Research On Incentive Strategy In Resource-constrained Opportunistic Networks

The concept of Opportunistic Networks mainly from the DTN and MANET. In Opportunistic Networks, nodes are free to move; they can’t predict the next node that will encounter with. Nodes in the network use a "store-carry-forwards" approach to forward data. Therefore, Opportunistic Network can be used to solve wireless network problems, like the delay, the network division, etc. It also can be used to meet the need of harsh environment.Opportunistic Network’s s features and the differences between traditional networks have been researched in resent years. Currently, there have been a lot of research results on Opportunistic Networks, but most studies assume such a precondition that node in the network volunteer to help other nodes to forward messages. However, the resources of nodes in Opportunistic Networks are limited. So in order to save their own resource, nodes show selfishness generally, the selfish behavior seriously affect the normal operation of the network. In order to encourage collaboration between the nodes, suppress selfish behavior of nodes, scholars have raised a lot of nodes incentive policies. However, the existing nodes incentive policies mainly concerning on how to stimulate selfish nodes cooperate with each other instead of thinking about the resources of the node itself. When stimulate nodes frequent mutual cooperation, some active nodes may have died because of energy depletion, this will affect the survival of the network. Therefore, this article will consider the node itself attributes and the attributes of message, using motivational strategies based on cost effective TPS algorithm to stimulate selfish nodes cooperation. The main contents of this paper are summarized as follows:First, we introduce some concepts of Opportunistic Network and applications as well as challenges in real-life. Then describe some existing incentive algorithm for node selfish behavior, such as the node incentive algorithm based on credit, the node incentive algorithm based on prestige and trading-based incentive algorithm. On this basis, we introduce the research work on these nodes incentive algorithm, analyze advantages and disadvantages of these three incentive algorithm. For these incentives algorithm does not consider the effects of each factor on node selfish behavior in the network, this paper describes a strategy of BIP which is based on business model. For BIP policy does not consider nodes encounter probability and the transaction price when selecting the relay node, this paper presents Cost-effective method to select the next node and design a kind of TPS algorithm which is based on trading model.When nodes encounter, both sides update their forwarding utility value, then according to the forwarding utility value list of each other to update itself to other nodes. When source node try to forward a message, it first compares the utility value of itself to the message destination node and the other node to the destination node, if the forwarding utility value of counterpart node is bigger, the source node attempts to deal with the counterpart node. Both sides calculate bid and asking price by its own resources and properties of the message, if bid is greater than the asking price, the transaction can proceed, otherwise, the transaction fails, the source node tries to deal the next message. When selecting forwarding node, the source node will consider the forwarding utility value and final deal price, cost-effective node selected as the next node to transfer message. By this way, we can increase the likelihood of message delivery, this transaction model can also effectively encourage the selfish nodes cooperation, solve nodes blind cooperation cause the degradation of network performance.In this paper, our research chooses ONE as simulation platform. In the simulation, we primarily compare BIP scheme with TPS, we can prove the effectiveness of TPS by simulation.By comparing three parameters in two kinds of program, like the delivery rate of messages, the average relative surplus energy of node, average relative delay of messages, we can evaluate the performance of the network. Experimental results show that TPS algorithm has a good incentive effect on node selfish behavior in resource-constrained opportunistic network, TPS algorithm can effectively improve the delivery rate of the message, reduce energy consumption of nodes, the results verify the effectiveness of this algorithm.