Adaptive sinkhole detection on wireless ad hoc networks

Wireless ad hoc technology is demanding and continually growing. Its dynamics and flexibility allow the network to be easily set up without the requirement of a predetermined infrastructure. The advancement of the technology itself draws attentions from intruders as well as researchers and developers. Consequently, detecting intrusions on wireless ad hoc network is one of the most challenging issues facing today's researchers. Its dynamic nature permits the existence of a great number of difficulties in implementing its security features. In a highly dynamic ad hoc network, it is inefficient to learn normal behaviours of each node from the network where nodes move in and out all the time. This research introduces an approach for sinkhole countermeasure. Sinkhole attacks occur when malicious nodes do not relay the received data to a destination and finally cause an availability problem to the network in the form of denial-of-services. Here, malicious nodes try to disable any services passing through it. We present a mechanism for adaptively detecting and defending against sinkhole attacks in a dynamic ad hoc system by applying trust-based algorithm on ad hoc nodes. Weights and thresholds are used for separating suspicious behaviours from normal ones.