Reconfigurable Intelligent Surfaces for Destructive Beamforming via Manifold Optimization

Luís Gustavo Zulai, José Carlos Marinello

DOI: 10.14209/sbrt.2025.1571157320

Evento: XLIII Simpósio Brasileiro de Telecomunicações e Processamento de Sinais (SBrT2025)

Keywords: Reconfigurable intelligent surface (RIS) destructive beamforming manifold optimization signal nulling

Abstract

Reconfigurable Intelligent Surfaces (RISs) are emerging as a key technology for future wireless systems, enabling control over the propagation environment. While most research focuses on enhancing desired signals, this paper inves- tigates the use of RIS for destructive beamforming - intentionally creating signal nulls at specific locations. This capability is crucial for applications like physical layer security, interference mitigation, and deliberate jamming. We formulate the problem of minimizing the signal-to-noise ratio (SNR) at a target user equipment (UE) in both single-input single-output (SISO) and multiple-input single-output (MISO) scenarios by optimizing the RIS phase shifts under practical unit-modulus constraints. For the MISO case, we propose an efficient method based on manifold optimization techniques, specifically the Trust-Region algorithm, to solve the non-convex optimization problem. This approach utilizes multiple random initializations to enhance the likelihood of finding a high-quality solution. Simulation results, considering realistic transmit power and a channel model incorporating Ri- cian fading, realistic path losses, and shadow fading, demonstrate that the proposed approach can significantly suppress the signal power at the target UE, with the suppression deepening as the number of RIS elements increases, validating the potential of RIS for precise destructive beamforming. For instance, with 256 RIS elements, a suppression of approximately 14.1 dB compared to the scenario without RIS is achieved.

Download