Channel Estimation for RIS-assisted MISO System using Vector Approximate Message Passing

Reconfigurable Intelligent Surface (RIS) is composed of a large number of adjustable passive electromagnetic reflection elements. In RIS-assisted multi-antenna communication systems, the quasi-passive characteristic prevents the reflection channel from being obtained piecewise. In this paper, we focus on the channel estimation over the uplink of RIS-assisted Multi-input Single-output (MISO) communication system. In angular domain, channel energy is concentrated at a few limited directions, which makes the cascaded channel sparse. By utilizing the sparsity of cascaded channel in the angle domain, we formulate the channel estimation problem as a sparse signal reconstruction problem. To solve this problem, we propose an Expectation Maximization-Vector Approximate Message Passing (EM-VAMP) algorithm to estimate the cascaded channel. Simulation results show that the proposed EM-VAMP algorithm achieves lower computational complexity and higher estimation accuracy than traditional algorithms.