Tensor-Based Modeling and Processing for Channel Estimation in Two-Hop V2X MIMO Systems

Efficient vehicle-to-everything (V2X) communications improve traffic safety, enable autonomous driving and help to reduce environmental impacts. To achieve these objectives, accurate channel estimation in highly mobile scenarios becomes necessary. In this paper, we propose a tensor modeling-based approach for channel estimation and receiver design in a two-hop multiple-input multiple-output V2X communication system. Specifically, by exploiting a Tucker-2 modeling of the received signals, and relying on the joint estimation of the two-hop link, we formulate simple tensor-based closed-form and iterative semi-blind receivers. Furthermore, motivated by the parallel factor analysis (PARAFAC) structure of the time-varying multipath channel, we develop an iterative algorithm for estimating key parameters of the two-hop channel – including angles of departure, angles of arrival and path gains – from the factor matrices of the estimated channel tensors. Simulation results illustrate the performance of the proposed channel estimation and receiver algorithms in selected V2X communications scenarios.