State-space approach to propagation path parameter estimation and tracking

Abstract

In this paper we address the problem of propagation path parameter estimation in channel sounding. Propagation parameter estimation is crucial in creating realistic channel models that may be used to study the performance of multiantenna (MIMO) transceivers as well as in network planning. The proposed approach employs a nonlinear state-space model in order to capture the dynamics of the channel parameters in time. Both specular and diffuse components are considered. Extended Kalman filtering is used to estimate the state. The computational complexity is reduced by applying the matrix inversion lemma. Hence, significant savings in computation compared to conventional iterative methods is obtained. The method gives insight into the dynamic behavior of the propagation parameters, allows parameter pairing over time and facilitates analyzing the path lifetime in different measurement scenarios. The performance of the proposed technique is demonstrated using real-world channel sounding measurements.