BenDi: An Energy-Efficient Quasi-Stochastic Systolic Architecture for Edge Bioelectronics

2026-06-10

AI Summary

Deploying high-accuracy convolutional neural networks (CNNs) on edge bioelectronic devices is challenged by stringent energy and resource constraints. This work proposes BenDi, a cross-layer co-optimized architecture spanning circuit to algorithm design, which integrates Bent-Pyramid quasi-stochastic multiplication, DiP systolic dataflow, and hardware-aware quantization to enable efficient inference at 0.5 V under a 22 nm process. Compared to state-of-the-art binarized-weight fixed-systolic architectures, BenDi reduces area by 3.35× and improves energy efficiency by 5×, while incurring only a 1%–3.3% drop in classification accuracy on the MIT-BIH and Apnea-ECG datasets. These results demonstrate a significant balance among accuracy, energy efficiency, and area utilization for edge biomedical applications.

Abstract

Continuous long-term monitoring and diagnosis of biomedical signals, such as electrocardiograms (ECGs), can help mitigate an increasing threat to public health. Artificial Intelligence (AI) models, such as Convolutional Neural Networks (CNNs), provide accurate monitoring and classification for relevant diseases; however, they require more computational resources than conventional AI hardware can typically afford, especially for a resource-constrained environment on the edge. In this work, we present BenDi, an energy-efficient quasi-stochastic systolic architecture for bioelectronic systems on the edge. BenDi leverages multiple levels of energy and power optimization, ranging from circuits to software quantization, including low supply voltage, the \underline{Ben}t-Pyramid data format for quasi-stochastic multiplication, the \underline{Di}P systolic dataflow, and hardware-aware quantization, to handle CNNs with high accuracy on the edge within limited hardware budgets. The hardware implementation results, using a commercial 22nm technology, show that BenDi architecture, at 0.5 Voltage and 100 MHz, offers 3.35x smaller area and 5x higher energy efficiency, compared to state-of-the-art binary-based weight-stationary systolic architectures. Regarding Bioelectronic edge systems, BenDi achieves an order-of-magnitude improvement in energy efficiency and another order-of-magnitude improvement in area efficiency, compared to its counterparts. This significant improvement comes at the cost of 1\% to 3.3\% accuracy loss on the MIT-BIH and Apnea-ECG benchmarks, respectively, compared with conventional computing using the 32-bit floating-point format.

Problem

• edge bioelectronics
• energy efficiency
• convolutional neural networks
• resource-constrained environment
• biomedical signal monitoring

Innovation

• quasi-stochastic computing
• systolic architecture
• energy-efficient AI
• edge bioelectronics
• hardware-aware quantization