Study the performance of different neural architectures for traffic admission control

• Main Author: Lim, Poh Keng.
• Other Authors: Quah, Tong Seng
• Format: Theses and Dissertations
• Published: 2008
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/4684
• Institution: Nanyang Technological University

The capability of neural networks to control connection admission in Asynchronous Transfer Mode (ATM) networks is investigated. The general problem of connection admission control (CAC) and its formulation as a functional mapping are discussed, leading to applications of neural networks and their associated algorithms to the solution of CAC problems. In particular, the use of the class of feed-forward neural networks with backpropagation learning rule and the learning vector quantization (LVQ) network are being studied. Various frameworks have been proposed for the ATM traffic control, but it is not easy to build an efficient traffic control system because of the diversity in multimedia traffic characteristics. This diversity complicates the traffic control system, and various assumptions and simplified traffic models are required to design a practical system based on the traditional mathematical calculations and computer simulations. Neural networks are thought to have many potential applications in ATM traffic control. In this research, the major aim is to present and to compare different neural architectures applicable to connection admission control*