用于持续机器监测的低功耗神经形态声学异常检测
Low-Power, Neuromorphic, Acoustic Anomaly Detection for Persistent Machine Monitoring
浏览论文内容
中文总结 AI 辅助
该研究在Intel Loihi 2神经形态处理器上实现低功耗声学异常检测，在两个基准测试中性能超基线，能耗比CPU、GPU低两个数量级，为持续机器监测提供实用方案。
中文摘要 AI 辅助
持续声学监测可在不接触机器的情况下检测故障，但始终在线的推理受限于功耗、延迟和部署复杂度。我们在Intel Loihi 2神经形态处理器上，针对干净和带噪条件，展示了基于自编码器的声学异常检测。对数梅尔特征在片外计算；归一化、自编码器推理、L1重构评分和阈值处理在片上运行。在干净的、麦克风位置不变的ToyADMOS ToyCar基准测试中，该片上模型在最大假阳性率0.1下，AUC达0.9959，标准化pAUC达0.9785；在DCASE 2026 Task 2 ToyCar带噪基准测试中，模型的源AUC为0.7990、目标AUC为0.6466、pAUC为0.6426，超过了已报告的基线指标。对16芯片Loihi 2 VPX系统的功耗分析显示，其实时吞吐量下每个样本的动态能耗为0.0406–0.0426 mJ，比CPU和GPU低两个数量级。这些结果表明，神经形态声学异常检测是低功耗、持续机器监测的实用候选方案。
英文摘要
Persistent acoustic monitoring can detect machine faults without physical contact, but always-on inference is constrained by power, latency, and deployment complexity. We demonstrate autoencoder-based acoustic anomaly detection on an Intel Loihi 2 neuromorphic processor under clean and noisy conditions. Log-mel features are computed off chip; normalization, autoencoder inference, L1 reconstruction scoring, and thresholding run on chip. In a clean, microphone-position-invariant ToyADMOS ToyCar benchmark, the on-chip model achieves 0.9959 AUC and 0.9785 standardized pAUC at maximum false-positive rate 0.1. In the DCASE 2026 Task 2 ToyCar noisy benchmark, the model achieves source AUC 0.7990, target AUC 0.6466, and pAUC 0.6426, exceeding reported baseline metrics. Power profiling on a 16-chip Loihi 2 VPX system shows real-time throughput with 0.0406$\unicode{x2013}$0.0426 mJ dynamic energy per sample, two orders of magnitude lower than both a CPU and GPU. These results support neuromorphic acoustic anomaly detection as a practical candidate for low-power, persistent machine monitoring.
发表机构
- Los Alamos National Laboratory（洛斯阿拉莫斯国家实验室）
- AeroVironment, Inc（AeroVironment公司）
- University of New Mexico COSMIAC Research Center（新墨西哥大学COSMIAC研究中心）
- Air Force Research Laboratory（空军研究实验室）
机构由 AI 辅助整理，请以论文原文为准。
