Joint Transmit and Pinching Beamforming Design for Pinching Antenna-assisted Symbiotic Radio

Ze Wang, Guoping Zhang, Hongbo Xu, Wei Liu, Ming Zeng, Fang Fang, Dusit NiyatoarXiv 2025 · DOI · arXiv

Abstract

This paper investigates a novel downlink symbiotic radio framework enabled by the pinching antenna system (PASS), designed to enhance both primary and secondary transmissions through reconfigurable antenna positioning. This reconfigurability introduces additional degrees of freedom for adaptive pinching beamforming, thereby enabling constructive signal enhancement and interference suppression tailored to the locations of the backscatter device, the Internet of Things (IoT) receiver, and the primary receivers. To fully exploit these benefits, we formulate a joint transmit and pinching beamforming optimization problem that maximizes the achievable sum rate while satisfying the IoT receiver's detection error probability constraint and feasible deployment constraints for the pinching antennas. The resulting problem is inherently nonconvex and highly coupled. To address this challenge, we develop two complementary solution approaches. The first is a learning-aided gradient descent method, where the constrained optimization is reformulated into a differentiable form and solved through end-to-end learning. The second approach is an optimization-based successive convex approximation-particle swarm optimization method, which first determines the transmit beamforming solution using successive convex approximation and subsequently optimizes pinching beamforming via a particle swarm optimization search over candidate pinching antenna placements.

Introduction

The explosive growth of the Internet of Things (IoT) demands unprecedented spectrum and energy efficiency. Symbiotic Radio (SR) offers a clever solution: passive IoT devices, such as backscatter tags, embed their data into the active signals of primary transmitters, creating a mutually beneficial "symbiotic" relationship without needing their own RF carriers. However, backscatter links suffer from severe "double fading" and are incredibly weak, especially if the line-of-sight is blocked. While researchers have explored Reconfigurable Intelligent Surfaces (RIS) or movable antennas to improve these channels, their range of motion and performance gains are often physically limited.

The Pinching Antenna System (PASS)

Enter the Pinching Antenna System (PASS). PASS utilizes a dielectric waveguide where "pinching" antennas can be activated and moved anywhere along the guide. This offers massive flexibility to physically reshape the wireless channel and dodge blockages by moving the radiation point to find a better line-of-sight. The core challenge addressed in this work is determining exactly where to place these antennas while simultaneously optimizing the transmit beamforming, all while keeping the IoT signal readable and preventing physical antenna collisions.

The proposed PASS-enabled symbiotic radio system architecture, showing the base station, waveguides with pinching antennas, and the interaction between primary receivers and IoT devices.

Proposed Optimization Frameworks

To solve the highly coupled, non-convex joint optimization problem, we propose two distinct algorithmic approaches:

1. Learning-aided Gradient Descent (LGD)

This approach treats the optimization variables—antenna positions and beamforming weights—as learnable parameters in a custom neural network architecture. Instead of training on a static dataset, the variables are optimized on the fly using automatic differentiation (via the Adam optimizer). To handle strict physical constraints, we employ a clever reparameterization: the network predicts the offsets between antennas rather than absolute coordinates, which automatically guarantees minimum spacing constraints without complex penalty functions.

Architecture of the Learning-aided Gradient Descent (LGD) approach, illustrating the reparameterization of antenna positions and beamforming weights for constrained optimization.

2. SCA-PSO Hybrid Method

Because gradient-based methods like LGD can get trapped in local optima due to the highly multimodal nature of antenna positioning, we developed a two-stage alternating optimization method. In the first stage, Successive Convex Approximation (SCA) is used to optimize the transmit beamforming matrix. In the second stage, Particle Swarm Optimization (PSO)—a swarm intelligence algorithm—searches the physical waveguide space for the optimal antenna positions.

Key Insights and Results

Our evaluation in simulated environments reveals several critical insights into the benefits of PASS-enabled symbiotic radio:

  • Physical Flexibility Beats Static Arrays: Moving antennas along a waveguide effectively mitigates severe large-scale path loss and double-fading. At a transmit SNR of 65 dBm, the SCA-PSO and LGD algorithms achieve sum-rate gains of 65.6% and 55.9%, respectively, compared to conventional massive MIMO systems.
  • Resilience to Distance: As the distance between the base station and receivers increases, conventional MIMO performance drops rapidly. In contrast, the PASS-enabled system maintains a highly stable sum rate by physically reconfiguring to maintain optimal paths.
  • Algorithm Trade-offs: The SCA-PSO algorithm improves performance by 17.1% over the LGD method and 35.5% over schemes where the pinching antennas are fixed. While LGD is faster and more lightweight, SCA-PSO strikes a superior balance between performance and computational complexity for practical deployment.

Conclusion and Future Work

This work demonstrates that the joint design of transmit and pinching beamforming can significantly enhance the performance of symbiotic radio systems. By leveraging the physical degrees of freedom provided by PASS, we can overcome the inherent weaknesses of backscatter communication. Future research will explore PASS in Parasitic Symbiotic Radio (PSR) scenarios and investigate multi-antenna receiver architectures to further exploit the path-loss reconfiguration capabilities of pinching antennas.

Summary updated

Abstract

This paper investigates a novel downlink symbiotic radio framework enabled by the pinching antenna system (PASS), designed to enhance both primary and secondary transmissions through reconfigurable antenna positioning. This reconfigurability introduces additional degrees of freedom for adaptive pinching beamforming, thereby enabling constructive signal enhancement and interference suppression tailored to the locations of the backscatter device, the Internet of Things (IoT) receiver, and the primary receivers. To fully exploit these benefits, we formulate a joint transmit and pinching beamforming optimization problem that maximizes the achievable sum rate while satisfying the IoT receiver's detection error probability constraint and feasible deployment constraints for the pinching antennas. The resulting problem is inherently nonconvex and highly coupled. To address this challenge, we develop two complementary solution approaches. The first is a learning-aided gradient descent method, where the constrained optimization is reformulated into a differentiable form and solved through end-to-end learning. In this approach, the pinching antenna position matrix is reparameterized to automatically satisfy minimum spacing constraints, while transmit power and waveguide length limits are enforced via projection and normalization. The second approach is an optimization-based successive convex approximation-particle swarm optimization method, which first determines the transmit beamforming solution using successive convex approximation and subsequently optimizes pinching beamforming via a particle swarm optimization search over candidate pinching antenna placements.
BibTeX
@misc{wang2025joint,
  author = {Ze Wang and Guoping Zhang and Hongbo Xu and Wei Liu and Ming Zeng and Fang Fang and Dusit Niyato},
  title = {{Joint Transmit and Pinching Beamforming Design for Pinching Antenna-assisted Symbiotic Radio}},
  year = {2025},
  eprint = {2508.07002},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2508.07002},
  howpublished = {arXiv},
}
Figure