Real-time processing of analog signals on accelerated neuromorphic hardware

2026-02-04

AI Summary

This work presents the first end-to-end on-chip real-time processing pipeline that directly injects continuous analog audio signals into the analog computing units of the BrainScaleS-2 mixed-signal neuromorphic chip, bypassing conventional event-based sensors or spike-conversion frontends that incur additional power and latency. Leveraging the system’s 1000× physical acceleration, a spiking neural network processes interaural time differences for sound source localization, and an embedded microprocessor actuates a servo motor to track transient acoustic sources. Demonstrated in real-world conditions, the system achieves a low-power, high-efficiency perception–decision–action closed loop, validating the potential of fully analog neuromorphic architectures for real-time control tasks.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Humans and AI: Brain-Sensing and Analysis
• Machine Learning: Hardware-aware ML

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Sensory processing with neuromorphic systems is typically done by using either event-based sensors or translating input signals to spikes before presenting them to the neuromorphic processor. Here, we offer an alternative approach: direct analog signal injection eliminates superfluous and power-intensive analog-to-digital and digital-to-analog conversions, making it particularly suitable for efficient near-sensor processing. We demonstrate this by using the accelerated BrainScaleS-2 mixed-signal neuromorphic research platform and interfacing it directly to microphones and a servo-motor-driven actuator. Utilizing BrainScaleS-2's 1000-fold acceleration factor, we employ a spiking neural network to transform interaural time differences into a spatial code and thereby predict the location of sound sources. Our primary contributions are the first demonstrations of direct, continuous-valued sensor data injection into the analog compute units of the BrainScaleS-2 ASIC, and actuator control using its embedded microprocessors. This enables a fully on-chip processing pipeline$\unicode{x2014}$from sensory input handling, via spiking neural network processing to physical action. We showcase this by programming the system to localize and align a servo motor with the spatial direction of transient noise peaks in real-time.

Problem

• neuromorphic computing
• analog signal processing
• real-time processing
• sensor-actuator integration
• spiking neural networks

Innovation

• neuromorphic computing
• analog signal injection
• spiking neural network
• real-time sensor processing
• BrainScaleS-2