Resource Allocation for Pinching-Antenna Systems (PASS)-enabled NOMA Communications

Songtao Xue, Jingjing Zhao, Kaiquan Cai, Xidong Mu, Zhenyu Xiao, Yuanwei LiuarXiv 2025 · DOI · arXiv

TL;DR

This paper introduces a novel communication framework that combines Pinching-Antenna Systems (PASS) with Non-Orthogonal Multiple Access (NOMA) to overcome large-scale path loss in wireless networks. By dynamically positioning antennas along dielectric waveguides and optimizing power and user scheduling, the authors achieve significant sum-rate improvements over conventional fixed-antenna and orthogonal access systems.

What Problem Does This Paper Address?

As we push toward 6G, reconfiguring the wireless environment is a top priority. Technologies like Fluid Antennas (FA) and Movable Antennas (MA) allow for local signal optimization, but they are often constrained to a very small area (a few wavelengths), making them ineffective against large-scale path loss. Reconfigurable Intelligent Surfaces (RIS) help, but they suffer from "double fading" (signal loss on both the hop to the RIS and the hop to the user).

PASS offers a different approach: it uses low-loss dielectric waveguides to carry signals, with "pinching" particles (antennas) that can be placed anywhere along the guide to radiate the signal directly toward a user. However, PASS has a hardware constraint: all antennas on a single waveguide must transmit the same signal. This creates a bottleneck when the number of users exceeds the number of waveguides. The authors address this by integrating NOMA, allowing multiple users to share a single waveguide through power-domain multiplexing.

Methods

The authors propose a Waveguide-Division (WD)-PASS-NOMA architecture. To manage the resulting complex, non-convex optimization problem (maximizing sum-rate), they decouple it into two main stages:

  1. Joint Power Allocation and Pinching Beamforming: The authors use a Penalty Dual Decomposition (PDD) algorithm. They transform the original problem into a Weighted Minimum Mean Square Error (WMMSE) form. To handle the "coupling" (where variables depend on each other), they use Augmented Lagrangian relaxation and solve it via Block Coordinate Descent (BCD), which iteratively updates power, antenna positions, and beamforming coefficients.
  2. User Scheduling: Assigning users to specific waveguides is a discrete "many-to-one" matching problem. The authors developed a low-complexity matching-based algorithm that avoids the prohibitive computational cost of an exhaustive search while maintaining near-optimal performance.

Datasets

The study relies on numerical simulations rather than a physical dataset. The simulation environment parameters include:

  • Carrier Frequency: 6 GHz.
  • Setup: A Base Station with 3 waveguides ($K=3$), each 10 meters long, serving 6 to 9 users.
  • Channel Model: Line-of-Sight (LoS) dominated links, typical for the intended PASS application.
  • Baselines: Compared against conventional fixed-position antenna systems (ULA), Random PA placement (RPP), and Orthogonal Multiple Access (OMA) schemes.

Key Insights and Ideas

  • The NOMA-PASS Synergy: NOMA is the perfect "software" solution for the PASS "hardware" limitation. Since a waveguide can only carry one data stream, NOMA allows that stream to be a superposition of multiple users' signals, effectively multiplying the system's capacity without adding more waveguides.
  • Combating Path Loss via Proximity: Unlike centralized arrays, PASS allows the "antenna" to physically move closer to the user along the waveguide, drastically reducing the distance the signal must travel through free space.
  • Scalability: The authors highlight that PASS is inherently scalable. Adding or removing pinching particles (PAs) to a waveguide is mechanically simple and doesn't require redesigning the internal RF circuitry of the base station.
  • Inter-cluster Interference Management: By optimizing the positions of PAs across different waveguides, the system can perform "pinching beamforming" to nullify interference between different user clusters.

Key Results

  • Sum-Rate Gains: The proposed WD-PASS-NOMA framework achieved a sum-rate of approximately 17.8 bps/Hz (with $N=14$ antennas and $M=6$ users), significantly outperforming the OMA baseline (~15.8 bps/Hz) and conventional fixed-antenna systems.
  • Near-Optimality: The low-complexity matching algorithm for user scheduling achieved 95% to 97% of the performance of an exhaustive search while being computationally feasible for real-time applications.
  • Degrees of Freedom: Increasing the number of pinching antennas ($N$) on a waveguide consistently improved the sum-rate, as it provided more "degrees of freedom" to fine-tune the beamforming and interference cancellation.

Limitations and Future Work

  • Hardware Realization: The paper is theoretical and simulation-based. The mechanical challenge of "pinching" dielectric particles and moving them rapidly in real-time remains a significant engineering hurdle.
  • Channel Estimation: The study assumes perfect channel state information (CSI). In a dynamic PASS system where antennas move, the overhead for constant channel estimation could be substantial.
  • Environment Complexity: The simulations focus on LoS-heavy environments. Performance in highly cluttered, non-LoS (NLoS) urban environments with significant multipath interference requires further investigation.

Summary updated

Abstract

Pinching-antenna systems (PASS) have emerged as a promising technology due to their ability to dynamically reconfigure wireless propagation environments. A novel PASS-based multi-user non-orthogonal multiple access (NOMA) framework is proposed by exploiting the waveguide-division (WD) transmission characteristic. Specifically, each NOMA user cluster is served by one dedicated waveguide, and the corresponding pinching beamforming is exploited to enhance the intra-cluster performance while mitigating the inter-cluster interference. Based on this framework, a sum-rate maximization problem is formulated for jointly optimizing power allocation, pinching beamforming, and user scheduling. To solve this problem, a two-step algorithm is developed, which decomposes the original problem into two subproblems. For the joint power allocation and pinching beamforming design, a penalty dual decomposition (PDD) algorithm is proposed to obtain the locally optimal solutions. Specifically, the coupling constraints are alleviated through augmented Lagrangian relaxation, and the resulting augmented Lagrangian (AL) problem is decomposed into four subproblems, which are solved by the block coordinate descent (BCD) method. For the user scheduling, a low-complexity matching algorithm is developed to solve the user-to-waveguide assignment problem. Simulation results demonstrate that 1) the proposed PASS-based NOMA framework under the WD transmission structure achieves significant sum-rate gain over conventional fixed-position antenna systems and orthogonal multiple access (OMA) scheme; and 2) the proposed matching-based user scheduling algorithm achieves near-optimal user-waveguide association with low computational complexity.
BibTeX
@misc{xue2025resource,
  author = {Songtao Xue and Jingjing Zhao and Kaiquan Cai and Xidong Mu and Zhenyu Xiao and Yuanwei Liu},
  title = {{Resource Allocation for Pinching-Antenna Systems (PASS)-enabled NOMA Communications}},
  year = {2025},
  eprint = {2512.03502},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2512.03502},
  howpublished = {arXiv},
}
Figure