Enhancing Physical Layer Security for SISO Systems Using Stacked Intelligent Metasurfaces

Due to the insufficient spatial degrees of freedom, the single-input single-out (SISO) systems face the challenges of physical layer security, reflected by the fact that most conventional schemes are accompanied by reduced spectral efficiency. Recently, stacked smart metasurfaces (SIMs) have emerged as a promising technology for directly processing electromagnetic (EM) wave signals. Through stacking multiple programmable metasurface layers, SIMs are capable of providing additional spatial degrees of freedom without introducing expensive radio frequency (RF) chains, which can be used to reduce the risk of information leakage in SISO systems. Therefore, in this paper, we propose to deploy SIMs at the transmitter of the SISO system to achieve joint modulation, beamforming, and the generation of artificial noise. Under this architecture, the artificial neural network-structured SIM attempts to convert the input carrier signal into the desired output signal. Subsequently, the fitting problem between the actual output signal and the desired signal is formulated. In order to reduce the computational complexity, an alternating optimization algorithm is provided to iteratively determine each variable, from which the closed-form expressions for each phase shift and transmit power are derived. Finally, simulation results show that compared with conventional schemes, SIM-assisted SISO systems can effectively achieve secure communications.