Research On Communication Network Effectiveness Evaluation Method Based On Neural Network

In recent years,communication technology has been widely used in the military and civilian fields.The effectiveness evaluating of the communication network helps to plan the network deployment reasonably and improve the efficiency,reliability,and economy of the network system.At present,there are some problems in the effectiveness evaluation of communication networks: firstly,the interpretability of the evaluation results is poor;secondly,the subjective factors of the effectiveness evaluation process are relatively large,and the evaluation results lack objectivity;thirdly,the current evaluation methods lack the factors that affect the network effectiveness.The analysis cannot provide guiding suggestions for network planning.In response to the above problems,the main contributions of this article are as follows:First,this thesis proposes a network effectiveness evaluation method SB-CNEE based on simulation technology.This thesis first clarified the confusion between the concepts of network effectiveness and network performance.Network effectiveness refers to measuring the supportability of a network system as a subsystem to achieve the goal of an overall mission system.On this basis,the calculation formula of network effectiveness and the SB-CNEE network effectiveness evaluation method is proposed.This method uses multi-system co-simulation to simulate the mission system,calculate the overall effectiveness of the mission system based on the deduction results,and then use the network effectiveness calculation formula to calculate the network effectiveness.The SB-CNEE method solves the problems of current effectiveness evaluation methods that are greatly affected by human factors and poor interpretability of evaluation results.Secondly,this thesis proposes a neural-network-based communication network effectiveness evaluation method,referred to as NB-CNEE.SB-CNEE requires cosimulation of multiple systems,and the evaluation process is complicated and timeconsuming.NB-CNEE employs the neural network to deal with these problems.Specifically,we first apply SB-CNEE to conduct extensive simulation to construct a data set.This data set is consequently used to train the neural network that represents the relationship between the network service indicators and network effectiveness.In order to be applied to networks of different sizes and ensure the generality of the neural network model,NB-CNEE uses the K-means algorithm to classify the network nodes and uses the class data,instead of the node data,for the training of the effectiveness evaluation neural network.With the neural network,NB-CNEE only needs to perform a simple inference to obtain the network effectiveness.Compared with SB-CNEE,NB-CNEE is simpler and faster,and can be applied to urgent network effectiveness evaluation tasks in emergency situations.