Joint Antenna Position and Transmit Power Optimization for Pinching Antenna-Assisted ISAC Systems

Yunhui Qin, Yaru Fu, Member, IEEE, Haijun Zhang, Fellow, IEEE · 2025

Read Joint Antenna Position and Transmit Power Optimization for Pinching Antenna-Assisted ISAC Systems by Yunhui Qin, Yaru Fu, Member, IEEE, Haijun Zhang, Fellow, IEEE (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-2503-12872.txt.

Introduction

Yunhui Qin, Yaru Fu, and Haijun Zhang explored a new antenna idea for wireless systems that must communicate and sense at the same time. Pinching antennas can be pinched onto a waveguide at chosen spots, which lets the system shape where signals leave and keep a clear line of sight. The team asked how to place the antennas and set transmit power so data rates stay high while sensing targets and an energy budget are met.

The joint problem is hard and non-convex, so they solved it with a reinforcement learning method that rewards both performance and continued exploration. In simulations their approach beat benchmark schemes on data rate, sensing quality, and stability.

Catalog record

CREATORYunhui Qin, Yaru Fu, Member, IEEE, Haijun Zhang, Fellow, IEEE
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE24 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-12872

Browse the full library · About the Tower of Babel

Full text

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

Joint Antenna Position and Transmit Power Optimization for Pinching Antenna-Assisted ISAC Systems Yunhui Qin, Yaru Fu, Member, IEEE, Haijun Zhang, Fellow, IEEE Abstract This letter explores how pinching antennas, an advanced flexible-antenna system, can enhance the performance of integrated sensing and communication (ISAC) systems by leveraging their adaptability, cost-effectiveness, and ability to facilitate line-of-sight transmission. To achieve this, a joint antenna positioning and transmit power optimization problem is formulated to maximize the total communication data rate while meeting the target sensing requirements and the system energy constraint. To address the complex non-convex optimization problem, we propose a maximum entropy-based reinforcement learning (MERL) solution. By maximizing cumulative reward and policy entropy, this approach effectively balances exploration and exploitation to enhance robustness. Numerical results demonstrate that the proposed MERL algorithm surpasses other benchmark schemes in cumulative reward, total data rate, sensing signal-to-noise ratio, and stability. I Introduction Integrated sensing and communication (ISAC) has emerged as a groundbreaking technology that plays a pivotal role in shaping future wireless communication systems [ 1 ] . It can incorporate radar sensing and wireless communication functions on a platform that leverages shared resources such as hardware, spectrum, and energy, thus improving operational efficiency, decreasing costs, and promoting sustainability [ 2 ] . This capability addresses the extensive application demands of future intelligent scenarios such as autonomous driving, smart cities, the industrial Internet of Things, and other domains [ 3 ] . The utilization efficiency of shared resources and the adaptability to dynamic environments present major challenges for ISAC systems, especially in their antenna systems. On one hand, the sharing of resources between sensing and communication functions necessitates the use of highly directional and flexible antennas to mitigate signal interference and improve resource efficiency. On the other hand, in dynamic and complex environments, antennas must be capable of dynamically adapting to environmental variations, thus ensuring high communication quality and sensing accuracy. However, traditional fixed-position antenna designs often struggle to adapt to these diverse requirements. In recent years, flexible antennas, such as reconfigurable intelligent surfaces [ 4 ] , intelligent reflecting surfaces [ 5 ] , fluid antenna systems [ 6 ] , and movable antennas [ 7 ] , have attracted considerable attention. Due to their ability to dynamically reconstruct the wireless channel, flexible antenna systems can significantly enhance performance compared to traditional fixed-position antenna systems [ 8 ] . However, in most existing flexible antennas, the variation range of antenna positions is generally limited to the wavelength scale, constraining their ability to counteract large-scale path loss. In addition, the high costs of many existing flexible antennas also restrict their applications in real-world scenarios. The emergence of pinching antennas overcomes these challenges, as they can create new line-of-sight (LoS) links or enhance existing transceiver channels by using low-cost dielectric materials placed at arbitrary positions on a dielectric waveguide [ 9 ] . Unlike traditional antennas, pinching antennas can be flexibly deployed, and increasing the number of pinching antennas incurs almost no additional cost [ 9 ] . It can be observed that both the pinching antenna positions and transmission power significantly affect the performance of the wireless systems. In [ 8 ] , the authors studied the optimal locations and number of pinching antennas to activate in order to maximize the throughput of a non-orthogonal multiple access-assisted pinching antenna system and proposed a practical and low-complexity solution. Later, the authors of [ 10 ] optimized the locations of the pinching antennas to maximize the downlink transmission rate. Although the effectiveness of pinching antennas is verified in [ 8 , 9 , 10 ] , to the best of the authors’ knowledge, the potential of optimizing pinching antenna positions and power allocation in ISAC systems has not been fully explored, which stimulate our work. For brevity, the main contributions of this letter are summarized below: • We investigate pinching antenna-assisted ISAC systems and formulate a joint optimization problem for pinching antenna positions and power allocation to improve communication rate and sensing signal-to-noise ratio (SNR). • We propose a maximum entropy-based reinforcement learning (MERL) algorithm to solve the total data rate maximization problem while meeting sensing requirements and the system energy constraint. By focusing on maximizing cumulative reward and policy entropy, this approach ensures a proper balance between exploration and exploitation, resulting in greater robustness. • The numerical results show that the proposed MERL algorithm significantly outperforms other benchmarks, achieving higher cumulative reward, total data rate, sensing SNR, and stability. Specifically, the proposed MELR algorithm has improved the normalized total data rate by 20.3% compared to the twin delayed deep deterministic policy gradient (TD3) [ 11 ] and by 44.4% compared to the deep deterministic policy gradient (DDPG) [ 12 ] . Additionally, it has improved the normalized sensing SNR by 16.7% over the TD3 and far surpasses that of the DDPG algorithm. The reminder parts of this letter are given as follows. In Section II , we elaborate on the considered system model and formulate the total data rate maximization problem subject to target sensing constraints. In Section III , we reconstruct the optimization problem and introduce a MERL solution. In Section IV , extensive numerical results are presented to evaluate the proposed algorithm compared with benchmarks. Finally, we conclude this paper in Section V . II System Model and Problem Formulation As illustrated in Fig. 1 , we consider a pinching antenna-assisted ISAC system that consists of a base station (BS) equipped with N pinching antennas, serving M single-antenna mobile users (UEs) and K sensing targets. The pinching antennas establish new LoS transceiver links. Let \psi_{m,t}=(x_{m,t},y_{m,t},0) denote the location of the m -th UE in time slot t , define \psi^{\rm{pin}}_{n,t}=(x^{\rm{pin}}_{n,t},0,d) as the location of the pinching antenna n , and denote by \psi^{\rm{sen}}_{k}=(x_{k,t},y_{k,t},0) the location of the sensing target k . To facilitate system description and algorithm design, the collections of all pinching antennas, mobile UEs, and sensing targets are represented by \mathcal{N}=\{1,2,\cdots N\} , \mathcal{M}=\{1,2,\cdots M\} , and \mathcal{K}=\{1,2,\cdots K\} , respectively. Fig. 1: Illustration of pinching antenna-assisted ISAC system, enabling both UE service and target sensing via LoS links established by pinching antennas. Based on the spherical wave channel model, in time slot t , t\in\mathcal{T}=\{1,2,\cdots T\} , the chennel vector of m -th UE can be expressed as follows [ 9 ] : {\boldsymbol{h}}_{m,t}=\left[\frac{\alpha e^{-j\frac{2\pi}{\lambda}{\lvert\psi_{m,t}-\psi^{\rm{pin}}_{1,t}\rvert}}}{\lvert\psi_{m,t}-\psi^{\rm{pin}}_{1,t}\rvert}\cdots\frac{\alpha e^{-j\frac{2\pi}{\lambda}{\lvert\psi_{m,t}-\psi^{\rm{pin}}_{N,t}\rvert}}}{\lvert\psi_{m,t}-\psi^{\rm{pin}}_{N,t}\rvert}\right]^{T}, (1) where \alpha=\frac{c}{4\pi f_{c}} , c denotes the speed of light, f_{c} is the carrier frequency, and \lambda represents wavelength. The N pinching antennas are located on the same waveguide, meaning that the signal transmitted by one pinching antenna is a phase-shifted version of that transmitted by another. Let x_{c,t} denote the information signal passed onto the waveguide, x_{s,t} represent the dedicated sensing signal. Therefore, the signal vector {\boldsymbol{x}} can be expressed as {\boldsymbol{x}}=\left[\sqrt{\frac{P}{N}}e^{-j\theta_{1,t}}\cdots\sqrt{\frac{P}{N}}e^{-j\theta_{N,t}}\right]^{T}(x_{c,t}+x_{s,t}), (2) where P is the total transmit power, \theta_{n,t} is the signal phase shift through antenna n and \theta_{n,t}=2\pi\frac{\lvert\psi^{\rm{pin}}_{0}-\psi^{\rm{pin}}_{n,t}\rvert}{\lambda_{0}} . Thereof, \psi^{\rm{pin}}_{0} is the location of feed point of waveguide, \lambda_{0} denotes the waveguide wavelength in a dielectric waveguide, i.e., \lambda_{0}=\frac{\lambda}{n_{\rm{neff}}} , in which n_{\rm{neff}} is the effective refractive index of a dielectric waveguide [ 13 ] , and \lambda=\frac{2\pi}{f_{c}} . The received signal of m -th UE can be denoted by y_{m,t}={\boldsymbol{h}}^{H}_{m,t}{\boldsymbol{x}}+w_{m,t}

Read the complete text