Filtering Reconfigurable Intelligent Computational Surface for RF Spectrum Purification

TL;DR

The paper addresses RF spectrum pollution caused by out-of-band interference by introducing Filtering Reconfigurable Intelligent Computational Surfaces (FRICS), a three-layer metasurface system ( Spatial Filtering Layer, Metamaterial Intermediate Layer, Control Layer) with two architectural designs. Design A achieves interference cancellation via bandstop filtering and absorbing metamaterials, while Design B achieves in-band amplification through a bandpass filter paired with a computational metamaterial intermediate layer; both designs are evaluated in D2D and V2X scenarios, showing superior SINR, SNR, and energy efficiency over conventional RIS variants. Key contributions include a tunable spatial filter leveraging varactor diodes controlled by an FPGA, absorber and computational metamaterials enabling simultaneous interference suppression and signal amplification, and a detailed performance comparison highlighting FRICS’ potential for energy-efficient spectrum purification. The work outlines practical challenges and future directions (energy harvesting, phase modulation, ISAC integration, and low-energy transceivers) to enable FRICS adoption in future 6G networks and related wireless systems.

Abstract

The increasing demand for communication is degrading the electromagnetic (EM) transmission environment due to severe EM interference, significantly reducing the efficiency of the radio frequency (RF) spectrum. Metasurfaces, a promising technology for controlling desired EM waves, have recently received significant attention from both academia and industry. However, the potential impact of out-of-band signals has been largely overlooked, leading to RF spectrum pollution and degradation of wireless transmissions. To address this issue, we propose a novel surface structure called the Filtering Reconfigurable Intelligent Computational Surface (FRICS). We introduce two types of FRICS structures: one that dynamically reflects resonance band signals through a tunable spatial filter while absorbing out-of-band signals using metamaterials and the other one that dynamically amplifies in-band signals using computational metamaterials while reflecting out-of-band signals. To evaluate the performance of FRICS, we implement it in device-to-device (D2D) communication and vehicular-to-everything (V2X) scenarios. The experiments demonstrate the superiority of FRICS in signal-to-interference-noise ratio (SINR) and energy efficiency (EE). Finally, we discuss the critical challenges faced and promising techniques for implementing FRICS in future wireless systems.

Figures (4)

• Figure 1: The structure of FRICS. It consists of a spatial filtering layer, an intermediate layer made of metamaterials, and a control layer. Design A on the left includes a bandstop filter layer plus an absorbing material layer. Design B on the right includes a bandpass filter layer plus a computational metamaterial layer.
• Figure 2: The left shows the signal propagation of the FRICS, and the right shows the CST simulation results of the spatial filtering layer, where the design A of FRICS is in (a), and the design B of FRICS is in (b).
• Figure 3: In (a), the application scenario of FRICS with Design A is shown on the left, wherein the outdoor interference signal at the frequency $f_1$ or $f_3$ is filtered and absorbed via FRICS. Concurrently, the signal $f_2$ is reflected to enhance wireless link quality. In contrast, passive/active/star-RIS can only compensate for interference signals by adjusting the reflection phase, as shown on the left of (b). The typical application scenario of FRICS with Design B is depicted on the right of (a), where the desired signal from the vehicle to the roadside unit (RSU) via $f_2$ can be filtered and amplified by FRICS. Meanwhile, the signals $f_1$ and $f_3$ are reflected to enhance the V2V transmission. For comparison, passive/active/star-RIS just reflects/refracted the desired signals to improve the receiving quality, as shown in the right of (b).
• Figure 4: The comparison of SINR and EE for the receiving device in an indoor environment after passing through FRICS (operating with Design A), passive RIS, active RIS, and star RIS (with simultaneous transmission and reflection mode) is shown in (a) and (b), respectively. The comparison of SNR for the RSU and the V2V communication after passing through FRICS (operating with Design B), passive RIS, active RIS, and star RIS (with full transmission mode) is presented in (c) and (d), respectively.