Estimation of Dispersive High-Doppler Channels in the RIS-Aided mmWave Internet of Vehicles.

Reconfigurable intelligent surfaces (RISs) have emerged as a promising candidate for improving the spectral- and energy-efficiency of millimeter-wave (mmWave) Internet of Vehicles (IoV) communications, but the conception of their accurate channel estimation poses. Hence, the existing estimation methods mainly focus on time-invariant channels, while ignoring the Doppler effect induced by the high-velocity vehicles, which will lead to significant performance degradation. In this article, we investigate the problem of channel estimation in RIS-aided mmWave IoV systems considering the deleterious Doppler effect. First, we derive the expression of the time-varying cascaded two-hop multiple-path channels, where each delay tap is subject to multiple paths instead of having a simple one-to-one correspondence. In order to decouple the paths, the problem is formulated in the delay-domain by a series of transformations and the cascaded two-hop channel can be estimated at each delay tap. Then, we propose a pair of estimation strategies by considering different hardware constraints depending on the number of receiver antennas at the base station (BS). When a large receiver array is employed at the BS, we can exploit its high angular selectivity for distinguishing each resolvable path at a certain delay tap because they arrive from different directions. However, this cannot be achieved for small arrays, given their more limited angular resolution. Thus, the RIS reflection patterns are delicately designed for distinguishing multiple resolvable paths. After separating the paths, Doppler estimation can be performed by calculating the phase difference of the adjacent symbols. Our simulation results demonstrate the superior performance of the proposed methods within a wide range of Doppler shifts.