A Secure Protocol against Selfish and Pollution Attacker Misbehavior in Clustered WSNs

Wireless sensor networks (<i>WSNs</i>) have been widely used for applications in numerous fields. One of the main challenges is the limited energy resources when designing secure routing in such networks. Hierarchical organization of nodes in the network can make efficient use of their r...

Full description

Bibliographic Details
Main Authors: Hana Rhim, Damien Sauveron, Ryma Abassi, Karim Tamine, Sihem Guemara
Format: Article
Language:English
Published: MDPI AG 2021-05-01
Series:Electronics
Subjects:
Online Access:https://www.mdpi.com/2079-9292/10/11/1244
Description
Summary:Wireless sensor networks (<i>WSNs</i>) have been widely used for applications in numerous fields. One of the main challenges is the limited energy resources when designing secure routing in such networks. Hierarchical organization of nodes in the network can make efficient use of their resources. In this case, a subset of nodes, the cluster heads (<i>CH</i>s), is entrusted with transmitting messages from cluster nodes to the base station (<i>BS</i>). However, the existence of selfish or pollution attacker nodes in the network causes data transmission failure and damages the network availability and integrity. Mainly, when critical nodes like <i>CH</i> nodes misbehave by refusing to forward data to the <i>BS</i>, by modifying data in transit or by injecting polluted data, the whole network becomes defective. This paper presents a secure protocol against selfish and pollution attacker misbehavior in clustered <i>WSNs</i>, known as (<i>SSP</i>). It aims to thwart both selfish and pollution attacker misbehaviors, the former being a form of a Denial of Service (<i>DoS</i>) attack. In addition, it maintains a level of confidentiality against eavesdroppers. Based on a random linear network coding (<i>NC</i>) technique, the protocol uses pre-loaded matrices within sensor nodes to conceive a larger number of new packets from a set of initial data packets, thus creating data redundancy. Then, it transmits them through separate paths to the <i>BS</i>. Furthermore, it detects misbehaving nodes among <i>CH</i>s and executes a punishment mechanism using a control counter. The security analysis and simulation results demonstrate that the proposed solution is not only capable of preventing and detecting <i>DoS</i> attacks as well as pollution attacks, but can also maintain scalable and stable routing for large networks. The protocol means 100% of messages are successfully recovered and received at the <i>BS</i> when the percentage of lost packets is around 20%. Moreover, when the number of misbehaving nodes executing pollution attacks reaches a certain threshold, <i>SSP</i> scores a reception rate of correctly reconstructed messages equal to 100%. If the <i>SSP</i> protocol is not applied, the rate of reception of correctly reconstructed messages is reduced by 90% at the same case.
ISSN:2079-9292