Introduction
As wireless communications push toward mmWave and THz frequencies, traditional fixed-antenna systems face severe path loss and frequent blockages. Pinching-Antenna Systems (PASS) offer a dynamic alternative, utilizing movable antennas along dielectric waveguides to maintain ideal Line-of-Sight (LoS) links. While promising, the integration of PASS with wideband Orthogonal Frequency Division Multiplexing (OFDM) introduces complex frequency-selective challenges. This paper presents the first rigorous model to account for waveguide dispersion and coupling variability in wideband OFDM-PASS.
System Architecture and Modeling
The proposed system integrates an Access Point (AP) with a dielectric waveguide. "Pinching" antennas (PAs) are positioned along this waveguide to serve users. Unlike traditional systems where signal loss follows a logarithmic decay in free space, PASS signals propagate through a waveguide with linear attenuation (measured in dB per meter).
To model this accurately, the authors employ:
- Waveguide Theory: Capturing frequency-dependent attenuation and phase constants for the TE10 mode.
- Coupled-Mode Theory (CMT): Modeling the frequency-selective energy transfer from the waveguide to the PAs, accounting for phase mismatches across subcarriers.
- Group Delay Analysis: Determining the delay spread caused by waveguide dispersion, which is critical for calculating the necessary Cyclic Prefix (CP) length.
The "Dispersion Tax" and Cutoff Effects
A primary finding is the impact of the waveguide's cutoff frequency ($f_0$). Below $f_0$, signals cannot propagate. Near $f_0$, the group velocity varies significantly across the bandwidth, leading to a high "dispersion tax." This necessitates a larger CP overhead—sometimes exceeding 10% of the symbol time—to prevent inter-symbol interference, thereby reducing spectral efficiency.
Performance Comparison: Linear vs. Logarithmic Loss
The most significant advantage of PASS is its loss characteristic. In traditional fixed-antenna setups, signal strength decays logarithmically with distance. In contrast, the waveguide-based PASS experiences linear loss.
As shown in the performance analysis, there is a distinct "crossover point" (approximately 150 meters in simulated mmWave scenarios). Beyond this distance, the gentle linear loss of a practical PASS system provides superior achievable rates compared to the steep logarithmic decline of traditional LoS or NLoS links.
Conclusion
OFDM-PASS represents a viable path for long-range high-frequency communications, provided the "dispersion tax" is managed. Future work will focus on "frequency-aware" antenna placement to mitigate phase misalignment and optimize performance across the entire OFDM bandwidth.
Reference: Xiao, J., et al. "Frequency-Selective Modeling and Analysis for OFDM-Integrated Wideband Pinching-Antenna Systems." arXiv:2505.20636v4 [cs.IT], 2025.
Summary updated