Effective denial of service attack detection using artificial neural network for wired lan

Nowadays, as information systems are more open to the Internet, the importance of secure networks is tremendously increased. Interconnected systems such as web server data servers are now under the threats of network attackers. Intrusion Detection System (IDS) is the most powerful system that can handle the intrusions of the computer environments by triggering alerts to make the analysts take actions to stop this intrusion. IDS are based on the belief that an intruders behavior will be noticeably different from that of a legitimate user. In the proposed system we capture on line networks packets using packets sniffer and analyzer from different libraries and then we apply k-means for parameter discretization for clustering then extract certain attributes from captured network packets and made training dataset and save it into database. Our training dataset include fifty values and eight attributes. Then in next step in real time IDS again capture real time network packets calculate features from captured packets then load training dataset then apply artificial neural network algorithm which is work in three layers input layer, output layer and hidden layer and provides two outputs normal packets and flood packets. ANN classifier give ninety six percent accuracy for our training data set. Proposed system performance evaluated using online network packets and normal packets and flood packets are classified correctly.