On the wideband nature of UTD-based propagation models

Abstract

UTD-based radiowave propagation prediction models are constantly gaining in popularity as versatile system planning tools, especially in the context of cellular and microcellular mobile radio systems. One of the frequently cited advantages of such models is that they can predict wideband radio channel characteristics, usually by predicting the position-dependent time-delay-spread channel transfer function, or radio channel impulse response. The position-dependent time-delay-spread transfer function can be used directly to produce the delay spread of the radio channel, or can be processed by Fourier transforming it to obtain the remaining channel transfer and correlation functions, which can be then employed to predict the overall radio system performance in the modelled environment. The authors seek to establish that UTD-based propagation models should be used with caution in producing wideband radio channel information, as there exist fundamental limitations in the procedure described for predicting the overall radio system performance. At this stage the authors are unable to establish validity bounds for the above procedure, as they have not conducted numerical simulations and controlled laboratory experiments for sufficiently complex multipath scenarios in order to be able to generalise the conclusions. In this sense, the paper is confining its aims to establishing that in principle the wideband predictions of UTD-based models could be incorrect.