Font Size: a A A

Research On Routing Algorithm Based On Neighbor Node Selection In Opportunistic Networks

Posted on:2019-09-25Degree:MasterType:Thesis
Country:ChinaCandidate:X Q XuFull Text:PDF
GTID:2428330545971630Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of science and technology,mobile smart terminals are gradually replacing the PC.The opportunistic networks gradually appears with the proliferation of mobile smart terminals,and it has become the research focus in the field of communications in recent years because of its numerous advantages.However,the communication mode of the opportunistic networks depends on its internal nodes,which is the reason that the topology structure of the opportunistic networks will be dynamically connected with the movement of nodes.At present,due to the lack of theoretical analysis tools,it is difficult to get an objective assessment to the performance of such networks so far.The transmission of data has great dependence on the grasping and understanding of the dynamic connectivity of the network,obviously,the traditional routing protocol is not applicable to this kind of special network,so the routing protocol applicable to the opportunistic network has been the focus of research in recent years.In the opportunistic networks,mobile devices usually play the roles of nodes,so the mobility characteristics of nodes reflect people's social attributes and movement rules to some extent.Therefore,In the opportunistic networks,there are many applications have constraints such as message delivery rates and end-to-end latency constraints,etc.PRoPHET routing protocol is a classical routing protocol in opportunistic networks,but it just focuses on improving the performance of the algorithm in normal scenes than considering that there are different constraint conditions in some special opportunistic network environments.Therefore,this paper uses PRoPHET routing as the basic algorithm,the following improvement was made in an application scenario with an opportunistic network with constraints.1.This paper proposes an end-to-end delay-limited opportunistic networks application scenario,and proposes a neighbor node selection strategy based on the encounter probability under the constraints of this scenario.Based on this idea,the protocol algorithm named Opportunistic Routing Protocol Based Encounter Probability(abbreviated as BP_PRoPHET)is proposed in this paper.On the premise that the node's movement condition meets the Poisson distribution,the concept of encounter probability isintroduced,the number of neighbor nodes that the node meets within the unit time per unit time is the encounter probability.Each node maintains a list of probability encounters,and this linked list update once every fixed time.The node carrying the message copy selects the node with the better probability of occurrence as the next hop node.As for the BP_PRoPHET routing algorithm,considering that when the neighbor nodes' encounter probability are all smaller than the carrying message node's own encounter probability,there is a case that the message can never be delivered.Therefore,the paper proposes a timeout processing policy what specifies that when the message is not transmitted within the specified time Tcast,and the method of treatment and determination of the size of the Tcast in the experiment.Finally,through a series of simulation experiments,we can see that the performance of BP_PRoPHET routing algorithm in terms of message delivery rate and overhead is better than the traditional routing algorithm in the presence of experimental constraints such as time-delayed and node sparse.2.The paper proposes a message value model and combines it with PRoPHET routing algorithm to propose a routing algorithm based on message value called Opportunistic Routing Protocol Based Message Value(abbreviated as BV_PRoPHET).Considering that the PRoPHET routing algorithm only calculates the connection prediction rate of neighboring nodes and ignores the message itself,it is easy to make effective information unable to be preferentially delivered,which is easy to cause the waste of resources.This algorithm determines the connection prediction rate of neighboring nodes based on the PRoPHET routing algorithm,after that,it is weighted with the size of the message copy and the remaining lifetime.The value obtained by the addition is used as the message value.Based on this,the message type that needs to be sent to the next hop node is selected,then the message is forwarded.Finally,the paper evaluated the performance through a series of simulation experiments.Simulation experiments show that the performance of the BV_PRoPHET routing algorithm is better than the PRoPHET routing algorithm in terms of message delivery rate and overhead in the environment of constraints such as overhead constraints and nodes sparse.
Keywords/Search Tags:opportunistic networks, routing algorithm, encounter probability, time-out processing mechanism, message value model
PDF Full Text Request
Related items