Real valued negative selection for anomaly detection in wireless ad hoc networks

Abstract

Wireless ad hoc network is one of the network technologies that have gained lots of attention from computer scientists for the future telecommunication applications. However it has inherits the major vulnerabilities from its ancestor (i.e., the fixed wired networks) but cannot inherit all the conventional intrusion detection capabilities due to its features and characteristics. Wireless ad hoc network has the potential to become the de facto standard for future wireless networking because of its open medium and dynamic features. Non-infrastructure network such as wireless ad hoc networks are expected to become an important part of 4G architecture in the future. In this paper, we study the use of an Artificial Immune System (AIS) as anomaly detector in a wireless ad hoc network. The main goal of our research is to build a system that can learn and detect new and unknown attacks. To achieve our goal, we studied how the real-valued negative selection algorithm can be applied in wireless ad hoc network network and finally we proposed the enhancements to real-valued negative selection algorithm for anomaly detection in wireless ad hoc network.