Call for papers - Special issue on privacy preserving analytics for IOT streaming systems

Internet of Things (IoT) systems are widely integrated in modern life, from industry applications to public transportation and personal health care. IoT systems continuously monitor the cyber physical and collect vast amount of data. Various analytics are developed to analyze IoT data and discover the business or societal values for their applications, e.g., home surveillance, smart readers, and location services. While we as a society greatly benefit from the utility of IoT data, an alarming concern of privacy breaching arises, i.e., data owners’ social and personal interests are revealed unknowingly. To address the conundrum of extracting utility from IoT data and protecting its privacy, the design of analytics is expected to jointly combine novel solutions in the security and machine learning fields. For example, encrypting generated data and enabling homographic computing can protect the data integrity; obfuscating data and disturbing the learning process with statistical noises are shown effective to guarantee the differential privacy. Moreover, due to the growing complexity and size of IoT systems, the implementation of analytical solutions needs to be scalable and adaptive to the streaming nature of IoT systems, i.e., data are continuously generated.