An improved method based on TS-BPNN for network intrusion detection
ZHOU Li-juan;Experimental Teaching Center,Shanxi University of Finance and Economics;
In order to improve the intrusion detection rate of the existing methods,an intrusion detection for network based on tab algorithm and BP network is proposed. The intrusion detection model based on BP network is established. The parameters of the network are adjusted according to the differences between the labels of the sample data and the real output. To avoid the local optimal value of the network structure parameters,the parameters of the network have to be optimized by Tabu algorithm. The optimal solution obtained from the Tabu algorithm is used to assign to the BP network. The detection result can be obtained by feeding the data to the neural network. The experiment is operated under the MATLAB environment. The proposed method is compared with BP neural network,AdaBoost and Elman,the result shows that the proposed method has the higher intrusion detection rate.
【CateGory Index】： TP181;TP393.08