Design and Evaluation of Memory Efficient Data Structure Scheme for Energy Drainage Attacks in Wireless Sensor Networks

Wireless Sensor Networks (WSN) are deployed on a large scale and require protection from malicious energy drainage attacks, particularly those directed at the routing layer. The complexity increases during critical operations like cluster head selection where detection of such attacks is challenging. The dependency of WSN on batteries elevates the concern posed by these threats, making detection and isolation crucial, especially within the framework of energy-efficient clustering protocols such as Low Energy Adaptive Clustering Hierarchy (LEACH). Various approaches have been proposed in prior research to deal with such attacks. However, the use of memory-efficient data structures has yet to be effectively addressed. In this article, considering the limitations of WSN, we utilize memory-efficient data structures named Bloom filters, count-min (CM) sketch, and cellular automata (CA) to address abnormal energy drainage. A CA-based trust model is used to choose the legitimate node as the cluster head. CM sketch is used to control the frequency of a node selected as a cluster head, achieving fairness in the cluster head selection process, and Bloom filters maintain the record of malicious nodes blocked from participating in the communication or cluster head selection process. CA and trust functions collectively keep a record of neighbors’ energy and their trust in the network. Grayhole, blackhole, and scheduling attacks are three well-known threats that lead to abnormal energy drainage in legitimate nodes. The proposed solution effectively detects and addresses abnormal energy drainage in WSN. Its impact is simulated and observed using ns2 IEEE 802.15.4 medium access control (MAC) and LEACH clustering protocols, specifically in the context of the mentioned attacks. The effectiveness of the proposed model was rigorously analysed, and it was observed that it reduces the energy consumption of WSN by approximately 16.66%, 48.33%, and 43.33% in the cases of grayhole, blackhole, and scheduling attacks, respectively. In terms of space/time complexity, its growth is linear O(n). The proposed solution also consumes 0.08-0.10 J more energy compared to the original LEACH as a cost of the solution, which is not more than 2% of the total initial energy. The trade-off of implementing heightened security is worthwhile, as the proposed approach outperforms the original LEACH and related methods, effectively mitigating abnormal energy drainage in WSN and extending network lifetime, especially in challenging environments with persistent battery recharging challenges.