The Research Of Intrusion Detection Model Based On Genetic Neural Network

With the development of computer network techniques and science technology, information industry and it's use have expanded greatly, the enterprise(for instance, government, finance, telegraphy.etc) and personal users have depended on networks more and more larger, at the same time, such has brighten lots of information security in hidden trouble, network security is increasingly paid attention to and concerned about, it is critical problem how to protect security in networks and information system. Traditional network security model could not fit development of network technology, PPDR model emerged, as the times require. Instruction detection technology is PPDR model importantly composed part, and it make up for absence about firewall and data security protection. This technology has not only distinguished from computer and network resources, but also has given important information in instruction; it has not only detected instructing action from out word, but also has controlled user's actions. Instruction detection technology is core in instruction detection system, it include abnormity instruction and abused instruction detection. Aim to the problem of high rate of false negatives and false positives of IDS, proposed the genetic neural network. This method based on the traits that the genetic algorithms are good in global searching, and the back propagation(BP)are effective on accurate local searching and joining the genetics algorithm and BP algorithm together and optimizing the initial weights of BP with GA, solve traditional BP algometry lie in absence about constringency rate slowly and immersion minim value. The result proved, this technology is well, it is in advantage about learning rate rapidity, classify nicety high.