dc.citation.conferencePlace |
US |
- |
dc.citation.title |
IEEE International Midwest Symposium on Circuits and Systems |
- |
dc.contributor.author |
Park, Keonhee |
- |
dc.contributor.author |
Jeong, Hoichang |
- |
dc.contributor.author |
Lee, Kyuho Jason |
- |
dc.date.accessioned |
2024-01-31T18:36:56Z |
- |
dc.date.available |
2024-01-31T18:36:56Z |
- |
dc.date.created |
2023-11-16 |
- |
dc.date.issued |
2023-08-09 |
- |
dc.description.abstract |
A highly energy-efficient compute-in-memory (CIM) processor for a low-power spiking neural network (SNN) is proposed in this paper. Most previous CIM processors were limited to binary neural networks with poor accuracy. Other CIM processors for multi-bit precision convolutional neural networks were developed to increase the accuracy, but they showed low energy efficiency. In addition, most previous works suffered from a power-hungry analog-to-digital converter (ADC) for partial sum computation. They consumed lots of energy due to the current-mode or voltage-mode analog computations. To resolve the issues, we propose a Time-Domain SNN CIM (TS-CIM) processor with 9T1C bitcell for highly energy-efficient time-domain computation with a compact area. The proposed Time multiply-and-accumulate circuit removes ADC that consumes a large portion of the system energy. In addition, the Analog Precision Reconstruction Unit is introduced for multi-bit reconstruction of the phase-coded input activations of SNN. Thanks to pipelined architecture, TS-CIM enables execution of the whole convolution layers without wasted cycles for the stall. The proposed TS-CIM is designed with 65 nm CMOS logic technology and achieves 701.7 TOPS/W energy efficiency. |
- |
dc.identifier.bibliographicCitation |
IEEE International Midwest Symposium on Circuits and Systems |
- |
dc.identifier.uri |
https://scholarworks.unist.ac.kr/handle/201301/74623 |
- |
dc.language |
영어 |
- |
dc.publisher |
IEEE |
- |
dc.title |
A 701.7 TOPS/W Time-Domain Spiking Neural Network Compute-in-Memory Processor with 9T1C Bitcell |
- |
dc.type |
Conference Paper |
- |
dc.date.conferenceDate |
2023-08-06 |
- |