Intrusion Recovery in NLOS WSNs with a Varying Network Density

Abstract

The Internet of Things (IoT) paradigm, supported by the operation of wireless sensor networks (WSNs), has gained a great attention lately and new security concerns and challenges are emerging from its usage. The rising number of cyber threats makes the quest for solutions to promote resilient operations a necessity. A major threat that continues to concern the research community is a denial of service (DoS) attack. In case the network’s communication path is compromised, it needs to be restored so that sensors can continue to propagate observations to a control center and support decision-making. The performance of intrusion recovery measures is critical in order to be able to recover and restore network communication effectively. The network deployment layout can affect the effectiveness of a DoS attack and its respective intrusion recovery measures. This research work investigates how and in what way shadowing that occurs in the communication paths and the network density, can impact the efficiency of a DoS attack and its respective intrusion recovery measure. In this work, a low duty cycle measure is applied by sensor nodes once the DoS attack is detected. Investigations focus on: a) identifying if the malicious nodes chances for compromisation increase or decrease when shadowing is considered, and b) how the intrusion recovery performance is affected. In order to support the objectives of the investigations, comparisons are performed between Line of Sight (LOS) and Non LOS (NLOS) scenarios.