Design And Simulated Implementation Of Routing Algorithm In Information-Centric Networking

Recently, with the increase of user’s number and the rich of web applications and the continuous improvement of technical, Internet terminals, hosted network and application has changed dramatically, and more and more attention of the application is paid to data-centric retrieval and services access, while current network was built in the end-to-end communication between hosts. Host-centric Internet no longer satisfies the increasing demand for highly scalable and efficient content distribution and retrieval. In order to fundamentally solve the problem facing the current Internet, domestic and foreign research scholars committed to the research of Information-Centric Networks. NDN, an important architecture of ICN, which is focused on "what" rather than "where" and route based on content name, can effectively reduce the data redundancy in the network, and improves the content retrieval rate. Therefore, based on NDN, routing algorithm around information was proposed in this thesis.Aim to decrease the data redundancy at all face forwarding strategy of NDN, two algorithms are put forward in this thesis, routing algorithm based on social relations and ant colony routing algorithm based on content popularity. Neighbors and friends relationship in social relations is used to the NDN routing, searching for neighbors and friends before the longest prefix match of FIB table, which can improve the efficiency and success rate of the routing. According to historical information, we establish friend relations and compute friend intimacy which is related to success rate, distance between nodes and so on, so that interest requests for content can be satisfied by a friend who has bigger intimate degree. In the process of data return to the request node, the content popularity and the cache volume of node itself and its neighbor are considered to choose which node should be cache, and content item which has low popularity would be replaced firstly when have not enough space.In order to make the system have self-organizing ability, and look for global path using local information through the interaction between nodes, the ideas of ants foraging is referenced to find the optimal path from the request to the data source by sending interest ants and updating pheromone along the path between request node and data source. Considering the QoS parameters, such as band, delay, delay jitter and so on, the adaptive pheromone updating and volatile formulas are introduced, which can avoid algorithm trapping in local optimum. According to the access frequency of content to calculate content popularity, the more popular content in the network has more frequent path update rate, so that more requests can get data from the optimal source node. Caching capacity and access frequency are considered to cache and cache replacement strategy is implemented based on utility value.In order to verify the effectiveness and feasibility of the routing algorithm in information-centric networks proposed in this thesis, the simulation program is developed respectively on Visual Studio platform, and the simulation and performance analysis has carried on cernet2 topology.The simulation results show that the design of the routing algorithm based on social relations and the ant colony routing algorithm based on content popularity has better performance, in aspect of improving network success rate, reducing the routing time, improving the cache hit ratio and reducing data redundancy.