Downlink and Uplink ISAC in Continuous-Aperture Array (CAPA) Systems

TL;DR

This paper introduces a novel continuous-aperture array (CAPA) framework for integrated sensing and communications (ISAC), providing new beamforming designs and analytical expressions that outperform traditional array systems in both downlink and uplink scenarios.

Contribution

The paper develops a comprehensive CAPA-based ISAC framework with new beamforming strategies, analytical performance metrics, and closed-form solutions, advancing the state-of-the-art in joint sensing and communication systems.

Findings

CAPA-based ISAC achieves higher communication and sensing rates.

The proposed beamforming designs optimize the CR-SR trade-off.

Numerical results show superior performance over conventional arrays.

Abstract

A continuous-aperture array (CAPA)-based integrated sensing and communications (ISAC) framework is proposed for both downlink and uplink scenarios. Within this framework, continuous operator-based signal models are employed to describe the sensing and communication processes. The performance of communication and sensing is analyzed using two information-theoretic metrics: the communication rate (CR) and the sensing rate (SR). 1) For downlink ISAC, three continuous beamforming designs are proposed: i) the communications-centric (C-C) design that maximizes the CR, ii) the sensing-centric (S-C) design that maximizes the SR, and iii) the Pareto-optimal design that characterizes the Pareto boundary of the CR-SR region. A low-complexity signal subspace-based approach is proposed to derive the closed-form optimal beamformers for the considered designs. On this basis, closed-form expressions are derived for the achievable CRs and SRs, and the downlink rate region achieved by CAPAs is characterized. 2) For uplink ISAC, the C-C and S-C successive interference cancellation-based methods are proposed to manage inter-functionality interference. Using the subspace approach closed-form expressions for the optimal detectors as well as the achievable CRs and SRs are derived. The uplink SR-CR region is characterized based on the time-sharing technique. Numerical results demonstrate that, for both downlink and uplink, CAPA-based ISAC achieves higher CRs and SRs as well as larger CR-SR regions compared to conventional spatially discrete array-based ISAC.