Deep Learning-based Target-To-User Association in Integrated Sensing and Communication Systems

TL;DR

Simulation results over different urban vehicular mobility scenarios show that the proposed T2U method provides a probability of correct association that increases with the size of the BS antenna array, highlighting the respective increase of the separability of the VEs in the beamspace.

Abstract

In Integrated Sensing and Communication (ISAC) systems, matching the radar targets with communication user equipments (UEs) is functional to several communication tasks, such as proactive handover and beam prediction. In this paper, we consider a radar-assisted communication system where a base station (BS) is equipped with a multiple-input-multiple-output (MIMO) radar that has a double aim: (i) associate vehicular radar targets to vehicular equipments (VEs) in the communication beamspace and (ii) predict the beamforming vector for each VE from radar data. The proposed target-to-user (T2U) association consists of two stages. First, vehicular radar targets are detected from range-angle images, and, for each, a beamforming vector is estimated. Then, the inferred per-target beamforming vectors are matched with the ones utilized at the BS for communication to perform target-to-user (T2U) association. Joint multi-target detection and beam inference is obtained by modifying the you only look once (YOLO) model, which is trained over simulated range-angle radar images. Simulation results over different urban vehicular mobility scenarios show that the proposed T2U method provides a probability of correct association that increases with the size of the BS antenna array, highlighting the respective increase of the separability of the VEs in the beamspace. Moreover, we show that the modified YOLO architecture can effectively perform both beam prediction and radar target detection, with similar performance in mean average precision on the latter over different antenna array sizes.

Figures (11)

• Figure 1: ISAC system with co-located mmWave radar sensor and communication base station. (a) provides a front view representation of the system, showing analog communication beams on the vertical direction at the BS, and highlighting the radar slant-range plane used for radar imaging. (b) shows a top view of the system, depicting analog communication beams on the horizontal direction at the BS.
• Figure 2: Hybrid MIMO sub-connected Tx. architecture.
• Figure 3: YOLOv8 architecture utilized for target class and beam indices prediction (upper-left block). The remaining blocks expand the layer components depicted in the architecture graph. Conv2D denotes a 2-dimensional convolution, BN indicates batch normalization, U describes an upsampling layer, and SiLU represents a Sigmoid Linear Unit activation function.
• Figure 4: Proposed method for the integration of deep learning-based radar multi-target detection and beam prediction aimed at T2U association. A sequence of RA radar images is provided to a modified YOLOv8 model to jointly infer radar targets' positioning and classification information and the per-target beamforming vectors' classes. The predicted beam indices are then matched with the ones used at the BS to perform T2U association.
• Figure 5: Communication and sensing channel simulation framework.