Recurrent Neural Radio Anomaly Detection

TL;DR

This paper presents a recurrent neural network approach for detecting radio anomalies, significantly improving detection of small anomalies in complex multi-user radio bands, with demonstrated effectiveness on real-world radio communication data.

Contribution

The paper introduces a novel recurrent neural network method for radio anomaly detection, enhancing sensitivity to small anomalies in complex radio environments.

Findings

High detection probability across various interference levels

Lower false alarm rates compared to baseline methods

Effective on real-world radio communication bands

Abstract

We introduce a powerful recurrent neural network based method for novelty detection to the application of detecting radio anomalies. This approach holds promise in significantly increasing the ability of naive anomaly detection to detect small anomalies in highly complex complexity multi-user radio bands. We demonstrate the efficacy of this approach on a number of common real over the air radio communications bands of interest and quantify detection performance in terms of probability of detection an false alarm rates across a range of interference to band power ratios and compare to baseline methods.