IoT Application-enabled Deep Learning Model With Secure ECC-Based Cloud Data Storage Optimization Strategy for Data Deduplication
Marx A.H., R Tanuja · Advances in computational intelligence and robotics book series · 2025
Due to the development of the “Internet of Things (IoT),” a huge quantity of data is transferred to the cloud architecture. As a consequence, expenses and storage charges are associated with cloud servers. Storage capacity can be enhanced by implementing a deduplication technique to spot duplicate information. Both cryptography and deduplication are carried out using the full hash values of the information chunks. The deduplication systems are vulnerable to file threats. So, we developed an effective optimal key-based deduplication model. The proposed model uses attributes including filename, size, block name, size, type of file, and data pattern for deduplication. The “Long Short-Term Memory (LSTM)” model is introduced for effective deduplication performance. The LSTM model separates the attributes and deduplicated attributes once the data file is not duplicated. Once it is proved that the data are deduplicated, the edge node (client) enciphers the data by utilizing the Optimal Key-Based Elliptical Curve Cryptography (OK-ECC). Further, it passes the data to the cloud system.