Research On Key Techniques Of Routing Protocols Based On Social Awareness In Delay Tolerant Networks

Delay Tolerant Network (DTN) is a kind of challenged network supporting interoperationamong various local area networks, and long delays can be tolerated. In DTN, message forwardingusually follows Store-Carry-Forward to deal with some problems in traditional networks such asintermittent connection, asymmetric data rate and high error rate. The intermittent connection ofDTN results in the non-availability of an end-to-end path. Also the resources of nodes in thenetwork are usually limited. Therefore, real-time data transmission cannot be provided. All of thesefacts lead to the result that traditional routing protocols cannot work properly in DTN. In the socialapplications of DTN, nodes’ movement possesses social features. With the proposal of sociallyaware computation, it is an important technology that the social characters are utilized to solve theproblems in DTN to meet the social application requirements. Therefore, researchers propose tointroduce social network analysis in DTN and investigate the social awareness based routingprotocols, so as to promote social interaction between nodes.Social awareness is defined as a sociological concept which describes various social phenomenaand human sociability. While in computer science area, the main idea of social awareness focuseson the awareness and response of the computer system to the social context.This thesis studies the routing technologies based on social awareness for Delay TolerantNetworks. Firstly, the social links are constructed according to nodes’ movement. The SLABR(Social Link Awareness Based Routing) algorithm is proposed to optimize the path selection inforwarding process. In addition, when congestion occurs, the SACC（Socially Aware CongestionControl）algorithm is designed to deal with the congestion issue and improve routing reliability.Finally, SAFM（Socially Aware Feedback Mechanism）is presented to reduce the redundant copiesand increase resource utility, so as to improve routing efficiency.The major contributions of our work are described as follows:(1) An improved routing algorithm based on social link awareness (SLABR) is proposed to dealwith the problems of intermitten connection and long delays. The social link of the nodes’ pair isdefined to indicate the social relationship. Then the friendship community of the node is constructedbased on its social links. SLABR is composed of two parts, inter-community forwarding andintra-community spreading. Different forwarding strategies are adopted to increase deliveryprobability, reduce overhead and transmission delay and improve routing efficency. (2) To deal with the problems of congestion issue and routing reliability, an improved SociallyAware Congestion Control algorithm (SACC) is desinged. In this algorithm, the social features andthe congestion level of the node are utilized to construct a social congestion metric (SCM). In theforwarding process, messages are forwarded to the nodes with higher SCM. When the congestionoccurs, a novel dropping strategy is adopted rather than random dropping to improve routingreliability and reduce overhead.(3) The limited resources and redundant copies reduce the utility rate of resources and decreasethe routing efficiency. To solve these problems, an improved Socially Aware Feedback Mechanism(SAFM) is presented. Acknowledgements are forwarded to the nodes whose social link is higherthan the threshold. After getting the acknowledgements, nodes will delete the copies of messagewhich has been received by the destination node to reduce the redundancy. With this mechanism,the utility rate of resources is improved with less overhead and finally routing efficiency isimproved.