Estimation of the User Position and Orientation in mmWave Cellular Networks Aided by an Active RIS

This paper considers an OFDM-based wireless cellular network operating at millimeter waves and studies the problem of estimating the position and orientation of a mobile station (MS) upon relying on the pilot symbols emitted by the serving base station (BS), which are received via a direct link and an indirect link provided by a reconfigurable intelligent surface (RIS). To counterbalance the multiplicative pathloss in the indirect link, an active RIS is employed, which is able to reflect and amplify the incident signal. Upon introducing a convenient signal model, which accounts for the additional noise generated by the reflective amplifiers in the active RIS, we derive the ML estimators of the MS position and orientation and the corresponding Cramér Rao Lower Bounds under three levels of system cognition at the MS concerning the BS beamforming matrix, the RIS response, and the channel amplitudes on the direct and indirect links. In two cases, we also derive a suboptimal estimator with a reduced implementation complexity. Finally, we provide a numerical analysis to show the merits of the proposed estimators, assess the achievable gains granted by the use of an active RIS (as compared to a passive one), and investigate the impact of the main system parameters, including the BS-RIS distance and the amplification gain at the RIS.