Optimal User and Target Scheduling, User-Target Pairing, and Low-Resolution Phase-Only Beamforming for ISAC Systems

Luis F. Abanto-Leon, Setareh Maghsudi · 2025

Read Optimal User and Target Scheduling, User-Target Pairing, and Low-Resolution Phase-Only Beamforming for ISAC Systems by Luis F. Abanto-Leon, Setareh Maghsudi (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-11593.txt.

Introduction

One radio system that both talks and senses must choose which users and targets to serve in each slot. It must also pair them up and aim its beams with cheap hardware that shifts phase only in coarse steps. Luis Abanto-Leon and Setareh Maghsudi solve that whole knot at once in a 2025 paper. The joint problem is a hard mixed puzzle of yes-no choices and curved math.

Their move is to restate the puzzle in a form that standard solvers can finish with a guarantee of the best answer. Against staged guesswork that takes the choices one at a time, the joint optimum wins clearly. The staged answers also swing wildly with the scene, while the exact method holds steady. The result gives system designers a benchmark that no shortcut should be judged without.

Catalog record

CREATORLuis F. Abanto-Leon, Setareh Maghsudi
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE44 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-11593

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Optimal User and Target Scheduling, User-Target Pairing, and Low-Resolution Phase-Only Beamforming for ISAC Systems Luis F. Abanto-Leon, Setareh Maghsudi Abstract We investigate the joint user and target scheduling, user-target pairing, and low-resolution phase-only beamforming design for integrated sensing and commmunications (ISAC) . Scheduling determines which users and targets are served, while pairing specifies which users and targets are grouped into pairs. Additionally, the beamformers are designed using few-bit constant-modulus phase shifts. This resource allocation problem is a nonconvex mixed-integer nonlinear program (MINLP) and challenging to solve. To address it, we propose an exact mixed-integer linear program (MILP) reformulation, which leads to a globally optimal solution. Our results demonstrate the superiority of an optimal joint design compared to heuristic stage-wise approaches, which are highly sensitive to scenario characteristics. I Introduction Integrated sensing and commmunications (ISAC) is an innovative technology offering several advantages, including improved radio resource utilization, reduced costs, and simplified system complexity [ 1 ] . However, ISAC also introduces new challenges, particularly in designing the radio resource allocation to jointly fulfill users’ and targets’ requirements. To enhance sensing accuracy, recent ISAC literature has focused on high frequencies, such as millimeter-wave and terahertz bands, using beamforming to mitigate the severe path loss [ 2 ] . However, the high costs of radio components for controlling signals’ amplitude and phase at these frequencies have led to the adoption of phase-only beamformers as a more cost-effective solution [ 3 ] . Despite the need for practical solutions, the literature primarily features designs with infinite-resolution phases, e.g., [ 4 , 5 , 6 , 7 ] , which are infeasible in real-world deployments. Only a few works have accounted for low-resolution phases, e.g., [ 8 , 9 ] , but these designs have relied on approximations to handle phase discretization resulting in inefficient radio resource utilization and highlighting the need for novel approaches that can achieve an optimal design. Practical systems often cannot simultaneously serve all users in a system due to limitations, e.g., in the number of radio-frequency (RF) chains, underscoring the need for user scheduling. While this aspect has been well investigated in communication systems, e.g., [ 10 ] , yielding substantial gains, it remains unexplored in ISAC . Hence, incorporating user scheduling into ISAC ’s resource allocation could significantly enhance system performance by enabling optimized decision-making on which users to serve jointly. Target scheduling is equally critical, aiming to optimize target selection for sensing while considering resource limitations. While a few studies, e.g., [ 11 ] , have delved into this topic, it remains largely under-explored. Fig. 1: ISAC system consisting of a BS and multiple users and targets. “Allocation #1” shows a favorable resource allocation strategy, featuring well-separated pairs and strong alignment between users and targets within each pair. This ensures high directivity, benefiting both users and targets. The sufficient spacing between scheduled users ( \mathsf{U}_{1} , \mathsf{U}_{3} , \mathsf{U}_{5} ) minimizes inter-user interference, while the adequate separation between targets ( \mathsf{T}_{1} , \mathsf{T}_{3} ) reduces cross-interference. In contrast, “Allocation #2” is less optimal due to closer pairs spread over a wider angular range, resulting in weaker alignment and reduced directivity. The angular proximity of users ( \mathsf{U}_{2} , \mathsf{U}_{3} , \mathsf{U}_{4} ) increases inter-user interference, and the limited separation between targets ( \mathsf{T}_{1} , \mathsf{T}_{2} ) heightens cross-correlation. While this scenario highlights the importance of alignment, a more comprehensive resource allocation approach is needed. Factors such as user channel conditions, target characteristics, and phase resolution significantly influence performance, necessitating a unified optimization strategy beyond fixed pairing and scheduling criteria. Most ISAC works have assumed that user-target pairing is predetermined and provided as prior information, e.g., [ 5 , 6 , 12 ] , which is employed for subsequent resource allocation. Recently, however, there has been a shift towards exploring flexible user-target pairing, e.g., [ 13 , 14 ] , allowing for improved performance by dynamically associating users and targets based on factors such as location and channel conditions. Yet, existing studies on this topic have only considered one user per time slot, e.g., [ 13 , 14 ] , thereby avoiding multiuser interference. Exploiting spatial multiplexing to service multiple users and targets simultaneously is crucial in modern radar and communication systems, as it improves radio resource utilization, making it a key aspect to consider. Jointly designing scheduling, pairing, and beamforming is a challenging task. Although performing these processes in separate stages can simplify the design, this decoupling often results in suboptimal performance. Particularly, pairing based on a single criterion, such as alignment, as shown in Fig. 1 , is a valid approach. However, pairing cannot be treated in isolation from scheduling, as both are influenced by users’ channel characteristics and targets’ features. Furthermore, phase resolution constrains the beamforming directions, further emphasizing the interdependence between scheduling, pairing, and beamforming. This paper investigates the optimal joint design of these three key processes, resulting in a novel resource allocation problem that is formulated as a nonconvex mixed-integer nonlinear program (MINLP) . We propose a reformulation that transforms the nonconvex MINLP into a mixed-integer linear program (MILP) , enabling global optimality through convex transformations that preserve the original solution space. Simulations demonstrate that our joint design provides significant performance gains over four baseline methods inspired by existing literature, which rely on heuristic scheduling and pairing. Notation : Matrices and vectors are respectively denoted by \mathbf{A} and \mathbf{a} . The transpose, Hermitian transpose, and trace of \mathbf{A} are denoted by \mathbf{A}^{\mathrm{T}} , \mathbf{A}^{\mathrm{H}} , and \mathrm{Tr}\left(\mathbf{A}\right) , respectively. The l -th row and i -th column of \mathbf{A} are denoted by \left[\mathbf{A}\right]_{l,:} and \left[\mathbf{A}\right]_{:,i} , respectively, and the l -th element of \mathbf{a} is denoted by \left[\mathbf{a}\right]_{l} . II System Model and Problem Formulation We consider a multi-user, multi-target downlink ISAC system, as illustrated in Fig. 1 , where the base station (BS) jointly performs sensing and communication tasks. Preliminaries: The BS has N transmit antennas and N receive antennas, indexed by set \mathcal{N}=\left\{1,\dots,N\right\} . Also, there are T targets and U single-antenna users, indexed by sets \mathcal{T}=\left\{1,\dots,T\right\} and \mathcal{U}=\left\{1,\dots,U\right\} , respectively. The BS has K RF chains, where K\leq U . Each RF chain supports a single user’s data stream. Following the current industry stance on ISAC , which advocates prioritizing communications while enabling sensing opportunistically, we predestine the use of RF chains primarily to serve user demands. Thus, users can be served individually or jointly with a target, but targets are not allocated RF chains for sensing only. The BS chooses J targets to sense from the total T , where J\leq\min\left\{K,T\right\} . In particular, each target is sensed simultaneously while servicing exactly one user 1 1 1 To maintain a practical perspective and prevent performance degradation [ 13 ] , we adopt a one-to-one pairing. Yet, our approach can be extended to a many-to-one pairing, allowing a user to be matched with multiple targets. . In the sequel, we denote the u -th user and the t -th target by \mathsf{U}_{u} and \mathsf{T}_{t} , respectively. User scheduling: As the RF chains are limited, only a subset of users can be scheduled at a given channel use 2 2 2 We assume that unscheduled users are queued for a subsequent channel

Read the complete text