Abstract

With the increasing demand for information anywhere, any time, wireless data service is becoming more important than before. In 2G mobile cellular systems, emphasis is put on circuit-switched voice services that can only supply limited data services, which cannot meet nowadays data requirement. To meet the new challenge, 2G mobile cellular systems have taken an evolutionary approach in developing data based services, which is referred to 2.5G or 2.75G. GPRS, EGPRS and EDGE are standardized by European Telecommunications Standards Institute (ETSI) and belong to 2.5G-2.75G standard. GPRS is only used for packet data service instead of voice traffic. In fact in a cell that supports GPRS, both GSM and GPRS systems operate in parallel to provide service for packet data and voice traffic. EDGE Phase one focuses on the non-real-time packet data services. This is also called EGPRS. The EDGE Phase Two aims at providing both real-time services such as voice and video delivering and non-real-time services in an end-to-end packet mode. To provide various services and to achieve more efficient utilization of the scarce frequency spectrum, research on radio resource allocation is becoming a hot topic. To evaluate the advantage of dynamic resource allocation, the performances of fixed and dynamic resource allocation are compared in this thesis. To simulate various services, four different data rate classes, and four different traffic models are assumed. The result i of the simulations show that: the Dynamic Resource Allocation Scheme can bring much better performance improvement than the Fixed Resource Allocation Scheme. At the same average data rate and the same variance, the performance is the same and does not change with the traffic model. The resource allocation schemes show robustness to different traffic models.