Confidentiality Awareness Computing

With the rapid growth of the number of Internet of Things connections, more and more types of Internet devices are used in people’s daily life and production activities. The security risks of these IoT devices are also increasing day by day.

The Internet of Things, as an important technological means of perceiving the physical world, its main function is to collect various perception data, perform local calculations on these perception data, and connect them to cloud center platforms. In the increasing number of applications in the era of intelligent Internet of Things (AIoT), it is also necessary to perform local or cloud edge collaborative AI model inference on these perception data.

Confidential aware computing, which applies confidential computing to the collection, calculation, and inference processes of IoT aware data, is a key supporting technology for ensuring the security of data and code usage on IoT devices.

This project will study the construction method of trusted IoT software based on confidential computing, including TEE based program segmentation, TEE based model inference, and WebAssembly based lightweight isolation mechanism.