This paper presents a stochastic 2times2 MIMO wideband channel modelling method that uses tapped delay lines (TDL) and a single channel spatial correlation matrix that represents a novel simplification with respect to other wideband MIMO models. The resulting models are accurate but also easy to implement and computationally efficient. Using this method, outdoor-indoor channel models are obtained based on a measurements campaign. These models will be very useful in the context of present and future broadband fixed wireless systems (BFWS), where a great deal of work has been done in recent years in order to develop and standardize new systems incorporating MIMO technology. In particular, the systems of the IEEE 802.16 family have evolved rapidly: version IEEE 802.16d was completed in the year 2004 and version IEEE 802.16e has been finalized recently.

In this contribution we analyze the performance of several transmission strategies in a MIMO-OFDM system deployed in a wireless metropolitan area network. With this aim, a measurement campaign has statistically characterized a typical outdoor-indoor scenario for wireless metropolitan area networks. The performance is measured in terms of spectral efficiency under several channel knowledge scenarios at the transmitter and power allocation matrices that lead to different transmission strategies. Also two alternatives (averaging in the time or frequency domains) are explored in order to provide the transmitter with the information needed to implement the transmission strategy when it is only possible to get information about the channel statistics.

This paper presents a system providing functionalities for cataloging multimedia content using MPEG-7 and accessing to content and descriptions. The cataloging application indexes content using MPEG-7 and creates annotated variations in order to have the capability of offering media content to a large amount of different terminals and through different access networks. The created multimedia database, both descriptions and content (original sources and variations), is used by, currently, two applications: a searching application, which allows a user selecting specific tags to find the desired media; and a filtering application for transparent access to the content database using profiles. These profiles can be selected from a profiles database or created by the user specifying content preferences, and network and terminal parameters.