[Paper Review] Near-Field Integrated Sensing and Communication: Performance Analysis and Beamforming Design

Kaiqian Qu, Shuaishuai Guo+1 more|arXiv (Cornell University)|Aug 12, 2023

Antenna Design and Optimization4 citations

TL;DR

This paper proposes near-field beamforming (NFBF) for integrated sensing and communication (ISAC) systems using extremely large-scale (XL-)arrays, modeling electromagnetic waves as spherical waves to exploit the distance dimension for enhanced beam focusing. It formulates a quadratically constrained quadratic program (QCQP) to minimize radar and communication beamforming errors under power constraints, achieving significant performance gains over far-field assumptions via Cramér-Rao bound (CRB) analysis and convex optimization-based power minimization.

ABSTRACT

This paper explores the potential of near-field beamforming (NFBF) in integrated sensing and communication (ISAC) systems with extremely large-scale arrays (XL-arrays). The large-scale antenna arrays increase the possibility of having communication users and targets of interest in the near field of the base station (BS). The paper first establishes the models of electromagnetic (EM) near-field spherical waves and far-field plane waves. With the models, we analyze the near-field beam focusing ability and the far-field beam steering ability by finding the gain-loss mathematical expression caused by the far-field steering vector mismatch in the near-field case. We formulate the NFBF design problem as minimizing the weighted summation of radar and the communication beamforming errors under a total power constraint and solve this quadratically constrained quadratic programming (QCQP) problem using the least squares (LS) method. Moreover, the Cramér-Rao bound (CRB) for target parameter estimation is derived to verify the performance of NFBF. Furthermore, we also perform power minimization using convex optimization while ensuring the required communication and sensing quality-of-service (QoS). The simulation results show the influence of model mismatch on near-field ISAC and the performance gain of transmit beamforming from the additional distance dimension of near-field.

Motivation & Objective

• Address the performance degradation in ISAC systems when assuming far-field plane waves for near-field scenarios with XL-arrays.
• Model electromagnetic waves as spherical waves in the near-field to exploit the additional distance dimension for beamforming gain.
• Formulate a joint beamforming design that minimizes radar and communication beamforming errors under a total power constraint.
• Derive the Cramér-Rao bound (CRB) for target parameter estimation to validate the performance of the proposed NFBF scheme.
• Enable power-efficient ISAC operation by solving a convex optimization problem to minimize transmit power while meeting quality-of-service (QoS) requirements.

Proposed method

• Model the near-field electromagnetic wave as a spherical wavefront, contrasting it with the traditional far-field plane wave assumption.
• Derive the gain-loss expression due to steering vector mismatch in the near-field, quantifying performance loss under far-field beamforming assumptions.
• Formulate the NFBF design as a quadratically constrained quadratic program (QCQP) to minimize the weighted sum of radar and communication beamforming errors.
• Solve the QCQP using the least squares (LS) method to obtain the optimal transmit beamforming vector.
• Derive the Cramér-Rao bound (CRB) for target parameter estimation to evaluate sensing accuracy and validate beamforming performance.
• Implement a convex optimization-based power minimization framework to reduce transmit power while satisfying SINR and CRB-based QoS constraints.

Experimental results

Research questions

• RQ1 How does the mismatch between far-field steering vectors and actual near-field wavefronts affect beamforming performance in ISAC systems with XL-arrays?
• RQ2 To what extent can near-field beamforming exploit the distance dimension to improve beam focusing and system performance compared to far-field assumptions?
• RQ3 What is the achievable trade-off between communication quality-of-service (SINR) and sensing accuracy (CRB) in near-field ISAC systems?
• RQ4 How can transmit power be minimized while maintaining required communication and sensing performance in near-field ISAC?
• RQ5 What is the theoretical performance limit of target parameter estimation in near-field ISAC, as quantified by the Cramér-Rao bound?

Key findings

• The proposed near-field beamforming (NFBF) scheme achieves significant performance gains over traditional far-field beamforming due to the exploitation of the distance dimension in spherical wave modeling.
• Model mismatch in the near-field leads to substantial beamforming gain loss when using far-field steering vectors, which the proposed NFBF effectively mitigates.
• The Cramér-Rao bound (CRB) analysis confirms that the proposed NFBF design enables accurate target parameter estimation, validating its sensing performance.
• The LS-based solution to the QCQP problem effectively minimizes the weighted sum of radar and communication beamforming errors under total power constraints.
• The convex optimization-based power minimization approach reduces transmit power while satisfying both communication SINR and sensing CRB requirements.
• Simulation results demonstrate that the NFBF design outperforms conventional far-field beamforming in both beam focusing and system spectral efficiency, particularly in dense near-field environments with XL-arrays.