Discriminability-Driven Spatial-Channel Selection with Gradient Norm for Drone Signal OOD Detection

TL;DR

This paper introduces a novel drone signal OOD detection method that adaptively weights features and uses gradient norms to improve detection accuracy and robustness across different drone types and signal qualities.

Contribution

It presents a discriminability-driven spatial-channel selection approach combined with gradient norm analysis for enhanced drone signal OOD detection.

Findings

Superior detection performance demonstrated in simulations

Robustness across different SNR levels and drone types

Enhanced discriminative power over existing methods

Abstract

We propose a drone signal out-of-distribution (OOD) detection algorithm based on discriminability-driven spatial-channel selection with a gradient norm. Time-frequency image features are adaptively weighted along both spatial and channel dimensions by quantifying inter-class similarity and variance based on protocol-specific time-frequency characteristics. Subsequently, a gradient-norm metric is introduced to measure perturbation sensitivity for capturing the inherent instability of OOD samples, which is then fused with energy-based scores for joint inference. Simulation results demonstrate that the proposed algorithm provides superior discriminative power and robust performance via SNR and various drone types.