Receiver-Aware Jamming Classification for OFDM Systems via Multi-Channel Signal Representation

Abstract

Accurate jamming classification is essential for reliable orthogonal frequency division multiplexing (OFDM) systems, especially under low jamming-to-noise ratio (JNR) conditions where weak interference is difficult to detect. Existing approaches mainly rely on time-frequency representations (TFR), which discard phase information and fail to exploit the structure of the receiver processing chain. In this contribution, we propose a receiver-aware jamming classification framework based on a multi-channel signal representation. The input is constructed from receiver-domain components that are consistent with the OFDM receiver processing chain, including frequency-domain symbols, pilot masks, and raw channel estimates, preserving complete complex-valued information while embedding communication system priors. We further show that channel estimates contain discriminative patterns induced by different jamming types, enabling robust classification under low JNR conditions. Simulation results demonstrate that the proposed method significantly outperforms TFR-based approaches across a wide JNR range.