AHAR: Adaptive CNN for Energy-efficient Human Activity Recognition in Low-power Edge Devices

TL;DR

This paper introduces AHAR, an adaptive CNN architecture designed for energy-efficient human activity recognition on low-power edge devices, outperforming traditional methods in energy consumption and maintaining high accuracy.

Contribution

The paper proposes a novel adaptive CNN architecture with an output block predictor, improving energy efficiency without sacrificing recognition performance.

Findings

Achieves comparable or better accuracy than state-of-the-art methods.

Significantly reduces energy consumption and memory usage on hardware.

Outperforms traditional early exit architectures in energy efficiency.

Abstract

Human Activity Recognition (HAR) is one of the key applications of health monitoring that requires continuous use of wearable devices to track daily activities. This paper proposes an Adaptive CNN for energy-efficient HAR (AHAR) suitable for low-power edge devices. Unlike traditional early exit architecture that makes the exit decision based on classification confidence, AHAR proposes a novel adaptive architecture that uses an output block predictor to select a portion of the baseline architecture to use during the inference phase. Experimental results show that traditional early exit architectures suffer from performance loss whereas our adaptive architecture provides similar or better performance as the baseline one while being energy-efficient. We validate our methodology in classifying locomotion activities from two datasets- Opportunity and w-HAR. Compared to the fog/cloud computing approaches for the Opportunity dataset, our baseline and adaptive architecture shows a comparable weighted F1 score of 91.79%, and 91.57%, respectively. For the w-HAR dataset, our baseline and adaptive architecture outperforms the state-of-the-art works with a weighted F1 score of 97.55%, and 97.64%, respectively. Evaluation on real hardware shows that our baseline architecture is significantly energy-efficient (422.38x less) and memory-efficient (14.29x less) compared to the works on the Opportunity dataset. For the w-HAR dataset, our baseline architecture requires 2.04x less energy and 2.18x less memory compared to the state-of-the-art work. Moreover, experimental results show that our adaptive architecture is 12.32% (Opportunity) and 11.14% (w-HAR) energy-efficient than our baseline while providing similar (Opportunity) or better (w-HAR) performance with no significant memory overhead.