Invention Grant
- Patent Title: In-memory computing circuit for fully connected binary neural network
-
Application No.: US17042921Application Date: 2019-10-30
-
Publication No.: US11335387B2Publication Date: 2022-05-17
- Inventor: Weiwei Shan , Tao Wang
- Applicant: SOUTHEAST UNIVERSITY
- Applicant Address: CN Nanjing
- Assignee: SOUTHEAST UNIVERSITY
- Current Assignee: SOUTHEAST UNIVERSITY
- Current Assignee Address: CN Nanjing
- Agency: Bayramoglu Law Offices LLC
- Priority: CN201910623458.5 20190711
- International Application: PCT/CN2019/114226 WO 20191030
- International Announcement: WO2021/003899 WO 20210114
- Main IPC: G11C7/06
- IPC: G11C7/06 ; G11C7/10 ; G06N3/063 ; G11C7/12 ; G11C8/08 ; H03K3/356 ; H03K19/21

Abstract:
An in-memory computing circuit for a fully connected binary neural network includes an input latch circuit, a counting addressing module, an address selector, a decoding and word line drive circuit, a memory array, a pre-charge circuit, a writing bit line drive circuit, a replica bit line column cell, a timing control circuit, a sensitive amplifier and a NAND gate array, an output latch circuit and an analog delay chain. A parallel XNOR operation is performed in the circuit on the SRAM bit line, and the accumulation operation, activation operation and other operations are performed by the delay chain in the time domain. Partial calculation is completed while reading the data, and the delay chain with a small area occupation can be integrated with SRAM, thus reducing the energy consumption of the memory access process. Multi-column parallel computing also improves system throughput.
Public/Granted literature
- US20210312959A1 IN-MEMORY COMPUTING CIRCUIT FOR FULLY CONNECTED BINARY NEURAL NETWORK Public/Granted day:2021-10-07
Information query