Privacy-Preserving of Edge Intelligence using Homomorphic Encryption

Edge intelligence has paved the way for a stronger and more secure system by combining security and artificial intelligence. A lot of research have been undertaken, revealing remarkable development and a variety of viewpoints on this novel phenomenon. We suggested a way enabling edge intelligence to move via homomorphic encryption-enabled gadgets and automobiles. This encrypted data will be delivered to an edge node for the building of a machine learning model in order to gather any potential information hidden in the data as well as any potential obstacles that may need to be addressed before any event happens. Several machine learning models, such as KNN, K-means, SVM, and others, are used to gain the best possible data analysis. Once secure, decisions will be taken, and the outcome will be visible. All machine learning training will be done in encrypted data, which will be encrypted to ensure that no one’s privacy is abused. This may be set up for a variety of machine learning modules.