A BACK PROPAGATION BASED BOTNET DETECTION ALGORITHM FOR ENHANCED NETWORK SECURITY
M. Kempanna, R. Jagadeesh Kannan, R. M. K Engg · 2014
The increased network computers makes botnet detection as a challenging one and it makes easier for intruders and attackers to generate mitigation attacks. The centralized propagation nature of botnet floods warms through different botnet clients and questions the network security. To overcome the challenges in identifying the botnet, we propose a new back propagation algorithm for botnet detection. The proposed back propagation method is a learning one, which keeps track of signature of identified warm and the list of hop it traversed. For each warm signature identified, it maintains bot matrix, in which set of hop addresses traversed by the data packet is stored. Whenever a warm packet is identified, its traversal path is tracked and compared with the list of hops present in the bot matrix for the occurrence of hop address present in the traversal path.