Secure Full-Duplex Communication via Movable Antennas

Jingze Ding, Zijian Zhou, Chenbo Wang, Wenyao Li, Lifeng Lin, Bingli Jiao

TL;DR

This paper proposes a novel wireless communication architecture that combines Movable Antennas (MA) with Full-Duplex (FD) technology to enhance physical layer security. By dynamically repositioning antennas at a base station and injecting artificial noise, the authors achieve a significantly higher sum secrecy rate for simultaneous uplink and downlink users compared to traditional fixed-antenna systems.

What Problem Does This Paper Address?

Wireless communication is inherently vulnerable because signals are broadcasted, making them easy for eavesdroppers (Eve) to intercept. Physical Layer Security (PLS) attempts to solve this by using multiple antennas to "steer" signals toward legitimate users and away from Eve.

However, traditional systems use Fixed-Position Antennas (FPAs), which are stuck in one place and cannot fully exploit the spatial variations of the wireless channel. Furthermore, most secure systems operate in Half-Duplex (HD) mode, meaning they can't handle secure uploads and downloads at the same time. The challenge is to maximize the Sum Secrecy Rate (SSR)—the total rate of secure data transmission—while managing the self-interference inherent in full-duplex systems and the complex geometry of moving antennas.

Methods

The authors propose an FD base station equipped with separate transmit and receive MAs. To solve the complex, non-convex optimization problem of maximizing the SSR, they developed an Alternating Optimization (AO) framework that breaks the problem into three iterative subproblems:

  1. Receive Beamforming Optimization: Determining how the base station should process incoming signals from the uplink user to maximize signal quality while suppressing interference.
  2. Transmit Beamforming and Artificial Noise (AN) Optimization: Designing the downlink signal and the "jamming" noise (AN) intended to confuse the eavesdropper. This is solved using Successive Convex Approximation (SCA) and Semidefinite Relaxation (SDR).
  3. Antenna Position Optimization: Finding the best physical coordinates for the MAs within a given region. Because this is a highly non-linear geometric problem, the authors employ Particle Swarm Optimization (PSO), a metaheuristic algorithm that "searches" the physical space for the best antenna layout.

Datasets

As is standard in theoretical wireless communications research, this paper does not use a static "dataset" like those found in computer vision. Instead, it uses a field-response based channel model to simulate realistic wireless environments.

  • Simulation Parameters: The authors assume a carrier frequency where the moving region is $2\lambda \times 2\lambda$ (where $\lambda$ is the wavelength).
  • Environment: Legitimate users and an eavesdropper are randomly distributed within a 50-meter radius of the base station.
  • Benchmarks: The proposed "MA-FD-PSO" scheme is compared against several baselines, including fixed-antenna systems (FPA-FD) and half-duplex systems (MA-HD-PSO).

Key Insights and Ideas

  • Exploiting Spatial Degrees of Freedom: By moving an antenna just a few centimeters (fractions of a wavelength), the system can find a "local sweet spot" where the legitimate channel is strong and the eavesdropping channel is weak.
  • The Synergy of AN and FD: In a full-duplex setup, Artificial Noise serves a dual purpose: it jams the eavesdropper and provides a degree of freedom to balance the interference between the simultaneous uplink and downlink.
  • Antenna "Saving": One of the most striking insights is that a few movable antennas can outperform a much larger array of fixed antennas. The paper shows that an MA system can achieve the same security performance as an FPA system while using significantly fewer antennas.

Key Results

The simulation results demonstrate the clear superiority of the movable antenna approach:

  • Performance Gains: The proposed MA-FD-PSO scheme achieved a 9.14 bps/Hz gain from the full-duplex capability and a 3.69 bps/Hz gain specifically from the ability to move the antennas, compared to benchmark schemes.
  • Efficiency: In one scenario, the MA-assisted system achieved a secrecy threshold with only 2 antennas that a fixed-antenna system required 8 antennas to reach.
  • Robustness: The system maintains high secrecy rates even when self-interference cancellation (the ability of the FD station to ignore its own transmission) is not perfect.

Limitations and Future Work

  • Perfect CSI Assumption: The paper assumes the base station has "Perfect Channel State Information" (CSI) for all parties, including the eavesdropper. In real-world scenarios, Eve is usually passive and her channel is difficult to estimate.
  • Computational Complexity: The use of PSO and iterative AO is computationally expensive. Implementing this in real-time as users move would require highly optimized hardware or faster heuristic algorithms.
  • Single Eavesdropper: The current model focuses on a single-antenna eavesdropper. Future work could explore scenarios with multiple, multi-antenna eavesdroppers or "colluding" attackers.
  • Hardware Constraints: The paper mentions that balancing the trade-off between the number of MAs and the difficulty of maintaining minimum inter-antenna distances during movement is a potential issue for future implementation.

Summary updated

Abstract

This paper investigates physical layer security (PLS) in a movable antenna (MA)-assisted full-duplex (FD) system. In this system, an FD base station (BS) with multiple MAs for transmission and reception provides services for an uplink (UL) user and a downlink (DL) user. Each user operates in half-duplex (HD) mode and is equipped with a single fixed-position antenna (FPA), in the presence of a single-FPA eavesdropper (Eve). To ensure secure communication, artificial noise (AN) is transmitted to obstruct the interception of Eve. The objective of this paper is to maximize the sum secrecy rate (SSR) of the UL and DL users by jointly optimizing the beamformers of the BS and the positions of MAs. This paper also proposes an alternating optimization (AO) method to address the non-convex problem, which decomposes the optimization problem into three subproblems and solves them iteratively. Simulation results demonstrate a significant performance gain in the SSR achieved by the proposed scheme compared to the benchmark schemes.
BibTeX
@misc{ding2024secure,
  author = {Jingze Ding and Zijian Zhou and Chenbo Wang and Wenyao Li and Lifeng Lin and Bingli Jiao},
  title = {{Secure Full-Duplex Communication via Movable Antennas}},
  year = {2024},
  eprint = {2403.20025},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2403.20025},
}
Figure