A PVT-Resilient Subthreshold SRAM-Based In-Memory Computing Accelerator with In-Situ Regulation for Energy-Efficient Spiking Neural Networks

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Kao, Shih-Hang, Hung, Yang-Chan, Wang, I-Wen, Liu, Bing-Han, Chen, Yu-Chia, Chang, Tian-Sheuan, Jou, Shyh-Jye, Liu, Chien-Nan, Chen, Hung-Ming, Chen, Wei-Zen
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910183352958976
author Kao, Shih-Hang
Hung, Yang-Chan
Wang, I-Wen
Liu, Bing-Han
Chen, Yu-Chia
Chang, Tian-Sheuan
Jou, Shyh-Jye
Liu, Chien-Nan
Chen, Hung-Ming
Chen, Wei-Zen
author_facet Kao, Shih-Hang
Hung, Yang-Chan
Wang, I-Wen
Liu, Bing-Han
Chen, Yu-Chia
Chang, Tian-Sheuan
Jou, Shyh-Jye
Liu, Chien-Nan
Chen, Hung-Ming
Chen, Wei-Zen
contents This paper presents a PVT-resilient, subthreshold SRAM-based computing-in-memory (CIM) macro tailored for energy-efficient spiking neural networks (SNNs). The macro integrates in-situ current sensors and distributed voltage regulators to enable robust large-scale (1024 wordlines, 1304 bitlines and 128 shared neuron cells) subthreshold current-mode CIM, mitigating energy overheads and process-voltage-temperature (PVT) sensitivity. The neuron cells adopt a programmable, memory cell-based firing threshold to enhance neuron robustness against PVT variations. The architecture uses a stride-tick batching schedule to significantly reduce buffer overhead with enhanced input data reuse. Exploiting the high sparsity of SNNs, the proposed system demonstrates significant improvements in energy efficiency and variation tolerance. Fabricated in 28-nm CMOS, the prototype attains 93.64\% accuracy on keyword spotting, delivers up to 1181.42 TOPS/W, and achieves 7.24 TOPS/mm^2, demonstrating a viable and efficient solution for high-performance edge SNN processing.
format Preprint
id arxiv_https___arxiv_org_abs_2605_00319
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A PVT-Resilient Subthreshold SRAM-Based In-Memory Computing Accelerator with In-Situ Regulation for Energy-Efficient Spiking Neural Networks
Kao, Shih-Hang
Hung, Yang-Chan
Wang, I-Wen
Liu, Bing-Han
Chen, Yu-Chia
Chang, Tian-Sheuan
Jou, Shyh-Jye
Liu, Chien-Nan
Chen, Hung-Ming
Chen, Wei-Zen
Hardware Architecture
This paper presents a PVT-resilient, subthreshold SRAM-based computing-in-memory (CIM) macro tailored for energy-efficient spiking neural networks (SNNs). The macro integrates in-situ current sensors and distributed voltage regulators to enable robust large-scale (1024 wordlines, 1304 bitlines and 128 shared neuron cells) subthreshold current-mode CIM, mitigating energy overheads and process-voltage-temperature (PVT) sensitivity. The neuron cells adopt a programmable, memory cell-based firing threshold to enhance neuron robustness against PVT variations. The architecture uses a stride-tick batching schedule to significantly reduce buffer overhead with enhanced input data reuse. Exploiting the high sparsity of SNNs, the proposed system demonstrates significant improvements in energy efficiency and variation tolerance. Fabricated in 28-nm CMOS, the prototype attains 93.64\% accuracy on keyword spotting, delivers up to 1181.42 TOPS/W, and achieves 7.24 TOPS/mm^2, demonstrating a viable and efficient solution for high-performance edge SNN processing.
title A PVT-Resilient Subthreshold SRAM-Based In-Memory Computing Accelerator with In-Situ Regulation for Energy-Efficient Spiking Neural Networks
topic Hardware Architecture
url https://arxiv.org/abs/2605.00319