Decoupling Networks and Super-Quadratic Gains for RIS Systems With Mutual Coupling

Dominik Semmler, Josef A. Nossek, Michael Joham, Benedikt Böck, Wolfgang UtschickIEEE Trans. Wireless Communications 2024

Abstract: We propose decoupling networks for the reconfigurable intelligent surface (RIS) array as a solution to benefit from the mutual coupling between the reflecting elements. In particular, we show that when incorporating these networks, the system model reduces to the same structure as if no mutual coupling is present. Hence, all algorithms and theoretical discussions neglecting mutual coupling can be directly applied when mutual coupling is present by utilizing our proposed decoupling networks. For example, by including decoupling networks, the channel gain maximization in RIS-aided single-input single-output (SISO) systems does not require an iterative algorithm but is given in closed form as opposed to using no decoupling network. In addition, this closed-form solution allows to analytically analyze scenarios under mutual coupling resulting in novel connections to the conventional transmit array gain. In particular, we show that super-quadratic (up to quartic) channel gains w.r.t. the number of RIS elements are possible and, therefore, the system with mutual coupling performs significantly better than the conventional uncoupled system in which only squared gains are possible. We consider diagonal as well as beyond diagonal (BD)-RISs and give various analytical and numerical results, including the inevitable losses at the RIS array. In addition, simulation results validate the superior performance of decoupling networks w.r.t. the channel gain compared to other state-of-the-art methods.

TL;DR

This paper introduces decoupling networks as a hardware-based solution to manage mutual coupling in Reconfigurable Intelligent Surfaces (RIS). By integrating these passive networks, the complex electromagnetic interactions between RIS elements are mathematically "simplified" into a standard uncoupled model, enabling closed-form optimizations and revealing that mutual coupling can actually be exploited to achieve super-quadratic channel gains (scaling up to $N^4$ instead of the usual $N^2$).

What Problem Does This Paper Address?

In the quest for 6G, RIS technology is a frontrunner for shaping wireless environments. Most existing research relies on a simplified "phase-shift" model, assuming that each reflecting element operates independently. However, in practical deployments, RIS elements are packed tightly together, leading to mutual coupling—where the electromagnetic field of one element influences its neighbors.

Traditionally, mutual coupling is viewed as a nuisance that makes system models mathematically "messy," requiring complex iterative algorithms to optimize. This paper tackles two main questions:

  1. Can we simplify the system model so that existing "uncoupled" algorithms can still be used?
  2. Can we stop treating mutual coupling as a problem to be mitigated and instead use it as a resource to boost performance?

Methods

The authors propose adding a decoupling network—a passive, lossless, and reciprocal multiport circuit—between the RIS antenna elements and the adjustable load impedances.

Proposed circuit model incorporating a decoupling network $Z_{DeN}$ to simplify mutual coupling effects.
  • Model Transformation: They mathematically demonstrate that incorporating this network transforms the system's impedance matrix. The resulting "equivalent" model has the exact same structure as a system without mutual coupling.
  • Closed-Form Solutions: Because the model is simplified, the authors derive a closed-form solution for maximizing the channel gain in Single-Input Single-Output (SISO) systems. This replaces the computationally expensive iterative methods used in prior work.
  • Beyond-Diagonal (BD)-RIS: The paper explores how decoupling networks relate to BD-RIS architectures. They show that a simple diagonal RIS equipped with a decoupling network effectively behaves like a specialized, high-performance BD-RIS.

Key Insights and Ideas

  • Coupling as a Feature, Not a Bug: The most striking insight is that mutual coupling allows for "super-quadratic" gains. While an uncoupled RIS typically sees gains scaling with $N^2$ (where $N$ is the number of elements), a coupled RIS with a decoupling network can achieve scaling of $N^3$ or even $N^4$ in specific "end-fire" configurations.
  • The Power of Legendre Polynomials: To prove the $N^3$ scaling bound for any user position, the authors utilized properties of Legendre polynomials, providing a rigorous mathematical foundation for these "super-gains."
  • Static vs. Reconfigurable Complexity: While BD-RIS usually requires a complex, reconfigurable interconnection matrix, a decoupling network is static. It only needs to be calibrated once during fabrication based on the physical array geometry.
  • The "Optimal Spacing" Trade-off: While theoretical gains explode as element spacing $d$ approaches zero, real-world Ohmic losses act as a dampener. The paper identifies that there is an optimal spacing where the benefits of coupling are maximized before losses take over.

Key Results

In simulations, the proposed decoupling approach consistently matched or outperformed all existing iterative state-of-the-art methods while being computationally simpler. The authors proved that in an "end-fire" scenario, the array gain reaches $N^4$, and established a lower bound showing that for a fully-connected BD-RIS, the gain scales at least with $N^3/8$ regardless of the user's angle.

Array gain $A$ as a function of element spacing $d/\lambda$ for different numbers of RIS elements $N$.

Limitations and Future Work

  • Hardware Complexity: Implementing a multiport decoupling network in hardware is non-trivial. It requires additional passive components (like inductors and capacitors) which increase the footprint and cost of the RIS.
  • Sensitivity to Calibration: Since the network is static and tuned to the mutual coupling of the array, any physical deformation or environmental changes affecting the coupling might require re-calibration.
  • Future Directions: Future research should focus on the practical hardware implementation of these networks and extending the "super-quadratic" analysis to more complex multi-user MIMO environments.

Summary updated

Abstract

We propose decoupling networks for the reconfigurable intelligent surface (RIS) array as a solution to benefit from the mutual coupling between the reflecting elements. In particular, we show that when incorporating these networks, the system model reduces to the same structure as if no mutual coupling is present. Hence, all algorithms and theoretical discussions neglecting mutual coupling can be directly applied when mutual coupling is present by utilizing our proposed decoupling networks. For example, by including decoupling networks, the channel gain maximization in RIS-aided single-input single-output (SISO) systems does not require an iterative algorithm but is given in closed form as opposed to using no decoupling network. In addition, this closed-form solution allows to analytically analyze scenarios under mutual coupling resulting in novel connections to the conventional transmit array gain. In particular, we show that super-quadratic (up to quartic) channel gains w.r.t. the number of RIS elements are possible and, therefore, the system with mutual coupling performs significantly better than the conventional uncoupled system in which only squared gains are possible. We consider diagonal as well as beyond diagonal (BD)-RISs and give various analytical and numerical results, including the inevitable losses at the RIS array. In addition, simulation results validate the superior performance of decoupling networks w.r.t. the channel gain compared to other state-of-the-art methods.
BibTeX
@article{semmler2024decoupling,
  author = {Dominik Semmler and Josef A. Nossek and Michael Joham and Benedikt Böck and Wolfgang Utschick},
  title = {{Decoupling Networks and Super-Quadratic Gains for RIS Systems With Mutual Coupling}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2024},
}
Figure