This paper deals with the design of measurementbased simulation models for wideband single-input single-output (SISO) mobile radio channels. We present an improved version of the iterative nonlinear least square approximation (INLSA) method for computing the parameters of measurement-based simulation models. The proposed method aims to fit the temporalfrequency correlation function (TFCF) of the simulation model to that of the measured channel. Unlike the original INLSA method, the proposed approach provides a unique optimal set of estimated model parameters. The proposed iterative procedure involves numerical optimization techniques to determine a set of parameters that minimizes the Euclidian norm of the fitting error. Our investigations show that the proposed method performs well in terms of the goodness of fit to the measured data. The new method is relatively simple and can be used for the design of measurement-based wideband channel simulators, which are important for the performance analysis of mobile communication systems under real-world propagation conditions.
In this paper we investigate the application potential of three known algorithms, namely the ESPRIT, SAGE and INLSA, to properly emulate the statistical properties of the mobile fading channel. The performance comparisons of those methods will be carried out with respect to their fitting accuracy to the autocorrelation function (ACF) of the reference model. The methods' accuracy will be assessed in the synthetic reference channel model, which is based on the sum-of-cisoids (SOC) principle. In our reference model, we consider the scenarios with equal gains and Rayleigh distributed gains. The obtained results indicate that in both scenarios the INLSA method is more preferable than the ESPRIT and the SAGE algorithms when emulating the correlation properties of the mobile fading channel. An excellent performance and the simplicity of the INLSA make the method a powerful tool for designing realistic channel simulators.
In this paper, we derive an analytical expression for the ACF of Rice processes in the general case of unsymmetrical Doppler power spectral densities. This expression, which is obtained based on the multidimensional Gaussian distribution approach, is shown to cover the ACF of Rayleigh processes as a special case. Various numerical examples are presented to illustrate the impact of the channel parameters on the ACF. Computer simulations, considering the von Mises distribution for the angle of arrivals, are also performed to check the validity of the analytical result. Finally, the analysis of the covariance spectrum is addressed.
In this paper, we present an improved version of the iterative nonlinear least square approximation (INLSA) method for designing measurement-based single-input single-output (SISO) wideband channel simulators. The proposed method aims to fit the time-frequency correlation function (TFCF) of the simulation model to that of a measured channel. The parameters of the simulation model are determined iteratively by minimizing the Frobenius norm, which serves as a measure for the fitting error. In contrast to the original INLSA method, the proposed approach provides a unique optimized set of model parameters, which guarantees a quasi-perfect fitting with respect to the TFCF. We analyze the performance of the proposed method in terms of the goodness of fit to the measured data. The investigations will be carried out with respect to the TFCF and the scattering function. We demonstrate that the proposed approach is a powerful tool for the design of measurement-based wideband channel simulators, which are important for the performance evaluation of mobile communication systems under real-world propagation conditions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.