Preserving the Privacy in IoT Medical data in Edge Computing using Modified AES Model with Discrete-Time Alternating Quantum Walk

The rapid growth of IoT edge devices has caused in an unprecedented explosion of data volume. Extremely often, the created data is gathered and disseminated via edge-based IoT structures. When hostile IoT devices make requests to steal private data from a cloud storage system via edge nodes, the data-sharing privacy exposure issue becomes even more concerning. We provide evolutionary privacy protection learning algorithms for an IoT data sharing scheme based on edge computing as a solution to the stated problem. Deep learning of medical models in an IoT-based healthcare scheme is described here via a federated learning approach. Using a combination of the alternating quantum walk (AQW) and the Advanced encryption standard (AES), this research suggests a new method of picture encryption. A keystream generator associated with AQW limits is used to build a likelihood distribution matrix, and the study makes use of quantum features to enhance the AES algorithm. To unlock the AES method, we must first extract certain singular values from the matrix. Elements of the probability distribution matrix stand in for Rcon in the AES algorithm. The AES algorithm's S-box and ShiftRow transformation mapping rules are then jumbled based on the increasing size of the clone probability distribution matrix. The preprocessing is done with a likelihood distribution matrix and a plaintext XOR operation, and the encryption is done with a modified version of the AES algorithm. To prevent an adversary from concluding confidential medical data via assaults like a perfect reconstruction attack, etc., this encryption technique may be applied to better safeguard local models.