⚡ electrical engineering

Spacing-Based Coupling Radiation Control in Pinching-Antennas Systems for Heterogeneous NOMA Users

This paper proposes a spacing-based coupling radiation control mechanism for pinching-antenna systems to optimize semantic spectral efficiency for heterogeneous NOMA users by jointly adjusting antenna positions and power allocation while satisfying bit-user QoS and interference constraints.

Original authors: Ishtiaque Ahmed, Leila Musavian

Published 2026-05-07

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ishtiaque Ahmed, Leila Musavian

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

Imagine a high-speed internet highway (the wireless network) that usually gets clogged because everyone is trying to drive at the same time. Now, imagine a new kind of traffic controller that doesn't just tell cars when to go, but can actually reshape the road itself to create a perfect, straight path for every driver.

This is the core idea behind the paper "Spacing-Based Coupling Radiation Control in Pinching-Antennas Systems for Heterogeneous NOMA Users."

Here is a breakdown of what the researchers did, using simple analogies:

1. The New Hardware: The "Clothespin" Highway

Traditional cell towers use big, fixed antennas. This paper introduces Pinching-Antenna Systems (PASS).

The Analogy: Imagine a long, clear plastic tube (a dielectric waveguide) carrying a signal like water flowing through a pipe. Instead of having one giant faucet at the end, you have a series of cheap "clothespins" clipped onto the side of the tube.
How it works: When you "pinch" the tube with a clothespin, some of the signal leaks out into the air to reach your phone. The magic is that these clothespins are reconfigurable. You can slide them closer or further apart, and you can adjust how tightly they pinch the tube. This changes how much signal leaks out at each spot, creating a custom "Line-of-Sight" path directly to the user.

2. The Problem: Two Different Types of Passengers

The researchers are trying to serve two very different types of users at the same time using the same signal:

The "Bit" User: This is your standard phone or computer. It needs to receive data exactly as sent (like a text message or a video stream). It's like a passenger who needs to hear every single word of a speech perfectly.
The "Semantic" User: This is a new, AI-driven type of communication. Instead of sending every single bit of data, the system sends the meaning or the "gist" of the message. It's like a passenger who only needs to understand the main idea of the speech, even if they miss a few words. This is more efficient but requires a specific decoding method (like a neural network).

The challenge is that these two users have different needs. If you give too much power to the "Bit" user, the "Semantic" user might get a weak signal. If you give too much to the "Semantic" user, the "Bit" user might miss their data.

3. The Solution: The "Adjustable Pinch"

The paper proposes a system where the "clothespins" (antennas) can change their strength based on how far apart they are spaced.

The Analogy: Think of the signal as water flowing through the tube. By moving the clothespins closer together or further apart, the researchers can control exactly how much water (signal power) sprays out at each point.
The Strategy: They use a technique called NOMA (Non-Orthogonal Multiple Access). This is like a radio station broadcasting two different songs at once, but one is louder than the other. The "Bit" user listens to the loud song and ignores the quiet one. The "Semantic" user listens to the quiet song, but first, they use a special filter (Successive Interference Cancellation) to remove the loud song so they can hear the quiet one clearly.

4. The Optimization: Finding the Perfect Spot

The researchers didn't just guess where to put the clothespins or how loud to make the signal. They created a mathematical "game" to find the best setup.

The Goal: Maximize the "Semantic Spectral Efficiency." In plain English: Get the most "meaning" delivered to the Semantic user without dropping the "Bit" user's connection.
The Method: They used a step-by-step computer algorithm (Alternating Optimization).
1. Step A: Keep the clothespin positions fixed and figure out the perfect power split between the two users.
2. Step B: Keep the power split fixed and move the clothespins to the perfect spots along the tube to boost the signal.
3. Repeat: Do this over and over until they find the absolute best combination.

5. The Results: Why It's Better

The researchers ran simulations to see how this new system compares to old, fixed-antenna systems.

The Finding: The "adjustable pinch" system (Proportional Power PASS) consistently outperformed the old systems.
The Analogy: Imagine a fixed antenna is like a sprinkler that sprays water in a fixed circle. If you stand in the wrong spot, you get wet; if you stand in the right spot, you get soaked. The new Pinching-Antenna system is like a smart sprinkler that can aim a specific stream of water exactly where you are standing, adjusting the pressure based on how far away you are.
Key Takeaway: By adjusting the spacing of the antennas, the system can deliver more "meaning" (Semantic data) to the AI users while still keeping the regular users happy, even in tricky environments.

Summary

This paper presents a smarter way to build wireless networks. Instead of using heavy, fixed antennas, it uses a flexible tube with movable "clothespins." By sliding these clothespins and adjusting how tightly they pinch, the system can perfectly tailor the signal for both standard users and next-generation AI users, delivering more useful information with less waste.

Technical Summary: Spacing-Based Coupling Radiation Control in Pinching-Antennas Systems for Heterogeneous NOMA Users

Problem Statement
Pinching-antennas systems (PASS) have emerged as a revolutionary architecture for 6G, utilizing low-cost dielectric waveguides and "clothespin" antennas to create reconfigurable line-of-sight (LoS) links. While existing PASS research has focused on bit-based communication with uniform power distribution or equal-power pinching antennas, there is a significant gap in exploring flexible radiation control for heterogeneous networks. Specifically, the potential of PASS to support simultaneous semantic communication (SC) and traditional bit communication via Non-Orthogonal Multiple Access (NOMA) remains unexplored. Current models often treat PASS similarly to conventional MIMO with uniform power, failing to leverage the unique physical property of PASS: the ability to control coupling strength and radiation power by adjusting the physical spacing between the waveguide and the pinching antennas.

Methodology
The paper proposes a downlink NOMA-assisted framework serving a heterogeneous pair of users: a semantic user (S) and a bit user (B). The core innovation is an adjustable power PASS model where the radiated power ratio of each antenna is governed by the coupling spacing ( $S_n$ ) between the dielectric waveguide and the antenna.

System Model:
- Radiation Control: The coupling coefficient $\kappa_n$ is modeled using coupled-mode theory (CMT), where the radiated power ratio $\beta_n$ is a function of the spacing $S_n$ . This allows for proportional power distribution across antennas rather than equal power.
- Channel Model: The channel is dominated by LoS, modeled via a spherical wavefront. The transmitted signal undergoes phase shifts due to propagation within the dielectric waveguide ( $\lambda_g$ ) before radiating into free space.
- NOMA Decoding: The system employs a "bit-to-semantic" decoding order. The bit user decodes its signal treating the semantic signal as interference. The semantic user employs Successive Interference Cancellation (SIC) to remove the bit user's signal before decoding its own message.
Optimization Problem:
The objective is to maximize the Semantic Spectral Efficiency (SE) for User S while guaranteeing the Quality of Service (QoS) for User B. The optimization variables are:
- Power Allocation ( $p_S$ ): The fraction of total power allocated to the semantic user.
- Antenna Positions ( $x^P$ ): The $x$ -coordinates of the $N$ pinching antennas along the waveguide.
The problem is subject to constraints including minimum adjacent antenna spacing ( $\Delta \geq \lambda/2$ ), bit-user QoS requirements, SIC feasibility, and total power limits. The objective function involves a sigmoid-shaped semantic similarity function $\epsilon_K(\gamma)$ , making the problem non-convex.
Solution Approach:
An Alternating Optimization (AO) algorithm is proposed to solve the non-convex problem by decoupling it into two subproblems:
- Power Allocation Subproblem: With fixed antenna positions, a closed-form solution for the optimal power coefficient is derived based on active constraints (QoS and SIC feasibility).
- Antenna Position Subproblem: With fixed power allocation, the antenna positions are optimized. This involves a two-step iterative process:
  - Large-scale placement: A bisection search places antennas to bias the geometry for enhanced channel gain for the semantic user while maintaining bit-user QoS.
  - Fine-scale alignment: Small adjustments are made to ensure phase alignment for both users, maximizing constructive interference.

Key Contributions

Adjustable Power PASS Model: The authors formalize a downlink model where unequal power radiation is dynamically controlled by the physical spacing between the waveguide and pinching antennas, moving beyond the uniform power assumption of prior works.
Heterogeneous NOMA Framework: The paper integrates SC and bit communication within a PASS-NOMA framework, justifying the bit-to-semantic decoding order and addressing the specific resource allocation challenges posed by neural-network-based semantic extraction.
Optimization Algorithm: A low-complexity AO approach is developed to jointly optimize power allocation and antenna locations, incorporating phase alignment constraints to handle the spherical wavefront and waveguide propagation effects.
Performance Validation: Numerical results validate that the proposed proportional power PASS significantly outperforms conventional fixed-antenna systems (CAS) and equal-power PASS schemes.

Simulation Results
Simulations conducted at 28 GHz with various parameters ( $N=3, 7$ ; $D=20, 40$ m) demonstrate:

Semantic SE Gains: The proportional power PASS achieves higher semantic SE compared to equal-power PASS and CAS, particularly as transmit power ( $P_{max}$ ) increases.
Phase Sensitivity: Performance is highly sensitive to phase alignment accuracy. Fine alignment for the semantic user ( $\delta_S$ ) yields the best results, while coarse alignment significantly degrades performance.
Robustness: The proposed system exhibits lower outage probability for bit-user QoS and SIC feasibility compared to CAS, attributed to the flexibility of repositioning antennas.
Geometric Impact: Smaller coverage areas and specific distance ratios between users yield better performance. Increasing the number of antennas ( $N=7$ vs $N=3$ ) distributes the radiation profile, which can impact feasibility under high bit-user QoS requirements but generally improves performance.

Significance and Claims
The paper claims that by exploiting the physical flexibility of PASS—specifically the ability to tune radiation ratios via antenna spacing—it is possible to effectively serve heterogeneous users with distinct communication needs (semantic vs. bit). The work demonstrates that flexible radiation control is a critical enabler for next-generation semantic communication systems. The authors conclude that their approach provides a clear performance gain over conventional benchmarks and suggests that future work could explore amplitude tunability to further enhance reconfigurable heterogeneous communication. The study positions PASS not just as a hardware innovation, but as a controllable resource for optimizing complex, mixed-service wireless networks.