Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

RIS-Assisted ISAC: Precoding and Phase-Shift Optimization for Mono-Static Target Detection

Özlem Tuğfe Demir, Emil Björnson

TL;DR

This paper derives the generalized likelihood ratio test (GLRT) detector and introduces a novel approach for jointly optimizing the configuration of RIS phase-shifts and precoding, which significantly enhances detection performance, especially when the clutter is strong.

Abstract

The reconfigurable intelligent surface (RIS) technology emerges as a highly useful component of the rapidly evolving integrated sensing and communications paradigm, primarily owing to its remarkable signal-to-noise ratio enhancement capabilities. In this paper, our focus is on mono-static target detection while considering the communication requirement of a user equipment. Both sensing and communication benefit from the presence of an RIS, which makes the channels richer and stronger. Diverging from prior research, we comprehensively examine three target echo paths: the direct (static) channel path, the path via the RIS, and a combination of these, each characterized by distinct radar cross sections (RCSs). We take both the line-of-sight (LOS) and the non-line-of-sight (NLOS) paths into account under a clutter for which the distribution is not known, but the low-rank subspace it resides. We derive the generalized likelihood ratio test (GLRT) detector and introduce a novel approach for jointly optimizing the configuration of RIS phase-shifts and precoding. Our simulation results underscore the paramount importance of this combined design in terms of enhancing detection probability. Moreover, it becomes evident that the derived clutter-aware target detection significantly enhances detection performance, especially when the clutter is strong.

RIS-Assisted ISAC: Precoding and Phase-Shift Optimization for Mono-Static Target Detection

TL;DR

This paper derives the generalized likelihood ratio test (GLRT) detector and introduces a novel approach for jointly optimizing the configuration of RIS phase-shifts and precoding, which significantly enhances detection performance, especially when the clutter is strong.

Abstract

The reconfigurable intelligent surface (RIS) technology emerges as a highly useful component of the rapidly evolving integrated sensing and communications paradigm, primarily owing to its remarkable signal-to-noise ratio enhancement capabilities. In this paper, our focus is on mono-static target detection while considering the communication requirement of a user equipment. Both sensing and communication benefit from the presence of an RIS, which makes the channels richer and stronger. Diverging from prior research, we comprehensively examine three target echo paths: the direct (static) channel path, the path via the RIS, and a combination of these, each characterized by distinct radar cross sections (RCSs). We take both the line-of-sight (LOS) and the non-line-of-sight (NLOS) paths into account under a clutter for which the distribution is not known, but the low-rank subspace it resides. We derive the generalized likelihood ratio test (GLRT) detector and introduce a novel approach for jointly optimizing the configuration of RIS phase-shifts and precoding. Our simulation results underscore the paramount importance of this combined design in terms of enhancing detection probability. Moreover, it becomes evident that the derived clutter-aware target detection significantly enhances detection performance, especially when the clutter is strong.

Paper Structure

This paper contains 10 sections, 1 theorem, 25 equations, 3 figures.

Table of Contents

  1. Introduction
  2. System and Channel Modeling
  3. Communication channel modeling and performance
  4. Mono-static target reflection modeling with unknown NLOS reflections
  5. Clutter modeling
  6. Mono-Static Target Detection
  7. ISAC Precoding and Phase-Shift Optimization
  8. Precoding optimization for given phase-shift configuration
  9. Numerical Results
  10. Conclusions

Key Result

Lemma 1

Let $\lambda_{1}\geq \cdots\geq \lambda_{r_2}\geq0$ be the nonnegative eigenvalues and $\mathbf{u}_{i}\in \mathbb{C}^{r_2}$, for $i=1,\ldots,r_2$ be the corresponding orthonormal eigenvectors of $\overline{\mathbf{A}}$. If $\overline{\mathbf{b}}=\mathbf{0}$, the optimal solution to the problem eq:op where $\gamma^{\star}>\lambda_1$ is the unique root of which can be solved using a bisection searc

Figures (3)

  • Figure 1: An ISAC transceiver with $K$ transmit antennas and $K$ receive antennas that want to transmit to a UE while also detecting the presence of a target. Both operations occur simultaneously with assistance from an RIS with $N$ elements. There are many paths in the channel. Solid lines represent paths leading to the target and dashed lines are paths leading back to the receiver. Dotted lines represent paths to the UE.
  • Figure 2: The detection probability with respect to the SNR for the received clutter-noise ratio of $20$ dB.
  • Figure 3: The detection probability with respect to the SNR for the received clutter-noise ratio of $40$ dB.

Theorems & Definitions (1)

  • Lemma 1