ImHQCN: Improved Harris Hawk Optimized Modified Elliptic Curve Digital Signature Encryption-based Malware Prediction and Secure Data Sharing Framework
Jolly Nikhade, Shrikant V. Sonekar · 2025
In recent days, Wireless sensor networks have been used in diverse applications because of their limited energy and portable nature. Furthermore, WSN becomes an easy target for intruders to attack the network due to its distributed and infrastructure-less nature. However, conventional malware detection techniques possessed the limitation of constrained resources, high power consumption, lack of security, and improper routing mechanisms. Therefore, to overcome these challenges, the research proposes the Improved Harris Hawk optimized quantum convolutional neural network (ImHQCN) for identifying malware and ensuring secure data sharing in the network. The model incorporates modified dynamic advance elliptic curve digital signature-based homomorphic encryption (MDAECH), which encrypts the data packet and stores in the blockchain for secure transmission. Moreover, the utilization of the Improved Harris Hawk Optimization Algorithm (ImHHO) finds the best cluster head within the SN and identifies the optimal path for routing. Moreover, the incorporation of these techniques in the proposed approach prevents the unauthorized access and exhibits encrypted data sharing in the network. The outcome of the ImHQCN framework attains maximum accuracy, F1 score, precision, and recall of 98.49%, 98.32%, 97.81%, and 98.83% respectively. Furthermore, the proposed model achieves a minimum energy loss of 0.182 and 0.95 privacy ratio