Reconfigurable Intelligent Surface-Aided Full-Duplex mmWave MIMO: Channel Estimation, Passive and Hybrid Beamforming

Yue Xiu, Songjie Yang, Wanting Lyu, Yunis Xanthos, Zhongpei Zhang, Chadi Assi, Chau YuenIEEE Trans. Wireless Communications 2024 · DOI

TL;DR

This paper tackles the challenge of boosting wireless communication capacity by combining full-duplex (FD) millimeter-wave (mmWave) MIMO with reconfigurable intelligent surfaces (RIS). They propose novel methods for joint uplink-downlink channel estimation and beamforming optimization, demonstrating significant improvements in spectral efficiency and self-interference cancellation, especially in complex mmWave environments.

What Problem Does This Paper Address?

The demand for higher wireless data rates is exploding, and millimeter-wave (mmWave) frequencies offer abundant bandwidth to meet this need. Full-duplex (FD) technology promises to double spectral efficiency by allowing simultaneous transmission and reception on the same frequency. However, FD systems face two major hurdles: self-interference (SI) from the transmitter to the receiver, and the inherent limitations of mmWave, such as limited coverage and susceptibility to blockages. Reconfigurable intelligent surfaces (RIS) offer a promising solution to enhance mmWave communication reliability and coverage by intelligently reflecting signals. This paper aims to integrate RIS into FD mmWave MIMO systems to overcome SI and improve overall performance, focusing on efficient channel estimation and beamforming strategies.

Methods

The authors introduce several novel techniques:

  • Angular-Domain Reciprocity: They demonstrate that FD antenna arrays exhibit angular-domain reciprocity under far-field planar wavefront assumptions. This property is crucial for sharing channel information between uplink (UL) and downlink (DL) transmissions, reducing estimation complexity.
  • Joint DL-UL Channel Estimation:
    • SI Channel Estimation: A Khatri-Rao product-based compressive sensing (KR-CS) framework is proposed for estimating the SI channel, offering improved performance and speed over Kronecker CS.
  • Direct Channel Estimation: A distributed CS (D-CS) framework is developed for jointly estimating DL and UL channels.
  • Cascaded Channel Estimation: A two-stage multiple measurement vector-based D-CS (M-D-CS) framework is presented for estimating the channels involving the RIS.
  • Passive Beamforming: An optimization solution is proposed to maximize the cascaded channel capacity by leveraging the angular-domain cascaded channel.
  • Hybrid Beamforming and SIC: A novel hybrid weighted minimum mean squared error (H-WMMSE-SIC) method is introduced for SI cancellation. This method iteratively optimizes analog and digital beamforming components within the WMMSE framework, aiming to maximize spatial multiplex gain while minimizing SI.

Datasets

Not discussed in the paper. The paper focuses on theoretical development and simulation-based evaluation.

Key Insights and Ideas

  • Angular-Domain Reciprocity is Key: The paper highlights that exploiting the angular-domain reciprocity of FD arrays is fundamental to achieving efficient joint DL-UL channel estimation and reducing training overhead. This is a more accurate assumption than simple channel reciprocity in FD systems.
  • RIS as a Game Changer for FD mmWave: Integrating RIS into FD mmWave systems is shown to be a powerful strategy for enhancing both coverage and reliability, effectively mitigating the limitations of mmWave and improving the overall system performance.
  • Joint Estimation Outperforms Separate: Jointly estimating DL-UL channels, rather than separately, leads to significantly improved estimation performance, especially when leveraging the angular-domain reciprocity.
  • H-WMMSE-SIC Balances Performance and Complexity: The proposed H-WMMSE-SIC method offers a compelling trade-off, achieving spectral efficiency comparable to fully-digital WMMSE-SIC, particularly at lower interference-to-noise ratios, while being more computationally tractable for hybrid beamforming architectures.
  • Off-Grid Refinement is Crucial: Addressing basis mismatch issues through off-grid refinement techniques significantly improves the accuracy of channel estimation, especially for complex mmWave environments.

Key Results

  • The proposed joint DL-UL channel estimation methods show significant performance improvements compared to separate DL/UL estimation.
  • The KR-CS framework for SI channel estimation is faster and performs better than K-CS.
  • The H-WMMSE-SIC method achieves spectral efficiency comparable to fully-digital WMMSE-SIC, especially when the interference-to-noise ratio is below 35 dB.
  • Simulations demonstrate that the proposed methods outperform conventional estimation techniques that do not consider angular-domain reciprocity.
  • The proposed H-WMMSE-SIC method shows excellent performance in terms of SI cancellation, even under high interference-to-noise ratios.

Limitations and Future Work

  • Computational Complexity: While the paper proposes efficient algorithms, the computational complexity of some methods, particularly for off-grid refinement and certain stages of cascaded channel estimation, can still be high. The paper mentions that off-grid methods can be referred to in [38], [39] for further extension.
  • Idealized Assumptions: The paper relies on certain assumptions, such as the far-field planar wavefront assumption and perfect knowledge of some parameters. Real-world scenarios might deviate from these idealizations.
  • Future Work: The paper suggests further extensions of the off-grid methods for variant LAOMP algorithms and the MMV-version off-grid method. The authors also implicitly suggest exploring more advanced RIS phase matrix adjustment strategies and optimizing the trade-off between passive and hybrid beamforming in more detail.

Summary updated

BibTeX
@article{xiu2024reconfigurable,
  author = {Yue Xiu and Songjie Yang and Wanting Lyu and Yunis Xanthos and Zhongpei Zhang and Chadi Assi and Chau Yuen},
  title = {{Reconfigurable Intelligent Surface-Aided Full-Duplex mmWave MIMO: Channel Estimation, Passive and Hybrid Beamforming}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2024},
  doi = {10.1109/twc.2023.3300719},
}
Figure