Reseach On Routing And Traffic Balancing Scheme In Low Earth Orbit Satellite Communication System

As an important part of mobile communication, low earth orbit(LEO) satellite communication is attracting more and more attention currently. The reason for this is LEO satellite communication can not only compensate the drawback of low coverage of ground communication system but also has irreplaceable advantages regard to emergency communication compares to other systems. However, although LEO satellite communication system has lower communication delay and transmission loss, it still has shortcomings. First, it is difficult to design routing algorithm due to satellites are highly dynamical. Second, the hazard of traffic congestion in network and satellite nodes failure, which result from the limitation of processing ability on satellite and awful communication environment in outer space, require reliable scheme to deliver packets within stipulated time. Currently, topology of satellite network is used to calculate data transmission routes with the help of periodicity and regularity of the whole constellation movement. It is notable that usually router maps of satellite network is the optimal solution of a couple of network indexs, thereby it has taken traffic balancing and congestion avoidance into consideration.Under the circumstance aforementioned, this thesis first establishes a simulation platform based on OPNET in order to simulate and analyze the end-and-end communication process, the network indexs of the whole system such as system delay, throughput, error rate and how system provide service to high dynamical terminals. Afterwards, routing algorithms for LEO satellite communication system are researched and simulated on the platform. Since time virtulization routing algorithm has the advantage of low communication delay and system complexity but awful adaptability to network emergency situation, thereby a routing algorithm with dynamic detection strategy is proposed via theoretical analysis. By periodically monitoring the transmission subqueue and acknowledgement message during communication to identify whether congestion occurs before making adjustment, this algorithm can not only inherit the advantages of time virtualization but also ease the impacts of emergency situation in the network according to simulation results. Last, traffic equilibrium algorithms are investigated on this platform, unlike traffic equilibrium within routing algorithms which works when packets transmitted on the satellite, algorithm presented in this paper is from another point of view. That is, in order to avoid traffic congestion and lower communication delay, packets from one terminal are distributed to a couple of access satellite nodes instead of one. It is proved that compares with the traditional ones, this algorithm can accommodate more users during the communication process at the same time, therefore it has a good value.