MEMORY DEVICE AND METHOD FOR MANUFACTURING THE SAME

    公开(公告)号:US20240365565A1

    公开(公告)日:2024-10-31

    申请号:US18306289

    申请日:2023-04-25

    IPC分类号: H10B63/00 H10B61/00

    摘要: A memory device and a method for manufacturing the same are provided. The memory device includes drain pillar structures, source pillar structures, memory structures surrounding the drain pillar structures respectively, a channel structure, and a gate structure surrounding the drain pillar structures, the source pillar structures and the channel structure. The channel structure is divided into arc channel parts by the drain pillar structures and the source pillar structures.

    Page buffer counting for in-memory search

    公开(公告)号:US12131787B2

    公开(公告)日:2024-10-29

    申请号:US17891589

    申请日:2022-08-19

    摘要: A memory such as a 3D NAND array, having a page buffer having page buffer cells coupled to bit lines has a search word input such as a search word buffer coupled to word lines. A circuit, such as string select gates, is provided to connect a selected set of memory cells in the array to the page buffer. The page buffer includes sensing circuitry configured to apply a match sense signal to a latch in a plurality of storage elements for a stored data word and an input search word. Logic circuitry uses storage elements in the plurality of storage elements of the page buffer to accumulate the match sense signals output by the sensing circuitry over a sequence matching a plurality stored data words to one or more input search words. A match for a search is based on a threshold and the accumulated match sense signals.

    IN-MEMORY COMPUTING (IMC) MEMORY DEVICE AND IN-MEMORY COMPUTING METHOD

    公开(公告)号:US20240355387A1

    公开(公告)日:2024-10-24

    申请号:US18303726

    申请日:2023-04-20

    IPC分类号: G11C13/00

    摘要: An in-memory computing memory device includes: a plurality of computing memory cells forming a plurality of memory strings, the computing memory cells storing a plurality of weight values; a loading capacitor; and a measurement circuit. In IMC operations, a plurality of input voltages, corresponding to a plurality of input values, are input into the computing memory cells; a plurality of effective resistances of the computing memory cells are corresponding to the input voltages and the weight values; when a read voltage is applied to the plurality of computing memory cells, the computing memory cells generate a plurality of cell currents which are summed into a plurality of memory string currents for charging the loading capacitor; and based a capacitor voltage of the loading capacitor, at least one delay time and a predetermined voltage, an operation result of the input values and the weight values is determined.

    Method for manufacturing memory device

    公开(公告)号:US12114514B2

    公开(公告)日:2024-10-08

    申请号:US18519230

    申请日:2023-11-27

    摘要: A memory device and a method for manufacturing the memory device are provided. The memory device includes a stack and a plurality of memory strings. The stack is disposed on the substrate, and the stack includes a plurality of conductive layers and a plurality of insulating layers alternately stacked. The memory strings pass through the stack along a first direction, wherein a first memory string in the memory strings includes a first conductive pillar and a second conductive pillar, a channel layer, and a memory structure. The first conductive pillar and the second conductive pillar respectively extend along the first direction and are separated from each other. The channel layer is disposed between the first conductive pillar and the second conductive pillar. The memory structure surrounds the second conductive pillar, and the memory structure includes a resistive memory material.

    MEMORY DEVICE AND METHOD OF FABRICATING THE SAME

    公开(公告)号:US20240324199A1

    公开(公告)日:2024-09-26

    申请号:US18186961

    申请日:2023-03-21

    发明人: Min-Feng Hung

    IPC分类号: H10B43/27 H10B43/10 H10B43/30

    CPC分类号: H10B43/27 H10B43/10 H10B43/30

    摘要: A memory device includes a stacked structure, a channel pillar, a plurality of conductive pillars, and a slit. The stacked structure is located on a dielectric substrate, and includes a plurality of conductive layers and a plurality of insulating layers stacked alternately. The channel pillar extends through the stacked structure. The plurality of conductive pillars are located in the channel pillar and electrically connected with the channel pillar. The charge storage structure is located between the plurality of conductive layers and the channel pillar. The slit is located in the stacked structure. The slit includes a body part and an extension part. The body part extends through the stacked structure. The extension part is connected to the body part and located between the stacked structure and the dielectric substrate. The memory may be applied in 3D AND flash memory.

    Push-pull output driver and operational amplifier using same

    公开(公告)号:US12101068B2

    公开(公告)日:2024-09-24

    申请号:US17349586

    申请日:2021-06-16

    发明人: Yih-Shan Yang

    IPC分类号: H03F3/45 H03K3/26 H03K3/356

    摘要: A voltage driver circuit for an output stage of an operational amplifier, or other circuits, includes a level shifter and an output driver including a source follower and a common source amplifier in a push-pull configuration. The level shifter generates a node voltage as a function of an input voltage on the input node. The output driver including a first transistor having a control terminal receiving the node voltage, and connected between a supply voltage and an output node, and a second transistor having a control terminal receiving the input voltage from the input node, and connected between the output node and a reference voltage, wherein the first and second transistors have a common conductivity type.

    DATA OPTIMIZATION FOR HIGH BANDWIDTH (HBW) NVM AI INFERENCE SYSTEM

    公开(公告)号:US20240281636A1

    公开(公告)日:2024-08-22

    申请号:US18112827

    申请日:2023-02-22

    IPC分类号: G06N3/04 G06N3/063

    CPC分类号: G06N3/04 G06N3/063

    摘要: A method for storing weight data used to compute node values during inferencing operations conducted by a neural network comprises receiving a neural network definition. The neural network definition defines a neural network having a plurality of layers, each having a plurality of nodes. A set of weights used to compute a neural network inferencing result for each neural network node of the plurality of network nodes in the layer is determined. The set of weights determined for the layer in a page of memory can be stored in a high bandwidth non-volatile memory (NVM), such that any weights used to compute the neural network inference result for each neural network node of the plurality of nodes in the layer are stored together in the page of memory for retrieval together. Weights can be stored in different arrays across multiple memory channels.