Personalized RIS for Tagging Users: Parallel Communications Based on Blind One-Shot Channel and Delay Estimation

When the phase shift variation of a reconfigurable intelligent surface (RIS) transitions from a quasi-static pattern to a rapidly changing pattern, it engenders a novel capability to support concurrent communications. This paper proposes a novel grouped RIS-enabled multi-user parallel communication framework, where each RIS group is assigned a specified spread spectrum (SS) sequence, e.g., Zadoff-Chu (ZC) code, to tag a distinct user. Although the RIS group establishes a customized virtual channel for its served user through the specified SS sequence, the received signal is still contaminated by other users’ signals. Thus, the quasi-static phase shift of the RIS is optimized to amplify the desired signal while suppressing inter-user interference. However, the transmission delays hinder the perfect detection of the SS sequences. Moreover, the optimization of phase shift necessitates the availability of channel state information (CSI) and exhibits a high sensitivity to the channel phase. Therefore, we propose a pilot-free subspace-based method combined with scalar ambiguity estimation to jointly estimate direct and cascaded channels without phase ambiguity, along with propagation delays. After obtaining the CSI and delays, the phase shift optimization problem could be reformulated as a generalized Rayleigh quotient maximization problem with closed-form solutions. Simulation results validate the accuracy of the blind channel estimation method. Meanwhile, the sum rate of the RIS-enabled parallel communications is ten times higher than that of the conventional RIS-assisted communication, attributed to the SS sequence design of the RIS.