Privacy Preserving Secure Data Storage scheme based on Adaptive ANN and Homomorphic Re-Encryption Algorithm for Cloud
Prasanta Kumar Bal, Sateesh Kumar Pradhan · 2019
Intrusion detection systems (IDSs) protect the communication networks and computer systems by supervising data traffic and finding malicious activities that represent uncommon network or system behavior. From data instances that correspond to the normal behavior of the network, the IDS learn the data model of normality which is known as inliers. Also, they detect the malicious activities that are capable of detecting intruders and generating the alarms when any illegitimate activity occurs in a cloud environment. As the cloud users are growing rapidly, the security and privacy concern also increase. As the users’ data are being managed by the third party, data protection has become a primary issue. So as to handle the mentioned issue, an IDS has been presented to find any malicious activity on a cloud environment that is capable of detecting intruders and sends a warning when any illegitimate activity occurs. For detecting normal and malicious data, the Adaptive artificial neural network (AANN) has been utilized in the proposed scheme. For optimally designing the training structure of presented AANN based IDS, the grasshopper optimization algorithm (GOA) has been utilized in the novel approach. This further enhances the intrusion detection. Then the homomorphic reencryption technique has been utilized for improving the security further. Further, a third party auditing (TPA) has been performed for checking the data periodically. Finally, the data have been stored in the cloud. Using the cloud simulator, the presented scheme has been implemented in the working platform of Java software.