An Asynchronous Mixed-Signal Resonate-and-Fire Neuron
🤖 AI Summary
Edge computing demands ultra-low-power, real-time analog neuromorphic computation to mitigate data storage, transmission overhead, and energy consumption—challenges exacerbated by the need for temporal-frequency-selective processing inspired by biological neurons.
Method: This work proposes a CMOS-based asynchronous mixed-signal Resonate-and-Fire (R&F) neuron circuit, integrating resonant dynamics modeling, asynchronous handshake protocols, and scalable frequency-detection units.
Contribution/Results: It presents the first hardware implementation of oscillatory R&F neurons tailored for large-scale brain-inspired systems. Silicon validation across typical process corners demonstrates exceptional frequency selectivity, robustness, sub-100 nW per neuron power consumption, and support for high-density asynchronous interconnects. This work establishes a scalable, energy-efficient hardware paradigm for spatiotemporal signal processing at the edge.
Application Category
📝 Abstract
Analog computing at the edge is an emerging strategy to limit data storage and transmission requirements, as well as energy consumption, and its practical implementation is in its initial stages of development. Translating properties of biological neurons into hardware offers a pathway towards low-power, real-time edge processing. Specifically, resonator neurons offer selectivity to specific frequencies as a potential solution for temporal signal processing. Here, we show a fabricated Complementary Metal-Oxide-Semiconductor (CMOS) mixed-signal Resonate-and-Fire (R&F) neuron circuit implementation that emulates the behavior of these neural cells responsible for controlling oscillations within the central nervous system. We integrate the design with asynchronous handshake capabilities, perform comprehensive variability analyses, and characterize its frequency detection functionality. Our results demonstrate the feasibility of large-scale integration within neuromorphic systems, thereby advancing the exploitation of bio-inspired circuits for efficient edge temporal signal processing.
Problem
Research questions and friction points this paper is trying to address.
Designing low-power CMOS neuron circuits for edge computing
Implementing asynchronous resonate-and-fire neurons for temporal signal processing
Enabling bio-inspired neuromorphic systems for efficient frequency detection
Innovation
Methods, ideas, or system contributions that make the work stand out.
CMOS mixed-signal Resonate-and-Fire neuron circuit
Asynchronous handshake integration for scalability
Frequency detection for edge temporal signal processing
🔎 Similar Papers
No similar papers found.
