Research On Distributed Routing Algorithm In Satellite Networks Based On Leo Polar Orbits Constellation

With increasing technological sophistication of satellite communication, the capability of onboard equipments involved to offer information processing and data forwarding on the orbit has long been recognized. Meanwhile, the broadband satellite link techniques such as laser and microwave have also contributed to make the high-bandwidth satellite network system into reality. By drawing a comprehensive comparison with the traditional ground network, it is definite that the utilization of satellite networks would become the inevitable tendency of future development of mobile communication due to their prominent performance in integrated fields. Such satellite-based networks can be available at providing a highly reliable, ample capacity services over a wide coverage area without limiting to the complicated geographical and natural conditions. Although the benefits are conspicuous, enormous challenges resulting from using the satellites as backbone nodes in the network begin to arisen. Investigating its reason, high-speed rotation of earth has been rendered as the top-issue which leads to the rapid topological change and frequent inter-satellite link switch. Additionally, because of the special working space, there exist surprisingly tough restrictions on the bulk and performance of onboard facilities. Under such circumstance, designing a feasible and effective routing algorithm for the satellite networks is in a great need. This paper conducts detailed research on distributed routing algorithm in satellite networks based on LEO constellation with polar orbits. The most significant contributions of this project are given as follows:(1)In view of the aim of the research, a model which can be applied in the study of routing strategy has been developed for satellite networks on the basis of polar-orbit constellation. And the analysis afterward reveals the fact that least-delay path would certainly belong to the set of the least-hop path in any given circumstances with specific parameters of the constellation. Then all these parts have been combined together to confirm that the least-hop path can also be obtained from the distributed routing algorithm by taking hops as routing cost. Considering the extraordinary circumstances in the space where the propagation delay is the major issue posed during long-distance communication, it seems that the shortest delay strategy for routing algorithm would provide an optimal solution. The distributed routing algorithm in satellite networks performs better in response to the topological change by effectively preventing the link status information from interacting among nodes. And it has also been strongly recommended to simplify the algorithm process. However, what needs to be mentioned here is that this method is limited only by setting the least-hop as the routing cost. As the facts stand, a sound conclusion has been achieved by dissecting the correlation between the least-delay path and the least-hop path. That is, the least-delay path would undoubtedly belong to the set of the least-hop path in a particular situation characterized by a list of parameters of the constellation. When it comes to the satellite networks employing appropriate constellation parameters, the least-hop path getting from the distributed routing algorithm by taking hops as routing coast can also be regarded as the shortest-delay path. Apart from that, much effort has been made to validate that the path delay and the latitude of horizontal links are closely coupled in the processing of theoretical analysis. And the outcome attained has acted a fundamental role in enabling the proposition of the distributed algorithm in accordance with horizontal transmitting priority. After passing through the entire simulation and analysis phase, it has proved that the linkage among the constellation parameters, the shortest-delay path and the least-hop path given by the theoretical analysis is in general accord with the actual conditions in inter-satellite links.(2) The research is practically devoted to advancing a distributed load-balancing routing algorithm based upon horizontal transmitting priority in this part. As a result, it had simplified the complexity of distributed routing algorithm and further gotten rid of the limitation in flow distribution. By making a systematic analysis of routing strategy concerning the LEO polar-orbit constellation, it has been observed that the propagation delay of path is badly affected by the altitude which the satellites with horizontal links located, but is insensitive to the relevant longitude. According to this situation, an optimized satellite network topological model has been constructed by taking the traditional Manhattan network structure into account. Specifically, the dynamic topology of satellite network is recognized as the combination of a static topology and a dynamic list of HTP. The procedure of the distributed routing algorithm is simplified by using such model. In addition, considering the practical reality that the flow of horizontal path is always grouped in high latitude, a load-balancing strategy on account of the HTP and horizontal forwarding probability has been proposed as required. It can be figured out from the simulation that the path obtained from our distributed routing algorithm coincides with that gotten from the traditional centralized algorithm when it is warranted that the LDP certainly conforms to a specialized LHP subset. Moreover, in the case where the LDP is not the part of the LHP group, the results drawn from these two algorithms seem to differ slightly. Also, the data of simulation has been analyzed in detail to prove that these two load-balancing strategies appeared in my paper, in a sense, are able to improve the problem of flow concentration in high latitude.(3) A survivability strategy founded on the distributed routing algorithm in satellite network has been brought forward to get over the problem of performance deterioration found in algorithm when referring to damage to network topology. Distributed routing algorithm copes with the rapid changes in the network topology by avoiding the exchange of link state information, thus simplifying the process of the algorithm. So it is more suitable for the equipments that work on satellites. But at the same time, the algorithm lacks strategies to deal with the unknown damages to network topology. In terms of the weakness of survivability existing in distributed routing algorithms, we have introduced a survivability strategy which is mainly built on the distributed routing algorithm, and is assisted by the use of dynamic routing algorithm when the network topology is subject to a certain degree of damage. In consideration of the impact to the performance of the algorithm itself caused by the flooding information of dynamic routing algorithm, the concept of self-healing area is brought up. The simulation results show that, when the satellite network topology is under damage to some extent, the strategy can make the protocol converges quickly to ensure the high efficiency of routing algorithm. Compared with traditional distributed routing algorithm, the distributed routing algorithm using such survivability strategy obviously has significant performance advantages under the circumstance where the network topology is under damage.