detection technology for underlying intrusion of large embedded network

The underlying intrusion accurate detection of large embedded network is studied. For the problem that low accuracy of underlying intrusion detection for the large embedded network, an underlying intrusion detection method for the large embedded network based on field rough set theory and BP neural network algorithm is proposed. Firstly, the concept of field is introduced on the basis of rough set theory to reduce the loss of information, field rough set theory is utilized to simplify data, the simplified data set are regarded as the input data of BP neural network, so as to simplify the structure of BP neural network, while reducing the training time of a sample, and improve the classification accuracy of BP neural network. The simulation results in Matlab show that the proposed algorithm for intrusion detection can achieve shorter training time for samples, while the false alarm rate of the invasion has been improved to meet the underlying intrusion detection needs of a large embedded network.