Pinching-Antenna Assisted Sensing: A Bayesian Cramér-Rao Bound Perspective

Hao Jiang, Chongjun Ouyang, Zhaolin Wang, Yuanwei Liu, Arumugam Nallanathan, Zhiguo DingarXiv 2025 · DOI · arXiv

TL;DR

This paper explores the fundamental sensing limits of Pinching-Antenna Systems (PASS)—a novel technology using long waveguides and movable antennas to create massive, flexible apertures. By applying a Bayesian Cramér-Rao Bound (BCRB) framework, the authors demonstrate that PASS significantly outperforms traditional fixed arrays and reveal a counterintuitive "mismatch" where the best place to put an antenna is not directly over the target's most likely location.

What Problem Does This Paper Address?

In the quest for 6G, "Integrated Sensing and Communication" (ISAC) is a major pillar. While technologies like Movable Antennas (MA) exist, they are usually limited to small-scale adjustments (a few wavelengths). PASS overcomes this by using long, low-attenuation waveguides (tens of meters) with "pinching antennas" (PAs) that can be repositioned to create large-scale reconfigurability.

However, optimizing these systems for sensing is tricky. Traditional performance metrics like the Cramér-Rao Bound (CRB) require knowing the target's exact location to calculate the bound—which is a bit of a "chicken and egg" problem if you are trying to sense that location. This paper uses the Bayesian CRB (BCRB), which incorporates prior statistical knowledge (where the target is likely to be) to provide a more practical and robust optimization objective for system design.

Methods

The authors analyze an uplink sensing scenario where targets transmit pilots to a PASS receiver. Their approach involves several key technical steps:

  • Statistical Modeling: They model target locations using Gaussian Mixture Models (GMMs), which can approximate almost any continuous probability distribution.
  • Numerical Integration: To handle the complex integrals required for the BCRB, they employ Gauss-Hermite Quadrature (GHQ), turning intractable math into a manageable summation.
  • Optimization Protocols: They propose two scheduling methods for multi-target sensing:
  • Pinch Switching (PS): Reconfiguring PA positions for every time slot (high performance, high overhead).
  • Pinch Multiplexing (PM): Using one fixed PA layout for all targets across all slots (lower complexity).
  • Algorithmic Solution: They use Karush-Kuhn-Tucker (KKT) conditions to solve for optimal power allocation and an element-wise 1D search algorithm to find the best PA positions along the waveguide.

Datasets

This is a theoretical and simulation-based study; no physical datasets were used. The evaluation is conducted via Monte Carlo simulations with the following parameters:

  • Carrier Frequency: 28 GHz (mmWave).
  • Waveguide Length: 10 meters.
  • Benchmarks: The proposed PASS algorithms are compared against Uniform PASS (evenly spaced), Centered PASS (clustered at the target mean), and Conventional Fixed-Position Arrays (FPA).

Key Insights and Ideas

  • The Centroid Mismatch: Perhaps the most surprising finding is that the optimal PA position does not align with the center of the target's location distribution. Because sensing in the near-field relies on the curvature of spherical waves, placing an antenna slightly offset (by a distance $\Delta$) from the target's expected center actually maximizes phase sensitivity.
  • Large-Scale Reconfigurability: Unlike traditional arrays that are "stuck" in one spot, PASS can move its "eyes" across meters of waveguide. This allows the system to drastically reduce path loss and improve resolution by physically moving antennas closer to the sensing region.
  • Protocol Trade-offs: Pinch Switching (PS) is superior because it tailors the "aperture" to each specific target. However, as the number of targets ($K$) increases, the computational cost of reconfiguring the PAs for every slot becomes a significant factor.

Key Results

  • Superiority over FPA: PASS consistently achieves a much lower root-BCRB compared to conventional fixed arrays, especially as the waveguide length increases.
  • Power Efficiency: To achieve the same sensing accuracy (BCRB threshold), PASS requires significantly less transmit power than fixed-position benchmarks.
  • Scalability: While fixed arrays struggle as targets spread out over a larger area, PASS maintains high performance because it can adapt its geometry to the expanded "service area."
  • Algorithm Convergence: The proposed element-wise search algorithm is shown to converge quickly, making it a viable candidate for real-time antenna repositioning.

Limitations and Future Work

  • Line-of-Sight (LoS) Focus: The study assumes a dominant LoS environment, which is common for high-frequency mmWave/THz bands but ignores complex multipath scattering found in cluttered indoor environments.
  • Single Waveguide/RF Chain: The model focuses on a single waveguide. Future work could explore multi-waveguide PASS architectures to enable 3D localization and interference cancellation between multiple RF chains.
  • Hardware Latency: The paper assumes PAs can be repositioned efficiently. In practice, the mechanical or electronic latency of moving "pinching" points needs to be accounted for in the scheduling protocols.

Summary updated

Abstract

The fundamental sensing limit of pinching-antenna systems (PASS) is studied from a Bayesian Cramér-Rao bound (BCRB) perspective. Compared to conventional CRB, BCRB is independent of the exact values of sensing parameters and is not restricted by the unbiasedness of the estimator, thus offering a practical and comprehensive lower bound for evaluating sensing performance. A system where multiple targets transmit uplink pilots to a single-waveguide PASS under a time-division multiple access (TDMA) scheme is analyzed. For the single-target scenario, our analysis reveals a unique mismatch between the sensing centroid (i.e., the optimal PA position) and the distribution centroid (i.e., the center of the target's prior distribution), underscoring the necessity of dynamic PA repositioning. For the multi-target scenario, two target scheduling protocols are proposed: 1) pinch switching (PS), which performs separate pinching beamforming for each time slot, and 2) pinch multiplexing (PM), which applies a single beamforming configuration across all slots. Based on these protocols, both the total power minimization problem under a BCRB threshold and the min-max BCRB problem under a total power constraint are formulated. By leveraging Karush-Kuhn-Tucker (KKT) conditions, these problems are equivalently converted into a search over PA positions and solved using an element-wise algorithm. Numerical results show that i)~PASS, endowed with large-scale reconfigurability, can significantly enhance the sensing performance compared with conventional fixed-position arrays, and ii)~PS provides more robust performances than PM at the cost of higher computational complexity.
BibTeX
@misc{jiang2025pinching,
  author = {Hao Jiang and Chongjun Ouyang and Zhaolin Wang and Yuanwei Liu and Arumugam Nallanathan and Zhiguo Ding},
  title = {{Pinching-Antenna Assisted Sensing: A Bayesian Cramér-Rao Bound Perspective}},
  year = {2025},
  eprint = {2510.09137},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2510.09137},
  howpublished = {arXiv},
}
Figure