Symbol-Level Precoding for Near-Field ISAC

TL;DR

This paper explores symbol-level precoding in near-field ISAC systems, leveraging spherical wavefronts for precise joint sensing and communication, with convex optimization and MUSIC algorithm for target estimation.

Contribution

It introduces a novel SLP design for near-field ISAC that enhances joint sensing and communication performance, with convex optimization and accurate target estimation methods.

Findings

SLP-based design outperforms block-level methods

Convex optimization simplifies the design process

2D MUSIC accurately estimates target parameters

Abstract

The forthcoming 6G and beyond wireless networks are anticipated to introduce new groundbreaking applications, such as Integrated Sensing and Communications (ISAC), potentially leveraging much wider bandwidths at higher frequencies and using significantly larger antenna arrays at base stations. This puts the system operation in the radiative near-field regime of the BS antenna array, characterized by spherical rather than flat wavefronts. In this paper, we refer to such a system as near-field ISAC. Unlike the far-field regime, the near-field regime allows for precise focusing of transmission beams on specific areas, making it possible to simultaneously determine a target's direction and range from a single base station and resolve targets located in the same direction. This work designs the transmit symbol vector in near-field ISAC to maximize a weighted combination of sensing and communication performances subject to a total power constraint using symbol-level precoding (SLP). The formulated optimization problem is convex, and the solution is used to estimate the angle and range of the considered targets using the 2D MUSIC algorithm. The simulation results suggest that the SLP-based design outperforms the block-level-based counterpart. Moreover, the 2D MUSIC algorithm accurately estimates the targets' parameters.