Enabling ISAC in Real World: Beam-Based User Identification with Machine Learning

TL;DR

This paper introduces a deep learning approach for user identification in integrated sensing and communication systems, achieving over 93% accuracy using real-world radar data, thus advancing practical ISAC deployment.

Contribution

It proposes a scalable deep learning method for user identification in ISAC systems, outperforming traditional model-based solutions on real-world data.

Findings

Deep learning approach achieves 93.4% accuracy.

Scalable method works with varying numbers of objects.

Real-world radar data validates the approach.

Abstract

Leveraging perception from radar data can assist multiple communication tasks, especially in highly-mobile and large-scale MIMO systems. One particular challenge, however, is how to distinguish the communication user (object) from the other mobile objects in the sensing scene. This paper formulates this \textit{user identification} problem and develops two solutions, a baseline model-based solution that maps the objects angles from the radar scene to communication beams and a scalable deep learning solution that is agnostic to the number of candidate objects. Using the DeepSense 6G dataset, which have real-world measurements, the developed deep learning approach achieves more than $93.4\%$ communication user identification accuracy, highlighting a promising path for enabling integrated radar-communication applications in the real world.