Online Anomaly Detection Based On Reservoir Sampling and LOF for IoT devices

TL;DR

This paper presents a resource-efficient anomaly detection method for IoT devices using reservoir sampling and LOF, enabling on-device training on microcontrollers with limited resources.

Contribution

It introduces a novel implementation of the LOF algorithm on microcontrollers and demonstrates on-device training capabilities for IoT anomaly detection.

Findings

LOF can be effectively implemented on microcontrollers.

On-device training of LOF is feasible on resource-constrained devices.

The proposed method enables real-time anomaly detection in IoT applications.

Abstract

The growing number of IoT devices and their use to monitor the operation of machines and equipment increases interest in anomaly detection algorithms running on devices. However, the difficulty is the limitations of the available computational and memory resources on the devices. In the case of microcontrollers (MCUs), these are single megabytes of program and several hundred kilobytes of working memory. Consequently, algorithms must be appropriately matched to the capabilities of the devices. In the paper, we analyse the processing pipeline for anomaly detection and implementation of the Local Outliner Factor (LOF) algorithm on a MCU. We also show that it is possible to train such an algorithm directly on the device, which gives great potential to use the solution in real devices.