A Study On Distributed Community-based Routing Algorithms For Opportunistic Networks

Opportunistic networks are the special mobile self-organizing networks, which can provide communication service when the networks are intermittent fracture and connected. Opportunistic networks are extensively applied in the wildlife monitoring, disaster emergency communications and vehicle network. With the increasing use of the personal mobile intelligent terminal services in daiy life, the distributed community-based routing algorithms can be designed by combining the social relationship and the routing algorithms in opportunity networks. The distributed community-based routing algorithms make full use of the social attributes of the nodes to forward messages, which can effectively improve the performance of the networks.Firstly, the research background and status of the opportunistic networks are expounded, and then the concept, the characteristics and the applications of the opportunistic networks are also introduced. Moreover, the distributed community-based routing algorithms for opportunistic networks are classified. At last, the characteristic of several representative algorithms are analyzed and compared in detail.In the following, to address the problems existing in the present community-based routing algorithm called CR(Community-based Routing algorithm for opportunistic networks), such as, existing the redundant control overhead when dividing the communities, the transmission delay is serious and the transmission rate is low when forwarding the messages, a new routing algorithm called EDCRA(Efficient Distributed Community-based Routing Algorithms for opportunistic networks) is proposed. EDCRA optimizes the interactive process of the community division, and transmits messages with the help of the neighbor node through the efficient cross layer listening mechanism. In the community, EDCRA firstly chooses the nodes with higher local centrality as the relay nodes, and when the local centrality is equal, the node whose tie strength with the destination node is higher is selected. Outside the community, EDCRA firstly chooses the nodes with higher global centrality as the relay nodes, and when the global centrality is equal, the nodes which has the higher meeting times with the destination community is selected. Theoretical analysis and simulation results show that EDCRA outperforms CR and Bubble Rap in terms of the control overhead, the delivery rate and the average end to end delay, and so on.Additionally, to address the unreasonable messages forwarding order and cache management strategy existing in the present community-based routing algorithm called CSAR(Connection Strength Aware Routing), a new routing algorithm called LDRADC(Low Delay Routing Algorithm based on Distributed Community for opportunistic networks) is proposed. LDRADC improves the messages forwarding order and cache management strategy, gives priority to send the important messages, and tries to save the messages that are ready to reach or have been reach the destination community. Simulation results show that LDRADC outperforms CSAR in terms of the delivery rate, the average hop counts and the average end to end delay, and so on.Finally, the future research directions of distributed community-based routing algorithms for opportunistic networks are proposed based on the conclusion.