Randomized RIS Signal Watermarking in FutureG Millimeter-Wave Wireless Communications

Reconfigurable Intelligent Surfaces (RISs) dynami-cally optimize the radio frequency (RF) environment for enhanced signal quality in FutureG millimeter-wave wireless communication. Despite their benefits, RISs can be exploited for malicious purposes, posing threats to the availability, integrity, and security of FutureG wireless communication systems. Conventional data encryption or authentication methods at data layer are not useful against malicious RIS attacks that are launched at physical (PHY) layer. To defend against malicious RIS attacks, we propose a novel RIS signal watermarking scheme that enables the authentication of a RIS deflected signal or the signal path that involves a legitimate RIS. The proposed scheme allows a legitimate RIS to embed a random bit into a deflected RF data symbol by conducting a differential binary phase shift keying (DBPSK) modulation in the middle of the RF symbol. In order to extract both the watermark and data with high reliability, we present a sequential decoding scheme that first decodes the watermark using non-coherent decoding and then data decoding from the recovered data symbol. We derive the bit error rates (BER) performance of both watermark and data decoding under a additive white Gaussian noise channel model. Analytical results show the efficiency of our proposed scheme.