Joint Beamforming Design for Double Active RIS-Assisted Radar-Communication Coexistence Systems

Mengyu Liu, Hong Ren, Cunhua Pan, Boshi Wang, Zhiyuan Yu, Ruisong Weng, Kangda Zhi, Yongchao HeIEEE Trans. Cogn. Commun. Netw 2024 · DOI

TL;DR

This paper tackles the challenge of interference in radar-communication coexistence (RCC) systems by proposing a novel approach using two active reconfigurable intelligent surfaces (RIS). They develop a low-complexity algorithm to jointly optimize radar and communication beamforming, maximizing communication data rates while ensuring radar detection capabilities and power constraints.

What Problem Does This Paper Address?

Next-generation wireless systems are moving towards integrating sensing and communication (ISAC) to improve efficiency. A key scenario is radar-communication coexistence (RCC), where radar and communication systems share spectrum. A major hurdle in RCC is the significant interference that communication signals can cause to radar, degrading its performance. While reconfigurable intelligent surfaces (RIS) have been proposed to mitigate this interference, passive RIS suffer from "multiplicative fading" due to signal attenuation. This paper addresses this by introducing double active RIS – RISs equipped with amplifiers to overcome fading – and tackles the complex optimization problem of designing their beamforming and reflecting coefficients for optimal coexistence.

Methods

The core of this paper lies in developing a novel optimization algorithm to solve a non-convex problem. Here's a breakdown:

  • Problem Formulation: They formulate an optimization problem to maximize the communication system's achievable data rate, subject to constraints on radar detection (SINR requirement) and power budgets for the radar and both active RISs.
  • Penalty Dual Decomposition (PDD): To handle the non-convexity and coupling of variables, they employ a PDD framework. This involves an outer loop for updating penalty and dual parameters and an inner loop for solving a Lagrangian subproblem.
  • Block Coordinate Descent (BCD): The inner loop of the PDD is further decomposed using BCD. This breaks down the complex optimization into smaller, more manageable subproblems, each focusing on a subset of variables (e.g., radar beamforming, RIS phase shifts).
  • Lagrange Dual Method & Bisection Search: Each subproblem is then solved using the Lagrange dual method. For certain constraints, a bisection search method is used to find optimal Lagrange multipliers.
  • Concave-Convex Procedure (CCP): A non-convex constraint within the PDD framework is handled using the CCP method, approximating it with a convex one.
  • Fractional Programming (FP) & Quadratic Transformation: The objective function is reformulated using FP and a quadratic transformation to obtain a lower bound, making it more amenable to optimization.

Datasets

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

Key Insights and Ideas

  • Active RIS Superiority: Active RISs are shown to be significantly better than passive RISs in RCC due to their ability to amplify signals and overcome multiplicative fading, leading to higher data rates.
  • Double RIS Synergy: Deploying two active RISs offers cooperative benefits, further enhancing the achievable data rate compared to a single active RIS. This suggests a synergistic effect from multiple intelligent surfaces.
  • Interference Mitigation Power: The proposed double active RIS system demonstrates a remarkable ability to suppress communication interference to the radar, leading to a higher probability of satisfying stringent radar SINR requirements.
  • Location Matters, but Double RIS Adapts: While RIS location impacts performance, the double active RIS system shows superior performance across various locations, indicating its robustness. Interestingly, as RISs get closer, their performance can be approximated as a single, larger RIS.

Key Results

  • Data Rate Enhancement: The double active RIS-assisted system achieves significantly higher achievable data rates compared to single active RIS and double passive RIS systems under the same power budget and number of reflecting elements.
  • Feasible Solution Probability: The double active RIS system achieves over 95% probability of finding a feasible solution even at high radar SINR requirements (29 dB), outperforming double passive RIS systems (around 60%).
  • Convergence Speed: The proposed PDD algorithm converges quickly, typically within about 10 iterations, reaching a constraint violation level of 10⁻⁶, ensuring feasibility.
  • Power Efficiency: The double active RIS system can satisfy radar detection requirements with lower radar power budgets, allowing more power to be allocated to RISs and the BS for improved data rates.

Limitations and Future Work

  • Self-Interference: The paper explicitly states that self-interference between the two active RISs was not considered. This is identified as a potential area for future research.
  • Channel State Information (CSI): The paper assumes perfect CSI is available. In practical scenarios, imperfect CSI would introduce additional challenges.
  • Computational Complexity: While the proposed algorithm is low-complexity compared to brute-force methods, the PDD framework still involves iterative loops, and the computational complexity can be significant for very large systems.
  • Simplified Models: The paper uses simplified models for radar detection and communication channels. Future work could explore more realistic and complex models.

Summary updated

BibTeX
@article{liu2024joint,
  author = {Mengyu Liu and Hong Ren and Cunhua Pan and Boshi Wang and Zhiyuan Yu and Ruisong Weng and Kangda Zhi and Yongchao He},
  title = {{Joint Beamforming Design for Double Active RIS-Assisted Radar-Communication Coexistence Systems}},
  journal = {IEEE Trans. Cogn. Commun. Netw.},
  year = {2024},
  doi = {10.1109/tccn.2024.3438350},
}
Figure