Research On The Key Technology Of Network Running Situation Awareness Based On Fuzzy Neural Network

In recent years,with the development of society,the network is becoming more and more complex,and the traditional network management is no longer suitable for the complex network environment.It is in urgent need of a technology that can monitor the network running status and predict the network change trend in real time,and the network running situation awareness is proposed under this environment.The network running situation awareness is to analyze and predict the network situation information.It is necessary to summarize,merge,correlate,synthesize and extract the runnig data from each network unit or each subnetwork under certain rules.It is essentially a complex data processing process.Fuzzy neural network is a combination of fuzzy theory and neural network technology of artificial intelligence technology,which can give full play to the advantages of both.It is an important development direction in this field at present,and it is also a powerful means to solve the network running situation awareness in complex environment.Therefore,in this thesis,fuzzy neural network has been applied to network running situation awareness.Surrounding the network running situation assessment model based on fuzzy neural network and the real-time link weight,assessment method of the real-time link weight,fuzzy neural network's structure and its learning algorithm and so on,this thesis has carried on the theoretical research and simulation experiment.The main contents are as follows:1.This thesis studies the overall model of network running situation awareness,focused on the method and model of network running situation assessment.And this thesis proposes a network running situation assessment scheme based on fuzzy neural network and real-time link weight.The fuzzy neural network is applied to evaluate the link running quality,and the real-time link weight evaluation method is used to calculate the real-time weight of each link.Finally,the assessment value of network running situation awareness is computed form the two part.2.This thesis studies the structure and learning algorithm of fuzzy neural network.By comparing the existing advantages and disadvantages of various structures and considering the real time demand of situation awareness,this thesis chooses a kind of four layers of fuzzy neural network structure,and combined with particle swarmoptimization algorithm and back propagation(BP)algorithm,this thesis puts forward a hybrid learning algorithm.The hybrid algorithm gives full play to the global optimization ability of the particle swarm algorithm and local optimization ability of BP algorithm.Simulation results show that the algorithm is efficient and correct.3.NS2 network simulation software is applied to build a network simulation environment,and gawk analysis tool was used to collect network runnig data.Simulation experiments of network running situation assessment are carried out according to the proposed scheme.