Target Detection in ISAC Systems with Active RISs: A Multi-Perspective Observation Approach

TL;DR

This paper proposes a multi-perspective observation framework for ISAC systems with active RISs, enhancing target detection accuracy by exploiting diverse observation paths and advanced signal processing techniques.

Contribution

It introduces a novel multi-perspective observation approach combined with joint waveform and RIS reflection design for improved target detection in ISAC systems.

Findings

Significantly improved target detection accuracy in simulations.

Effective joint optimization of waveform, RIS coefficients, and receive filters.

Enhanced target-clutter separation using multi-perspective observations.

Abstract

Integrated sensing and communication (ISAC) has emerged as a transformative technology for 6G networks, enabling the seamless integration of communication and sensing functionalities. Reconfigurable intelligent surfaces (RIS), with their capability to adaptively reconfigure the radio environment, have shown significant potential in enhancing communication quality and enabling advanced cooperative sensing. This paper investigates a multi-RIS-assisted ISAC system and introduces a novel multi-perspective observation framework that leverages the diversity of multiple observation paths, each exhibiting distinct spatial, delay, and Doppler characteristics for both target and clutter. The proposed framework integrates symbol-level precoding (SLP) and space-time adaptive processing (STAP) to fully exploit the benefits of multi-perspective observations, enabling superior target-clutter separation and significantly improving detection accuracy. The objective is to jointly design the transmit waveform, reflection coefficients of multiple active RISs, and spatial-temporal receive filters to maximize the radar output signal-to-clutter-plus-noise ratio (SCNR) for target detection, while ensuring the quality-of-service (QoS) requirements of communication users. To address the resulting non-convex optimization problem, an effective iterative algorithm is developed, combining fractional programming (FP), majorization-minimization (MM), and the alternating direction method of multipliers (ADMM). Extensive simulation results validate the effectiveness of the proposed multi-perspective observation strategy, demonstrating its advantages in improving target detection performance in challenging environments.