Wireless ATM network management modeling
Wireless personal communication services (PCS) and broadband networking for delivery of multimedia information represent two well-established trends in telecommunications. Given that ATM is now viewed as a universal base technology for broadband networks, it is reasonable to consider extension of standard ATM services into next-generation microcellular wireless and PCS scenarios. In this M. Sc. thesis, I focus on the mobility management and quality of service (QoS) management in such a wireless ATM system. Several algorithms are investigated for the wireless ATM model, which were developed within OPNET. One is an error-free handoff algorithm that allows data to follow the mobile user. When a mobile user moves from one micro-cell to another, the error-free handoff algorithm is implemented to guarantee continuous communication without loss of packets. A replication algorithm was implemented when the user transmits a handoff request to make the data available at the next location of the mobile user. Dynamic-power-control and automatic-frequency-hopping algorithms were investigated to guarantee the quality of ser ice. When at least one channel is idle, the automatic-frequency-hopping algorithm is implemented to avoid interference. If there are not enough channels to hop to, a dynamic-power-control algorithm is used to ensure the quality of service. The decision control algorithm is neural network based. Simulation studies indicate that the intelligence of a neural network provides for more efficient management than alternative techniques. Finally, a variance fractal dimension method was investigated to estimate the traffic and provide control descriptors, which can be used to make network management more efficient.