Design of measurement-based wideband mobile radio channel simulators

In this paper, we have proposed a new method of determining the parameters of measurement-based simulation models for wideband mobile radio channels. The method is grounded on fitting the temporal autocorrelation function (TACF) and the frequency correlation function (FCF) of the simulation model to those of measured channels. We have studied the possibility of exploiting the diversity in the frequency and time domains to improve the estimates of the TACF and FCF of the channel, obtained from the time-varying frequency response of the measured channel. The proposed method of determining the parameters of the channel simulator uses numerical optimization techniques and is called the iterative nonlinear least squares approximation (INLSA) algorithm. The simulation results show that the quality of the approximation of the TACF and FCF achieved by applying the INSLA algorithm is better than that attained by using the well-known subspace-alternating generalized expectation-maximization (SAGE) algorithm.