Harnessing personal data from Internet of Things: Privacy enhancing dynamic information monitoring

Summary form only given. The Internet of Things (IoT) brings significant potential to enable many applications ranging from smart metering to remote health monitoring. The amount of personal or sensitive information collected in devices and sensors is constantly growing. There is an increasing need to harness and aggregate such data in real time for monitoring and analytics while ensuring data integrity and guaranteeing privacy and security of the data. In this talk, I will present our PREDICT project on privacy and security enhancing dynamic Information Collection and Monitoring to address these challenges. I will present a novel framework for aggregating personal data collected from individual sensors in real time with the state-of-the-art differential privacy guarantee. In addition, I will present a comparative study of distributed differential privacy schemes in which data can be aggregated from distributed data sources without a trusted aggregator using multiple party secure computation protocols such that the sources do not reveal their data to each other and the aggregation result guarantees differential privacy. I will present several case studies using real-world data to show that our proposed approach can be used for a variety of monitoring applications such as flu and traffic monitoring while ensuring privacy and security of the data.