A packet marking with fair probability distribution function for minimizing the convergence time in wireless sensor networks

Wireless sensor networks (WSNs) contain a great number of nodes with sensing, processing, and wireless communicating capabilities. WSNs are expected to become the basic building blocks of the ubiquitous computing environments. However, inherited from its designed nature with limited resource constraints, WSNs exposed themselves to serious security threats. Their precious resources (e.g., low bandwidths and battery power) make a malicious node easy to launch the DoS flooding attacks by sending extra unnecessary packets. A DoS/DDoS attack may result in network disasters due to the energy exhaustion of the nodes along the attacking path. In the conventional IP network, edge sampling is a well known traceback algorithm to countermeasure DoS/DDoS attacks. Unfortunately, edge sampling is not effective enough for WSNs because it requires a lot of packets to reconstruct the attacking path, which may consume considerable energy and bandwidth. In addition, a shorter convergence time can reduce the failure rate of a traceback process due to mobility. This paper proposed an equality approach to deal with the traceback problem, called the edge sampling algorithm with probability distribution fairness (ESA–PDF), which reduces the convergence time of the conventional edge sampling algorithm. The salient features of the proposed ESA–PDF algorithm include: (1) able to produce faster convergence time, (2) capable of working as optimal in certain conditions, and (3) susceptible of integration with AODV routing protocol. Such a technique can provide a key answer required for advancing the state-of-the-art in DDoS mitigation and defenses in a realistic environment.