Reservoir-Based Distributed Machine Learning for Edge Operation

TL;DR

This paper presents a reservoir computing architecture called Deepdelay Loop Reservoir Computing (DLR) that enables efficient, in-situ machine learning training on resource-constrained edge sensors for RF spectrum analysis, improving accuracy and reducing complexity.

Contribution

The paper introduces DLR, a novel hardware-efficient reservoir computing approach for in-situ training on edge devices, with demonstrated applications in RF emitter identification and protocol recognition.

Findings

DLR outperforms state-of-the-art neural networks in accuracy.

DLR reduces hardware complexity and latency.

Distributed training with DLR maintains high accuracy with low communication overhead.

Abstract

We introduce a novel design for in-situ training of machine learning algorithms built into smart sensors, and illustrate distributed training scenarios using radio frequency (RF) spectrum sensors. Current RF sensors at the Edge lack the computational resources to support practical, in-situ training for intelligent signal classification. We propose a solution using Deepdelay Loop Reservoir Computing (DLR), a processing architecture that supports machine learning algorithms on resource-constrained edge-devices by leveraging delayloop reservoir computing in combination with innovative hardware. DLR delivers reductions in form factor, hardware complexity and latency, compared to the State-ofthe- Art (SoA) neural nets. We demonstrate DLR for two applications: RF Specific Emitter Identification (SEI) and wireless protocol recognition. DLR enables mobile edge platforms to authenticate and then track emitters with fast SEI retraining. Once delay loops separate the data classes, traditionally complex, power-hungry classification models are no longer needed for the learning process. Yet, even with simple classifiers such as Ridge Regression (RR), the complexity grows at least quadratically with the input size. DLR with a RR classifier exceeds the SoA accuracy, while further reducing power consumption by leveraging the architecture of parallel (split) loops. To authenticate mobile devices across large regions, DLR can be trained in a distributed fashion with very little additional processing and a small communication cost, all while maintaining accuracy. We illustrate how to merge locally trained DLR classifiers in use cases of interest.