Research On Data Forwarding And Cooperation Scheme Of Social Opportunistic Networks

The demand for wireless communication is increasing due to the widely use of handheld and vehicle terminals,popularity of mobile smart devices and enhancement of functions.Wireless networks are gradually extended to the application environment with dynamically changing topology and channel.In these scenarios,there are problems such as the sparse distribution of nodes,fast movement,limited communication capacity,dynamic changes of network topology and other issues.It is difficult to establish a complete end-to-end communication path due to the intermittent connection characteristics.To solve these questions,researchers proposed the architecture of opportunistic networks,where nodes do not need to establish an end-to-end path,not relying on the infrastructure,but relying on the communication opportunities brought by node encounters to transmit messages in the flexible "store-carry-forward" manner.Many researches show that the opportunistic networks have the characteristics of "big world and small world".According to the social relationship between each other,the nodes in the network show the characteristics of dynamic aggregation,and nodes form communities in a self-organization way.The internal links within a given community are relatively close and the external links between the communities are relatively sparse.This type of opportunistic network,which takes full account of the social attributes of nodes and their interrelationships,is called social opportunistic network.Opportunistic networks will play more and more significant role in the fields of social communication,crowd sensing,disaster emergency,portable wear equipment networking,etc.Its research and development will greatly promote the intelligentization,ubiquity and integration of Internet communication.In order to realize the social adaptive opportunistic message forwarding and multi-user collaboration,this paper analyzes the environmental information of opportunistic network to perceive network state information and user behavior parameters under the mathematical model related to learning inference,the classical theory and method of game theory based on the social network theory.Besides,this paper studies key technologies such as data forwarding mechanism,routing strategy and multi-user cooperation mechanism.Specific research results include the following aspects:1)In view of the network performance degradation due to the dynamical characteristics of node mobility in opportunistic networks,a semi-Markov model was established to describe the movement process of nodes.The proposed model can predict the encounter probability of nodes in a certain geographical area and the contact utilization value in a certain period of time in the future.Meantime,the status sensing mechanism can assess the accuracy of the predicted results of nodes.Finally,this paper proposes a message forwarding strategy,which can adapt to the change of node mobility.The simulation results show that the routing strategy is more suitable for social opportunity network,which can improve the delivery probability and reduce the overhead ratio and latency average.2)We propose a message transmission strategy using partial connected subnets and considering the stability of subnets,which is based on the assumption that the mainstream message transmission strategy in opportunistic network does not take into account the inefficient message transmission caused by link disconnection and existence of locally connected stable subnets.Firstly,each node is associated with two attributes: the connected probability and relative moving speed,which evaluates the connection of local topology.Then within the connected subnet,the appropriate path is selected by estimating the stability of the message transmission path,and the message is transmitted in a single copy.According to the similarity and the service ability of nodes,the appropriate relay nodes are selected to suppress the flooding of message replicas in the network.The simulation results show that compared with other message transmission strategies,the proposed adaptive message transmission strategy can effectively improve the probability of successful delivery,while greatly reducing the network delay and network overhead.3)Aiming at the contradiction between the number of duplicates and network load in opportunistic network,this paper proposes an adaptive copy number data forwarding mechanism based on node connection relationships and strength.In particular,node connection relationships can be quantified by employing an analytical approach based on the social pressure metric.By considering the encounter frequency,the total and average connection time,and the average interval time between network nodes,the connection strength between nodes can be evaluated.Simulation results show that the proposed mechanism can effectively reduce the message delivery delay and network overhead.4)To avoid the network performance degradation during the message transmissions between communities in opportunistic networks,the routing strategy based on differential game model in social selfishness environment is proposed.Firstly,the utilization function and state function are exploited to evaluate the collaboration of neighbor nodes.Then the appropriate relay node can be selected through the optimal control method of the differential game model.Simulation results show that the routing strategy based on differential game can improve the message delivery ratio between communities and reduce both the network overhead and transmission delay,when compared with similar routing strategies.5)In order to limit the interference of D2 D users and load off the energy consumption of D2 D users without compromising the communication quality,an interference limited multi-user cooperation scheme is proposed for D2 D users to solve the energy and interference problem in this paper.Multiple D2 D users use non-orthogonal spectrums to form clusters in a self-organized way,and then are divided into different cooperative units.There is no interference among different cooperative units so as to limit the interference between D2 D users in cooperative units.When the link capacity cannot meet the requirements of user rate,an interrupt event occurs.In order to evaluate the communication quality,the probability of link interrupt is derived by considering link delay threshold,data rate and interference.The simulation results show that the interference limited multiple D2 D user cooperation scheme can help offload energy consumption,limit the interference of D2 D users,enhance the energy efficiency,and improve the capacity efficiency.