A Hybrid Edge Classifier: Combining TinyML-Optimised CNN with RRAM-CMOS ACAM for Energy-Efficient Inference

2025-02-14

AI Summary

Conventional digital-only TinyML inference approaches face fundamental energy-efficiency bottlenecks in resource-constrained extreme-edge sensing scenarios (e.g., wearables), where ultra-low power and real-time operation are critical. Method: This work proposes a hardware–software co-designed two-tier near-sensor classification architecture: a lightweight TinyML CNN frontend for feature extraction, coupled with a backend analog content-addressable memory (ACAM) implemented via RRAM-CMOS heterogeneous integration for template matching. Contribution/Results: It pioneers a hybrid digital–analog inference paradigm—integrating digital TinyML with analog ACAM—enabling synergistic optimization of accuracy and energy efficiency. Experimental results demonstrate a single-inference energy consumption of only 97.68 nJ, representing a 792× reduction versus the original teacher model (78.06 μJ → 97.68 nJ), while maintaining competitive classification accuracy—thus meeting stringent requirements for real-time, ultra-low-power operation at the extreme edge.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Analogy
• Computer Vision: Learning & Optimization for CV

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

In recent years, the development of smart edge computing systems to process information locally is on the rise. Many near-sensor machine learning (ML) approaches have been implemented to introduce accurate and energy efficient template matching operations in resource-constrained edge sensing systems, such as wearables. To introduce novel solutions that can be viable for extreme edge cases, hybrid solutions combining conventional and emerging technologies have started to be proposed. Deep Neural Networks (DNN) optimised for edge application alongside new approaches of computing (both device and architecture -wise) could be a strong candidate in implementing edge ML solutions that aim at competitive accuracy classification while using a fraction of the power of conventional ML solutions. In this work, we are proposing a hybrid software-hardware edge classifier aimed at the extreme edge near-sensor systems. The classifier consists of two parts: (i) an optimised digital tinyML network, working as a front-end feature extractor, and (ii) a back-end RRAM-CMOS analogue content addressable memory (ACAM), working as a final stage template matching system. The combined hybrid system exhibits a competitive trade-off in accuracy versus energy metric with $E_{front-end}$ = $96.23 nJ$ and $E_{back-end}$ = $1.45 nJ$ for each classification operation compared with 78.06$mu$J for the original teacher model, representing a 792-fold reduction, making it a viable solution for extreme edge applications.

Problem

• Develop energy-efficient edge ML classifier.
• Combine TinyML CNN with RRAM-CMOS ACAM.
• Optimize accuracy and energy for edge systems.

Innovation

• Hybrid software-hardware classifier
• TinyML-optimized CNN front-end
• RRAM-CMOS ACAM back-end