E-MAC: Enhanced In-SRAM MAC Accuracy via Digital-to-Time Modulation
In this article, we introduce a novel technique called E-multiplication and accumulation (MAC) (EMAC), aimed at enhancing energy efficiency, reducing latency, and improving the accuracy of analog-based in-static random access memory (SRAM) MAC accelerators. Our approach involves a digital-to-time wo...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal on Exploratory Solid-State Computational Devices and Circuits |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10804123/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In this article, we introduce a novel technique called E-multiplication and accumulation (MAC) (EMAC), aimed at enhancing energy efficiency, reducing latency, and improving the accuracy of analog-based in-static random access memory (SRAM) MAC accelerators. Our approach involves a digital-to-time word-line (WL) modulation technique that encodes the WL voltage while preserving the necessary linear voltage drop for precise computations. This eliminates the need for an additional digital-to-analog converter (DAC) in the design. Furthermore, the SRAM-based logical weight encoding scheme we present reduces the reliance on capacitance-based techniques, which typically introduce area overhead in the circuit. This approach ensures consistent voltage drops for all equivalent cases [i.e., <inline-formula> <tex-math notation="LaTeX">$(a { \times} b) = (b \times a)$ </tex-math></inline-formula>], addressing a persistent issue in existing state-of-the-art methods. Compared with state-of-the-art analog-based in-SRAM techniques, our E-MAC approach demonstrates significant energy savings (<inline-formula> <tex-math notation="LaTeX">$1.89\times $ </tex-math></inline-formula>) and improved accuracy (73.25%) per MAC computation from a 1-V power supply, while achieving a <inline-formula> <tex-math notation="LaTeX">$11.84\times $ </tex-math></inline-formula> energy efficiency improvement over baseline digital approaches. Our application analysis shows a marginal overall reduction in accuracy, i.e., a 0.1% and 0.17% reduction for LeNet5-based CNN and VGG16, respectively, when trained on the MNIST and ImageNet datasets. |
|---|---|
| ISSN: | 2329-9231 |