Detection of Greedy Behavior in WSN Using IEEE 802.15 Protocol

On a major security issue in WSN is the availability as these networks suffer from Denial of service Attacks. One of the most aggressive DoS attacks in WSN are greedy behaviors. This attack consists in increasing the bandwidth by a compromised node at the expenses of other nodes by not respecting the access network procedure. In this paper, we tackle the greedy behavior detection in the context of WSN using non slotted 802.15.4. Detecting of such attacks is not evident as it requires assessment tools that decide of a potential attack among normal network behaviors. Formal methods can be very useful to model and simulate the Network and therefore to provide a feedback of its behavior. We propose in this study to use Time Petri nets to detect greedy attaks. For this effect, we characterize the behavior of a greedy node, then we propose a modeling based on Time Petri nets for both a sane and a greedy node. Then we simulate the model under TINA platform by considering different scenarios and confront the obtained values of some parameters (as collision rate, the number of failings, and the energy consumption). Numerical results are used to detect misbehavior directly by nodes only by overhearing the transmissions.