Abstract
We present a fuzzy satellite routing strategy (FSRS) for NGEO satellite IP networks to meet multi-class QoS demands. A set of fuzzy rules is established to map the congestion degrees and the linguistic variables. For every queue in a satellite, an output of the fuzzy logic is present to reflect the queue congestion degree called fuzzy queue congestion indicator (FQCI). For a satellite, three FQCIs are synthesised to obtain a Fuzzy Satellite Congestion Indicator (FSCI). One shortest end-to-end delay path and one alternative path are pre-calculated. And the real-time queuing delay is predicted by means of queuing theory and naive bayesian method. With the FSCI, the neighbouring satellites adjust transmission modes. Moreover, packets of different QoS could be treated discriminatively. A series of simulations were conducted in Network Simulator (Version 2.34) to verify the congestion alleviation effect. Simulation shows that FSRS performs better in terms of end to end delay, throughput, packet loss rate and traffic distribution, especially, it generates more a smooth shape for high class traffic.