Coordinated Multipoint Transmission in Pinching Antenna Systems

TL;DR

This paper introduces a novel optimization approach for coordinated multipoint transmission in pinching antenna systems, leveraging gradient-based meta-learning to enhance sum rate performance over traditional methods.

Contribution

It formulates a non-convex optimization problem for CoMP with PASS, and applies a GML strategy to effectively solve it, achieving near-optimal results.

Findings

Achieves 92% of the optimal sum rate.

Outperforms benchmark methods.

Provides higher upper bounds on achievable rate.

Abstract

We study a coordinated multi-point (CoMP) transmission where two base stations (BSs), each supported by a pinching antenna system (PASS), are deployed to jointly serve communication users under spatial division multiple access (SDMA) technology. Pinching Antenna technology was introduced as a promising solution to overcome the large-scale fading that has been shown to be an impediment in multiple-input multiple-output (MIMO) systems. To realize the advantages of this technology in CoMP systems, which suffer from an upperbound rate limitation when traditional uniform linear arrays (ULAs) are adopted, we formulate an optimization problem with the aim of maximizing the achievable sum rate by jointly determining the transmit beamforming vectors and pinching locations on the waveguides while respecting the quality of service (QoS) requirements of users. This problem is inherently non-convex due to the strong coupling among its decision parameters, making it challenging to solve using traditional optimization methods. Thus, we utilize a gradient-based meta-learning (GML) strategy specifically designed for large-scale optimization tasks. Finally, numerical analysis demonstrates the effectiveness of the proposed GML approach, achieving 92 percent of the optimal solution, and the superiority of the solution presented compared to other benchmarks. In addition, it achieves a higher upper bound on the achievable rate compared to conventional CoMP systems.