A Fuzzy Advance Reservation Mechanism of Network Bandwidth in Video Grid

Abstract

Next generation grid networks will be required to support video delivering services which need bandwidth is reserved at an earlier time so that desired QoS requirement could be ensured in the future. In this paper, we propose a new intelligent admission control algorithm for advance reservation. The main design principle underlying our algorithm is delay and jitter are playing more and more important roles in providing high quality video to users, but they are often fuzzy and indefinite expression from clients in advance and it is therefore hard to get the satisfied advance reservations for network resources requirements using the accurate mathematic models like traditional schemes. So, we propose a fuzzy neural network model to cluster priority of advance reservation requests according to delay and jitter variables and then allocate the bandwidth based on priority preempting principle. Experimental results show our algorithms achieve well-balanced performance.