Answers are grounded in this paper and cite their sources.
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., <tex-math notation="LaTeX">(a × b) = (b × a) </tex-math>], 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 ( <tex-math notation="LaTeX">1.89× </tex-math>) and improved accuracy (73.25%) per MAC computation from a 1-V power supply, while achieving a <tex-math notation="LaTeX">11.84× </tex-math> 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.
No takes yet. Share an insight, caveat, or question.
Seyedfaraji et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: