Research On Wireless Network Link Situation Awareness Technology

Wireless ad hoc network has no infrastructure,it can be deployed quickly and nodes can move freely.However,due to the mobility of nodes,the link will be broken and the path will be broken.Therefore,the awareness of link situation is the basis of improving link and path stability.Many researches focus on the change data of radio signal based on hardware and the statistical information of data packet transmission based on software.However,due to the complexity of network environment,it is not easy to build a rigorous and complex mathematical model by using analytic method.In this paper,from the point of view of local topology,the parameters related to link situation are collected and analyzed by mathematical tools such as artificial neural network.The long-term and short-term situation awareness of links are studied respectively.The simulation results show that the proposed method can predict the trend of link stability in the long term,improve the stability of local links and paths and improve the performance of the network in the short term.The main works are as follows:First,the node motion model is established through simulation experiments,and the parameter set related to link situation are collected from the perspective of node motion.The remaining life of the link is used as an index to evaluate the long-term stability of the link.By constructing the artificial neural network,the relationship between the node motion state parameter set and the remaining life of the link is fitted.Taking the given node motion state as the input parameter,the trained artificial neural network is used to predict the link lifetime.The simulation results show that the trained artificial neural network can provide a more accurate lifetime of links for those nodes which are similar to the training samples.After a long time of movement,when the state and training sample set of the nodes change greatly,the trained ANN can also give the trend of the remaining life of the link.The validity of using the remaining life of the link as the evaluation index of the long-term stability of the link is proved.Secondly,a method is proposed to evaluate the link situation from the stability situation of neighbor nodes.Through the evaluation of the node stability,the link stability can be evaluated.Through the simulation experiment,the node walking model is established,the parameters related to the node stability are collected,and the node stability situation is used to evaluate the short-term stability situation of the link.The probability of link maintaining a certain lifetime is used as the evaluation index of link short-term stability.The artificial neural network is used to fit the relationship between the probability and the node motion state.At the same time,a stable routing algorithm based on this link stability evaluation method is proposed.The simulation results show that the path generated by the stable routing algorithm can greatly improve the lifetime of the path compared with the same end-to-end path generated by the shortest path algorithm when only 0 ~ 2 hops are spent.The effectiveness of the proposed short-term link stability evaluation index and the stable routing algorithm based on link stability is proved.The proposed method can accurately evaluate the stability situation of neighbor nodes,increase the lifetime of path,and improve the stability of network and end-to-end data transmission.