Countermeasure for Reinforcement Swap Attack against Directed Diffusion in Wireless Sensor Networks

Directed Diffusion is a data centric protocol that focuses on the energy efficiency of the networks. It is interest based routing protocol in which communication occur hop-to-hop rather than end-to-end in wireless sensor networks. Hop-to-hop communication provides link diversity which helps to overcome obstacles like failure of intermediate node on communication path. Directed Diffusion protocol suffers from many attacks like suppression, cloning, path influence, selective forwarding and data tempering which degrades the performance of the network. In addition to this, reinforcement swap attack is being introduced under the path influence attack, which excludes the good route and includes the bad route in reinforcement path for routing.

This attacker behavior activated/deactivated periodically to prolong the time of attack and affect the performance of network. Reinforcement swap attack is being performed by attacker in two modes: Norm mode and Halt mode. In norm mode, the attacker node alternates between behaving bad and good in the on/off cycles respectively. In halt mode the attacker fluctuates either behaves bad or put itself in sleep mode to hide its own presence in the on/off cycles. In this paper, countermeasure is being proposed for the said attack in both modes of attack. This mechanism detects malicious node on the basis of various performance metrics. In network, senor node which has the lowest value of performance metrics is considered to be malicious and kept at dead state. During the attack, the sensor nodes which have been included in the routing path by the attacker node are being placed at sleep mode.