A Linear Implementation of an Analog Resonate-and-Fire Neuron

TL;DR

This paper presents a linear, silicon-based resonate-and-fire neuron that aligns with state-space-model principles, demonstrating robustness, efficiency, and suitability for neuromorphic hardware in machine learning applications.

Contribution

It introduces a linear RAF neuron implemented in 22nm technology, analyzing its dynamics and resilience, and validates its effectiveness in a keyword-spotting task.

Findings

Robustness to process, voltage, and temperature variations.

Energy-efficient with favorable power and area trade-offs.

Effective in a practical keyword-spotting application.

Abstract

Oscillatory dynamics have recently proven highly effective in machine learning (ML), particularly through State-Space-Models (SSM) that leverage structured linear recurrences for long-range temporal processing. Resonate-and-Fire neurons capture such oscillatory behavior in a spiking framework, offering strong expressivity with sparse event-based communication. While early analog RAF circuits employed nonlinear coupling and suffered from process sensitivity, modern ML practice favors linear recurrence. In this work, we introduce a resonate-and-fire (RAF) neuron, built in 22nm Fully-Depleted Silicon-on-Insulator technology, that aligns with SSM principles while retaining the efficiency of spike-based communication. We analyze its dynamics, linearity, and resilience to Process, Voltage, and Temperature variations, and evaluate its power, performance, and area trade-offs. We map the characteristics of our circuit into a system-level simulation where our RAF neuron is utilized in a keyword-spotting task, showing that its non-idealities do not hinder performance. Our results establish RAF neurons as robust, energy-efficient computational primitives for neuromorphic hardware.