Due to the exponential growth of network attacks, it is extremely essential to detect and prevent the manin- the-middle attacks both in wired and wireless networks. Most of the network systems are vulnerable to distributed denial of service (DDoS) and TCP flood attacks due to massive data, complex protocol structure and raw packets. A low-rate DDoS attack can easily restricts the communication channel and its services in the distributed systems. Traditional DDOS attack detection models generate a large set of patterns (or signatures) in which most of them are inaccurate due to high false alarm rate. Traditional packet correlation approaches require a large number of network packets along with expert knowledge to prevent complex DDOS attacks. Also, detection and prevention of dynamic DDOS attacks are difficult in the real-time distributed LAN/WLAN networks. In order to overcome these problems in dynamic LAN/WLAN networks, a novel packet monitoring based probabilistic DDOS attack defection and prevention model was proposed. Experimental results proved that the proposed model has high computational efficiency in dynamic DDOS attack detection and prevention compared to the traditional models.
Volume 9 | Issue 2
Pages: 272-286