Communication-Centric Integrated Sensing and Communications With Mixed Fields

Xiao, Yun; Wang, Enhao; Chen, Yunfei; Sun, Hongjian; Ikhlef, Aissa

Abstract

Dear editor, Integrated sensing and communications (ISAC) is expected to play a vital role in the sixth-generation (6G) wireless networks [1]. By utilizing high frequencies, extremely large-scale antenna arrays, and new antenna designs, the 6G-empowered ISAC faces several new challenges, one of which is the considerable near-field (NF) region [1]. Compared with far-field (FF) plane wave model, the NF model assumes spherical wave. This will affect beamforming (BF) designs in ISAC. Most previous works on ISAC beamforming assume the traditional FF model [2]. Recently, the BF design for NF ISAC has also attracted interest [3]. All these studies have considered only the NF or FF model. Reference [5] considered mixed NF and FF scenario for communications, but without sensing. In practice, the targets and the communication users (CU) could be in different fields. In the mixed near-and far-field, due to their different operating ranges of communications and radar or different mobilities of users and targets, assuming only NF or only FF models causes potential performance loss [4]. Inspired by the observations above, this letter considers a mixed-field ISAC scenario, where the dual-functional base station (BS) serves multiple NF and FF CUs and detects one NF target using mono-static setting. The contributions of the letter include the design of beamformers for the mixed near-and far-field in ISAC and the investigation of the sensing performance loss due to mismatched channel models. The system model is shown in Appendix A. Specifically, a communications performance fairness profile (FPO) problem is formulated, and a Dinkelbach-type successive convex approximation (SCA) algorithm is proposed to solve the problem. Note that using NF models for FF CUs or targets may increase the complexity with little performance improvement. It is beneficial to use FF model for FF target or CU in a mixed-field scenario. Channel Model. Assume that the dual-functional BS employs a uniform linear array (ULA) with Nt transmit antennas and Nr receive antennas. The adjacent antenna spacing is set to d = λ 2 , where λ is the carrier wavelength. Assume that the center of the BS is the origin, and the coordinate of the n-th antenna is [δ (n) d, 0], where n = 0,. .. , Nt − 1, δ (n) = n − N t −1 2. Assume that the target or CU has distance r and angle θ ∈ [ −π 2 , π 2 ] from the origin. The complex channel gain is β = β 0 r e −j 2π λ r , where β 0 is the 1-meter reference free-space path loss [5]. The Rayleigh distance, denoted as d F = 2D 2 λ , determines the boundary between the NF and FF regions, where D is the antenna aperture [3]. When r < d F , the NF channel vector is modeled as hnear = βa (r, θ), where a (r, θ) is the NF beam focusing vector that can be modeled as a (r, θ) = [e −j 2π λ r (0) −r ,. .. , e −j 2π λ r (n) −r ,. .. , e −j 2π λ r (N t −1) −r

Citation

Xiao, Y., Wang, E., Chen, Y., Sun, H., & Ikhlef, A. (2024). Communication-Centric Integrated Sensing and Communications With Mixed Fields. SCIENCE CHINA Information Sciences, 67(12), Article 219301. https://doi.org/10.1007/s11432-024-4152-9

• Journal Article Type: Article
• Acceptance Date: Sep 12, 2024
• Online Publication Date: Oct 15, 2024
• Publication Date: Oct 15, 2024
• Deposit Date: Oct 16, 2024
• Publicly Available Date: Oct 16, 2024
• Journal: SCIENCE CHINA Information Sciences
• Print ISSN: 1674-733X
• Electronic ISSN: 1869-1919
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• Volume: 67
• Issue: 12
• Article Number: 219301
• DOI: https://doi.org/10.1007/s11432-024-4152-9
• Public URL: https://durham-repository.worktribe.com/output/2959793