Near-Field Modelling and Performance Analysis of Modular Extremely Large-Scale Array Communications

Xinrui Li, Haiquan Lu, Yong Zeng, Shi Jin, Rui Zhang

Introduction

As 6G wireless communication evolves, the demand for higher capacity is driving the development of Extremely Large-Scale MIMO (XL-MIMO). Traditional approaches involve either collocated arrays, which face physical space constraints on single building facades, or Distributed Antenna Systems (DAS), which suffer from high backhaul complexity. This paper introduces a "modular" architecture for extremely large-scale antenna arrays (XL-arrays) that seeks a middle ground, balancing deployment flexibility with processing simplicity.

A critical challenge in XL-arrays is the transition from far-field to near-field propagation. Standard models assume Uniform Plane Waves (UPW), where signals arrive at all antennas with uniform strength and phase. However, for XL-arrays, users are often in the near-field, where signals arrive as Non-Uniform Spherical Waves (NUSW). In this regime, distance, signal strength, and phase vary significantly across the array elements, requiring more sophisticated modelling.

System Model and Methodology

The authors propose the Modular XL-array, where antennas are grouped into discrete modules. These modules are arranged on a platform with an inter-module spacing ($D$) significantly larger than the signal wavelength ($\lambda$).

Modular XL-array architecture featuring sub-modules and inter-module spacing.

To analyze this system, the researchers:

  1. Model the Near-Field: They employ the NUSW model to account for both phase and amplitude variations across thousands of array elements.
  2. Derive Closed-Form SNR: By applying Maximum Ratio Combining (MRC) beamforming, they convert the discrete signal summations into a continuous double integral.
  3. Mathematical Approximation: They solve this integral to provide a manageable closed-form formula (Theorem 1) that predicts Signal-to-Noise Ratio (SNR) based on the array's geometry and the user's spatial coordinates.

The study validates these mathematical derivations through numerical simulations at 2.4 GHz, comparing the modular near-field model against conventional collocated and far-field UPW models.

Key Insights: The Near-Field SNR Ceiling

A primary finding of this research is the existence of an SNR ceiling. Unlike far-field models, which suggest that SNR increases linearly and infinitely with the number of antennas, the near-field model reveals that SNR eventually approaches a constant limit as the array size grows. This occurs because antennas positioned far from the user contribute negligible additional signal power due to increased path loss.

Furthermore, the research demonstrates that:

  • Geometry is Critical: SNR depends heavily on inter-module spacing. Increasing the gap between modules generally reduces the maximum achievable SNR by pushing elements further from the user.
  • UPW Inaccuracy: The standard plane wave assumption (UPW) can significantly miscalculate performance in the near-field, either overestimating or underestimating SNR depending on the user's angular position relative to the array.

Key Results and Validation

The paper provides the first closed-form SNR approximation for modular XL-arrays in the near-field. A key result is the asymptotic scaling law: as the number of modules $N \to \infty$, the SNR converges to a value inversely proportional to the module spacing and user distance.

SNR performance versus the number of array elements, illustrating the near-field SNR ceiling.

As shown in Figure 3, the closed-form approximation (red and blue dashed lines) perfectly matches the exact numerical summation (solid lines). The graph clearly illustrates the saturation of SNR as the number of elements increases, contrasting sharply with the linear growth predicted by the UPW model (magenta line). For extremely large arrays, a modular configuration may see a performance gap of approximately 3 dB compared to a collocated array, representing the trade-off for its deployment advantages.

Conclusion and Future Directions

This work establishes a theoretical foundation for modular XL-arrays in the near-field. While it provides a robust analytical framework, several avenues for future research remain:

  • Multipath Environments: The current model assumes Line-of-Sight (LoS) conditions; future studies should incorporate Non-Line-of-Sight (NLoS) scenarios.
  • Multi-User Scenarios: Extending the analysis to multi-user interference is necessary for practical 6G applications.
  • Hardware Realities: Accounting for hardware impairments like phase noise and calibration errors will be crucial for real-world implementation.

Summary updated

BibTeX
@article{li2022near,
  author = {Xinrui Li and Haiquan Lu and Yong Zeng and Shi Jin and Rui Zhang},
  title = {{Near-Field Modelling and Performance Analysis of Modular Extremely Large-Scale Array Communications}},
  journal = {IEEE Communications Letters},
  year = {2022},
  doi = {10.1109/lcomm.2022.3172437},
}
Figure