The Research On Data Transmited Scheme In Opportunistic Network Based On Community

With the mass popularization of equipments which have the capability of short distance wireless communication,such as smart phone,tablet,PADs,etc.,the opportunistic network arises at the historic moment.End-to-end network connection doesn't always exist in such networks,and there is no requirement for real-time connectivity.So it is more suitable for the actual ad-hoc network demand,and has won close attention of the academic circles in recent years.Opportunistic network has been widely applied in social network,vehicular network,wildlife tracking and providing network coverage in remote areas,among which the opportunistic network that composed of equipments carried by people and with the capability of short distance communication,is a main applying scenario in opportunistic network.Owing to people has a certain social attributes,such as people with common interests often assemble together to form communities,resuliting in forming community opportunistic network.Such kind of network has now become a major research focus in the field of opportunistic network.Data transmission in opportunistic network has the following characteristics:1)owing to the movement of nodes,it makes the network topology time-varying,and there does not exist stable end-to-end communication path between nodes and data transmission has the characteristics of higher delay and lower rate;2)data transmission among nodes rely mainly on communication opportunities by node moving,so it is a key technology of deciding data transmission performance to using the laws of the encounter among nodes;3)limited network resources,due to the opportunistic network uses the data transmission mode of "storage-carry-forward",data need to be cached in the node for a long time,then there is a more urgent demand for effictive managent of node cache space.There are two big challenges for data transmission in the opportunistic network:nodes moving is unpredictable and network resources are limited.This paper focuses on the nodes moveing reflecting the nature of the community in community opportunistic network,aimed at the circumstances of network resource limited,research two key issues as follow:efficient data transmission strategy and cache management strategies.The main work and innovative achievements are summarized as follows:(1)Based on the idea of cross-layer design,we present an energy-efficient MAC protocol suitable for opportunistic network.The protocol changes the control frame format of underlying MAC protocol by useing information of routing layer protocol.It uses fully and reduces overhearing monitor,at the same time simplifies the control overhead of the multi-hop transmission,then improves the utilization of energy of node,and may further reduce the data transfer delay.(2)According to the node movement has the characteristics of heterogeneity and community in community opportunistic network;we present an efficient routing algorithm for community opportunistic network.The algorithm sets the target community activity and local community activity of node according to the node' meeting history with target community,and rely on the nodes,which have a higher degree of target community activity,to send the message to the target community.At the same time by an appropriate increasement in the number of message copies,it reduces latency effictively of message coming into the target communities.As for delivering message in target community,it uses the historical contact information with destination node to set the activity of destination node,then to determine the timing of forwarding the message.At the same time,it increases the number of message copies appropriatly to ensure high message delivery rate,and removes redundant message copies timely by transmission feedback,then further reduces the consumption of network resources and improves using efficiency of resources.(3)Based on the community characteristics of node mobility,we present an efficient low-latency routing algorithm for community opportunistic network.The algorithm can control the number of message copies adaptively,and depend on nodes,which have more contact with the target community of message,to forward the message to its target community.After message entering its target community,the algorithm generates forwarding tokens for the message again,to improve the successful delivery rate of message and reduce the delay of message transmission,and it uses a transitive function of time utility to complete the message forwarding to the destination node.At the same time,the algorithm sends messages which belonging to the last hop firstly,to further reduce the delay of message forwarding.(4)Based on the characteristics of broadcast when wireless node sending messages,we present a kind of data transmission mechanism which using overhearing for community opportunistic network.When a wireless node sends messages,the other nodes within its transmiting scope can overhearing and receive this information.It is decided by analysising the communities attribute of node whether the node receives the overhearing messages,which may reduce latency of message delivery and improve message delivery rate.When message has arrived in target community or reach the destination node,it'll generate feedback information to confirm the message.By means of overhearing acknowledgment message,the mechanism may remove redundant copy of the message quickly,and reduce consumption of network resources.We realize the community detection by adopting Weighted Network Analysis method based on node contact durations,and verify such transmission mechanism in three sets of real dataset.(5)We propose an adaptive buffer management strategy based on community movement model.According to community property of node in community opportunistic network,we design a self-adaptive lightweight redundant copies erasing mechanism.When a message enters the target community,which the destination node belong to,or reaches the destination node,It'11 generate a delivery status beacon for the message,and redundant copies of the message will be cleared by exchange these beacons,then reduces the node cache occupancy and unnecessary message delivery,thereby reducing the consumption of network resources.At the same time,it uses the community property of node to schedule and transmit cached messages adaptivly,which reduces the delivery latency and improves the message delivery rate.This paper presents a community detection method by using the node encounter frequency and contact duration,and verifies the buffer management strategy by testing it with the real data set of InfoCom06.Finally,we summarize the paper,and suggest the further research directions.