Next Generation Secure and Collaborative Internet of Things (IoT) System - Leveraging Blockchain for Decentralized Control and Privacy Preserving Machine Learning

点击翻译按钮获取中文摘要

When massive Internet of Things (IoT) devices are connected to the network in near future, we can foresee the surge of people's IoT data and the huge requirements of data analysis and processing for better humancentric services. It requires the IoT devices to collaborate and protect the people privacy. The next generation secure and collaborative IoT system can be widely used to provide the automation and intelligence in many fields from smart grid to autonomous driving, to supply chain and warehouse inventory management, and to industry 4.0. However, existing information system may not be able to support this goal. This is not only due to the explosion of the number of IoT devices and the amount of data to be transmitted, saved, and analyzed, but also because of the privacy concerns as people cannot fully control their data. This is because servers may suffer from the distributed denial of service attacks and/or single-point-of-failure vulnerabilities and users' data would be abused if the database is compromised. The blockchain with decentralized data storage and processing abilities is promising to fix this gap. In blockchain, data is not saved in the database but recorded in blocks permanently upon the consensus algorithm. Besides, the smart contract can run the same computer code anywhere to process users' data near its source in a decentralized and ubiquitous manner. Along with the evolving technology and users' demands, we had adopted the distributed system as the solution to deal with the burst of smart mobile devices in the last decade. Now, the decentralized architecture shall be the solution to deal with the device and data explosions in the next ten years. Considering authentication and authorization requests from the huge number of devices as well as the date storage and processing demands with privacy and integrity requirements, classical techniques have difficulties to deal with these problems. Our overall objective is to enable billions of IoT devices to work in a secure and collaborative way to provide the harmonic services to humans. Correspondingly, we shall develop the framework for the decentralized security control, develop the coordination protocol to optimize the efficiency of decentralized privacy preserving machine learning, develop protocols with game theories to incentivize data-sharing and truth-telling in the learning system. The techniques developed in this project will form the basis of future industry partnership projects where these techniques can be first evaluated based on the practical scenarios and then implemented in testbed for experimental verification of our work and proof-of-concept technology transfer. This project will provide an excellent opportunity to train the engineers and researchers working with blockchain technology who are knowledgeable on the use of contemporary machine intelligence and decentralization techniques to address the complexity and dynamic nature of the next generation IoT system.