CSI-based Simultaneous Location and Velocity Estimation in mmWave Systems

Abstract

In the development of mobile communication systems, localization has become a desirable feature. Channel state information (CSI)-based localization proves extra promising due to the fact that CSI measurement is already a built-in function in most wireless communication systems, which allows a cost-effective integrated design of communication and localization. However, robust localization methods in this novel integrated architecture are yet largely explored. In this paper, we propose a simultaneous user location and velocity estimation algorithm based on the multi-path CSI model. Specially, we first extend the newtonized orthogonal matching pursuit algorithm to extract propagation paths and corresponding channel parameters from the uplink CSI. Then, line-of-sight components are selected out and a weighted least square estimator is built accordingly. We finally propose a channel tracking module to accelerate the algorithm in applications where continuous estimating is required. Simulations show that the proposed algorithm is able to provide accurate location and velocity estimations within a few seconds.