Identifying and locating multiple spoofing attackers using clustering in wireless network

Wireless networks are vulnerable to identity-based attacks, including spoofing attacks, significantly impact the performance of networks. Conventionally, ensuring the identity of the communicator and detecting an adversarial presence is performed via cryptographic authentication. Unfortunately, full-scale authentication is not always desirable as it requires key management, coupled with additional infrastructural overhead and more extensive computations. The proposed non cryptographic mechanism which are complementary to authenticate and can detect device spoofing with little or no dependency on cryptographic keys. This generalized Spoofing attack-detection model utilizes MD5 (Message Digest 5) algorithm to generate unique identifier for each wireless nodes and a physical property associated with each node, as the basis for (1) detecting spoofing attacks; (2) finding the number of attackers when multiple adversaries masquerading as a same node identity; and localizing multiple adversaries. Cluster-based mechanisms are developed to determine the number of attackers. The proposed model can be explored further to improve the accuracy of determining the number of attackers, by using Support Vector Machines (SVM).