Evaluation of Encoding Schemes on Ubiquitous Sensor Signal for Spiking Neural Network

TL;DR

This paper evaluates four sensor signal encoding schemes for spiking neural networks in ubiquitous computing, analyzing their accuracy, robustness, and energy efficiency using a gym activity recognition case study on the Loihi2 processor.

Contribution

It provides a comprehensive comparison of encoding schemes for SNNs in real-world sensor applications, guiding optimal scheme selection based on performance metrics.

Findings

Time-to-first-spike encoding reduces firing rate and maintains accuracy but is less robust.

Rate encoding achieves the highest classification accuracy (91.7%).

Multi-threshold delta modulation offers the best robustness against noise.

Abstract

Spiking neural networks (SNNs), a brain-inspired computing paradigm, are emerging for their inference performance, particularly in terms of energy efficiency and latency attributed to the plasticity in signal processing. To deploy SNNs in ubiquitous computing systems, signal encoding of sensors is crucial for achieving high accuracy and robustness. Using inertial sensor readings for gym activity recognition as a case study, this work comprehensively evaluates four main encoding schemes and deploys the corresponding SNN on the neuromorphic processor Loihi2 for post-deployment encoding assessment. Rate encoding, time-to-first-spike encoding, binary encoding, and delta modulation are evaluated using metrics like average fire rate, signal-to-noise ratio, classification accuracy, robustness, and inference latency and energy. In this case study, the time-to-first-spike encoding required the lowest firing rate (2%) and achieved a comparative accuracy (89%), although it was the least robust scheme against error spikes (over 20% accuracy drop with 0.1 noisy spike rate). Rate encoding with optimal value-to-probability mapping achieved the highest accuracy (91.7%). Binary encoding provided a balance between information reconstruction and noise resistance. Multi-threshold delta modulation showed the best robustness, with only a 0.7% accuracy drop at a 0.1 noisy spike rate. This work serves researchers in selecting the best encoding scheme for SNN-based ubiquitous sensor signal processing, tailored to specific performance requirements.