🤖 AI Summary
This work addresses the limited flexibility of conventional Pinching antennas, whose fixed structures hinder independent control of complex radiation weights across array elements. To overcome this, the authors propose a novel analog beamforming architecture based on phase-mismatch engineering of guided waves under single-mode excitation. By introducing tunable propagation constants within the coupled-mode theory framework, the method enables independent and simultaneous configuration of both amplitude and phase for each element—a capability not previously achieved in Pinching antennas. The design is compatible with mechanical tuning mechanisms and integrates weighted minimum mean square error digital precoding with genetic algorithm–based antenna configuration optimization. Evaluated in multiuser downlink scenarios, particularly under interference-limited conditions, the proposed programmable architecture significantly outperforms both traditional arrays and existing Pinching antennas in terms of sum rate, demonstrating its effectiveness and robustness.
📝 Abstract
Pinching antenna systems (PASS) enable reconfigurable radiating elements and extended line-of-sight communication, mitigating path loss effects. However, existing designs lack fully controllable radiation weights, as they are governed by structural parameters rather than explicitly assigned variables. In this paper, we introduce a new degree of freedom (DoF) for PASS by enabling radiation weight control through phase-mismatch manipulation of guided waves under single-mode excitation within a coupled-mode framework. By tuning the propagation constants of pinching antennas, independent complex-weight control of individual elements is achieved, transforming PASS into a weight-adaptive analog beamforming architecture. Based on this principle, we present a physics-based hardware model that provides a unified framework for both amplitude-tunable pinching beamforming and conventional equal-power radiation models, ensuring compatibility with existing PASS implementations, such as movable setups. To evaluate the proposed model, we formulate a sum-rate maximization problem for hybrid precoding in multiuser downlink systems and solve it using an alternating optimization framework that combines weighted minimum mean square error-based digital precoding with genetic algorithm-based optimization of PASS configurations, including various scenarios such as weight tuning, antenna movability, and discrete activation. Numerical results demonstrate that the amplitude-tunable PASS architecture achieves consistent performance gains over conventional arrays and existing PASS schemes, with pronounced improvements in interference-limited regimes under practical constraints.