Modeling and Prediction of mm Wavelength Channels, Scheduling and Adaptive Data Rate for LMDS

The next generation of fixed wireless access networks, consisting of local multipoint distributed systems, is a means to provide everyone with true broadband services. Operating in the 20-45 GHz frequency bands the bandwidth will not be as scarce as for l ower frequencies, but the channel conditions might some times be challenging. The capacity of the radio link depends strongly on the weather and can change rapidly due to multipath effects in the atmosphere, vegetation or other reflecting objects. To obta in high data rates and reliable links over such fading channels one has to rely on adaptive data rate algorithms and radio resource scheduling. This is especially true when using low cost equipment with single antennas. To design appropriate countermeasu res to the fading, the dynamical behavior of the channel has to be understood. Long term slowly varying fading for fixed radio links has been thoroughly studied, but we aim to investigate the fast fading. An important challenge is to create realistic mode ls for the dynamics of the channel, applicable for system investigation and simulations. Insights in the fading process, as well as the models developed, will allow us to design prediction algorithms for the radio channel for different time scales. In ad dition to a deep understanding of the channel characteristics, we need knowledge on the system design in order to develop a good concept for adaptive data rate allocation, and radio resource management. A good dialog with the BROADWAN project is therefore necessary and planned. No solutions can be an obstacle for low cost equipment implementation. Furthermore, scalability is important in order for deployment of this type of system to get started. The approaches developed in this project will influence the BRAODWAN standard for low cost fixed area networks.