a non-volatile memory includes executing at least two program loops on memory cells in a selected word line, generating a fail bit trend based on a result of executing each of the at least two program loops, predicting a plurality of program loops comprising an N program loop to be executed last on the memory cells, based on the generated fail bit trend, and changing, based on a result of predicting the plurality of program loops, a level of an N program voltage provided to the memory cells when the N program loop is executed.