Abstract
The characterisation of the indoor radio channel has been an active area of research in the last decades, and this is likely to continue with the increasing importance of wireless techniques for the provision of connectivity within homes and offices. This thesis gives a survey of existing models for the indoor radio channel, noting that site-specific predictions are often of limited use owing to the difficulty of determining accurate details of the indoor environment. While ray tracing models, for instance, are potentially capable of offering highly accurate prediction, this promise is seldom fulfilled in practice owing to the difficulty of capturing the necessary input data for such models. As a consequence, statistical treatments of the indoor channel are popular, and two such models for the temporal response of the wideband channel are discussed in some detail. These models, however, rely on the use of empirical parameters that will either represent only the generality of buildings, or will need to be determined by measurement for each specific case. An alternative model was therefore sought which combined physical and statistical elements to introduce a greater degree of site-specificity. The initial modelling reported in this thesis took, as its basis, the methods developed within the acoustic community for the prediction of impulse response within reverberant rooms. The thesis then proposes a new model for the prediction of the average power delay profile (PDP) within a room, taking the extreme room dimensions as parameters to allow the accommodation of site-specific detail. A measurement campaign, using a channel sounder developed for the purpose, is described, and the results compared with the predictions offered by the new model. Finally, the thesis suggests ways in which the new model might be used as a component of other, more general, models for the indoor radio channel.