⚡ electrical engineering

Pinching Antennas for Multiple Access in Multigroup Multicast Communications

This paper proposes and evaluates three multiple access schemes—treating interference as noise, non-orthogonal multiple access, and time-division multiple access with distinct reconfiguration protocols—to jointly optimize pinching antenna placement and resource allocation, demonstrating that the proposed flexible architecture significantly enhances max-min fairness compared to traditional fixed-antenna systems.

Original authors: Shan Shan, Chongjun Ouyang, Yong Li, Yuanwei Liu

Published 2026-02-26

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Shan Shan, Chongjun Ouyang, Yong Li, Yuanwei Liu

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: The "Magic Wire" vs. The "Rigid Pole"

Imagine you are a radio station trying to broadcast a message to several different groups of people scattered across a large park.

The Old Way (Fixed Antennas): You have a giant, rigid metal pole with antennas stuck in fixed positions. You can't move them. If the wind blows or people move, the signal gets blocked or weak. It's like trying to shine a flashlight through a keyhole; if the hole doesn't line up with the door, the light doesn't get through.
The New Way (Pinching Antennas - PASS): Imagine a long, flexible garden hose (a waveguide) running along the ground. Along this hose, you have tiny "pinch points" (the antennas) that can pop out at any spot along the hose to release the signal. You can slide these pinch points anywhere to get the best angle. This is the Pinching-Antenna System (PASS). It's like having a flexible hose that can hug the terrain to deliver water (data) exactly where it's needed, bypassing obstacles.

The Challenge: The "Group Chat" Problem

The paper tackles a specific problem: Multigroup Multicast.
Imagine you have three different groups of people in the park:

Group A wants to watch a soccer game.
Group B wants to watch a cooking show.
Group C wants to listen to a podcast.

They are all in the same park, and you only have one hose (one signal source). How do you send three different shows to three different groups without them getting mixed up, while making sure everyone gets a good signal, even the person standing furthest away?

The authors tested three different "rules" (Multiple Access Schemes) to manage this chaos:

The Three Strategies (The "Rules of the Road")

1. TIN (Treating Interference as Noise)

The Analogy: The "Loud Party" Approach.
Everyone shouts their message at the same time. The people in Group A try to hear the soccer game, but they also hear the cooking show and the podcast. They just treat the other voices as annoying background noise and try to tune them out.

Pros: It's simple. You don't need complex equipment at the receiver.
Cons: If the park is crowded (high power), the noise becomes so loud that no one can hear anything clearly. It hits a "ceiling" where making the signal louder doesn't help because the interference is too strong.
Verdict: Good for simple, low-power situations, but hits a wall quickly.

2. NOMA (Non-Orthogonal Multiple Access)

The Analogy: The "Layered Cake" or "Russian Doll" Approach.
You stack the signals on top of each other.

Group A (the "strong" group) gets the whole cake.
Group B (the "medium" group) gets the cake minus the top layer.
Group C (the "weak" group) gets the bottom layer.
How it works: The strong group listens to the whole mess, figures out what the other groups are saying, and "cancels out" (eats) the layers they don't need to hear their own message. The weak group just listens to the bottom layer.
Pros: Very efficient. It uses the same time and space for everyone.
Cons: It requires the receivers to be smart enough to do the "cancellation" math. If the groups are very different (some far away, some close), this works surprisingly well because the "strong" group can easily cancel the "weak" group's signal.

3. TDMA (Time-Division Multiple Access)

The Analogy: The "Relay Race" or "Time Slots."
You don't shout at the same time. You give each group a specific time slot.

1:00 PM: Only Group A talks.
1:05 PM: Only Group B talks.
1:10 PM: Only Group C talks.
The Twist (Two Protocols):
- PS (Pinching Switching): The "Flexible Hose" version. Between 1:00 and 1:05, you physically slide the pinch points on the hose to the perfect spot for Group A. Then, you slide them again for Group B. This is the champion of performance because it eliminates interference completely and optimizes the position for every single group.
- PM (Pinching Multiplexing): The "Fixed Hose" version. You pick one spot for the pinch points and keep them there for everyone. It's easier to build (no moving parts), but it's a compromise.

The Secret Sauce: Moving the Antennas

The most important discovery in this paper isn't just which rule is best, but how the "Pinching Antennas" move.

In old systems, antennas are stuck in a grid. In this new system, the antennas are like sliders on a curtain rod.

The computer calculates exactly where to slide the antennas to make the signal strongest for the person with the weakest connection in each group.
The Goal: "Max-Min Fairness." The system doesn't care about the person with the great signal; it cares about the person in the corner with the bad signal. It moves the antennas to help that person, ensuring everyone gets a fair share.

The Results: Who Won?

The Gold Medalist (TDMA-PS):
If you can afford the mechanical engineering to slide the antennas around between time slots, this is the winner. It eliminates all interference and finds the perfect spot for every group. It's like having a personal spotlight for every audience member.
The Smart Runner-Up (NOMA):
If you can't move the antennas quickly, NOMA is the best choice. It's surprisingly powerful, especially when the groups are scattered in different directions. It uses "channel disparity" (the fact that some people are closer than others) as a superpower to cancel out interference.
The "Good Enough" Option (TIN):
If you have cheap hardware and low power, just treat the noise as noise. It's simple, but it won't win a race against the others in a crowded stadium.
The Comparison to Old Tech:
The paper proves that this "Pinching Antenna" system (PASS) crushes traditional fixed antennas. Why? Because a fixed antenna is like a lighthouse that can't turn its head. The PASS system is like a swarm of fireflies that can move to wherever the light is needed most, bypassing buildings and trees to create a direct line of sight.

Summary in One Sentence

This paper shows that by using a flexible "hose" of antennas that can slide to perfect positions, we can send different messages to different groups of people much more fairly and efficiently than ever before, especially if we use a "relay race" (TDMA) or a "layered cake" (NOMA) strategy to manage the traffic.

1. Problem Statement

The paper addresses the challenge of designing efficient Multiple Access (MA) schemes for Multigroup Multicast communications using Pinching-Antenna Systems (PASS).

Context: PASS is a flexible-antenna paradigm where "pinching antennas" (PAs) are deployed along a low-loss dielectric waveguide. Unlike traditional movable antennas confined to small regions, PASS can span large distances to establish strong Line-of-Sight (LoS) links, effectively mitigating large-scale path loss.
Challenge: In a multigroup multicast scenario, a single Radio Frequency (RF) chain feeds the entire waveguide, transmitting a common signal stream. This architecture limits spatial multiplexing degrees of freedom (DoFs). When serving multiple groups simultaneously, inter-group interference becomes a dominant bottleneck.
Objective: The goal is to maximize Max-Min Fairness (MMF)—ensuring the worst-performing user across all groups achieves the highest possible data rate—by jointly optimizing:
1. PA Placement: The physical locations of the antennas along the waveguide.
2. Resource Allocation: Power and time distribution among groups.

2. Methodology

The authors propose and analyze three distinct MA schemes to manage interference, developing specific optimization algorithms for each:

A. System Model

Architecture: A base station with a single waveguide of length $D_x$ and $N$ PAs.
Channel: Free-space LoS model dominates (high-frequency bands). The effective channel depends on the PA locations ( $x$ ).
Optimization Problem: A non-convex, non-smooth joint optimization problem maximizing the minimum group rate subject to power and location constraints.

B. Proposed Schemes & Algorithms

1. Treating Interference as Noise (TIN)

Concept: Groups transmit simultaneously; receivers treat interference from other groups as noise.
Methodology:
- Power Allocation: Derived a closed-form solution for optimal power allocation based on the bottleneck channel gain of each group.
- PA Placement: Developed a Sequential Element-wise Optimization (SEO) algorithm. It iteratively updates one PA location at a time to maximize the harmonic mean of channel gains, using a grid search over discretized waveguide points.
Insight: In the high-SNR regime, performance is bounded by interference, making PA optimization less effective.

2. Non-Orthogonal Multiple Access (NOMA)

Concept: Groups share time and frequency but are separated in the power domain. Receivers use Successive Interference Cancellation (SIC) based on a decoding order determined by channel conditions.
Methodology:
- Two-Group Case: Derived an analytical closed-form solution for optimal power allocation.
- General Multigroup Case: Proposed a recursive power allocation framework using bisection search to find the optimal MMF rate.
- PA Placement: Introduced a Hierarchical Objective Evaluation (HOE) mechanism combined with SEO.
  - Stage I: Quickly discard candidate PA positions using a theoretical upper bound.
  - Stage II: Only evaluate promising candidates with the full, computationally expensive bisection search.
Insight: NOMA leverages channel disparity; in high-SNR regimes, it benefits from optimizing the strong group's channel to support high-order SIC.

3. Time-Division Multiple Access (TDMA)

Concept: Groups are served in orthogonal time slots, eliminating inter-group interference. Two protocols are analyzed:
- Pinching Switching (PS): PA locations can be reconfigured for each time slot (group-specific).
- Pinching Multiplexing (PM): A single PA configuration is shared across all time slots.
Methodology:
- TDMA-PS: The problem is decoupled. PA locations are optimized independently for each group (SEO), followed by a convex optimization for time and power allocation.
- TDMA-PM: PA locations and resources are jointly optimized. The authors use an HOE-enhanced SEO where the inner loop solves a resource allocation problem using Karush-Kuhn-Tucker (KKT) conditions and bisection search.
Insight: TDMA-PS offers the best performance (interference-free + full flexibility) but requires complex mechanical switching. TDMA-PM is a practical compromise.

3. Key Contributions

Framework Development: Established a comprehensive MA framework for PASS-enabled multigroup multicast, explicitly characterizing the trade-offs between TIN, NOMA, and TDMA under MMF criteria.
Algorithmic Innovations:
- Derived closed-form solutions for power allocation in TIN and specific NOMA/TDMA scenarios.
- Developed HOE-enhanced SEO algorithms to efficiently solve the non-convex PA placement problems for NOMA and TDMA-PM, significantly reducing computational complexity.
Theoretical Insights:
- Identified that in low-SNR regimes, all schemes share a common optimization principle: maximizing the harmonic mean of group channel gains.
- Proved that in high-SNR regimes, the optimal strategy diverges: TIN is interference-limited, while NOMA exploits channel disparity for SIC.
Performance Bounds: Established analytical upper bounds for TIN and demonstrated the theoretical limits of PASS in high-power scenarios.

4. Numerical Results

Extensive simulations compared the proposed PASS schemes against traditional Fixed-Array Systems:

PASS vs. Fixed-Array: PASS consistently outperforms fixed arrays, especially as the service area ( $D_x$ ) increases. PASS mitigates large-scale path loss by physically moving antennas closer to users, whereas fixed arrays suffer from increased average distance.
Scheme Comparison:
- TDMA-PS: Achieves the highest performance by eliminating interference and fully utilizing spatial reconfigurability.
- NOMA: Consistently outperforms TDMA-PM. In high-power regimes with heterogeneous (spatially disjoint) group distributions, NOMA can even surpass TDMA-PS by effectively exploiting channel disparity.
- TIN: Serves as a practical lower bound. While it has the lowest complexity, its performance saturates in high-SNR due to interference.
Convergence: The proposed SEO algorithms converge rapidly (within a few iterations), making them suitable for real-time applications.

5. Significance

This paper is significant for several reasons:

Advancing Flexible Antenna Tech: It extends the application of Pinching-Antenna Systems from unicast to complex multigroup multicast scenarios, addressing a critical gap in current literature.
Practical MA Design: It provides a comparative analysis of MA strategies tailored to the unique constraints of a single-RF-chain waveguide, offering design guidelines for future 6G systems.
Fairness Optimization: By focusing on Max-Min Fairness, the work ensures that the system benefits the most disadvantaged users, a crucial requirement for equitable wireless services.
Hardware-Software Co-design: The study highlights the trade-off between hardware complexity (mechanical switching in TDMA-PS) and algorithmic complexity (NOMA/TDMA-PM), guiding practical system implementation.

In conclusion, the paper demonstrates that PASS is a superior architecture for multicast communications, capable of significantly outperforming traditional fixed arrays, with the optimal MA scheme depending on the specific power regime and user distribution.