A Spatio-Temporal-Frequency Transformer Framework for Near-Field Target Recognition

TL;DR

This paper introduces a novel spatio-temporal-frequency transformer framework for near-field target recognition in integrated sensing and communication systems, leveraging electromagnetic features and spherical wavefront signatures for high accuracy.

Contribution

It proposes a new transformer-based framework that effectively fuses spatial, temporal, and frequency features for improved near-field target recognition.

Findings

Achieves high recognition accuracy with limited bandwidth.

Utilizes a lightweight spatial encoder for feature extraction.

Demonstrates effectiveness through simulation results.

Abstract

A target recognition framework relying on near-field integrated sensing and communication (ISAC) systems is proposed. By exploiting the distance-dependent spatial signatures provided by the near-field spherical wavefront, high-accuracy sensing is realized in a bandwidth-efficient manner. A spatio--temporal--frequency (STF) transformer framework is introduced for target recognition using electromagnetic features found in the wireless channel response. In particular, a lightweight spatial encoder is employed to extract features from the antenna array for each frame and subcarrier. These features are then fused by a time-frequency transformer head with positional embeddings to model temporal dynamics and cross-subcarrier correlations. Simulation results demonstrate that strong target recognition performance can be achieved even with limited bandwidth resources.