The prevalence of Internet of Things(IoT)applications has generated an unprecedented volume of industrial data that create the main source of big data[1,2].How to effectively and efficiently preprocess,integrate,and analyze big IoT data from multiple sources is still a fundamental challenge[3,4].Fortunately,Artificial Intelligence(AI)has recently emerged as a key technology to achieving intelligent data analyses and scientific business decision-making.AI algorithms can process the streaming data generated by distributed IoT devices and provide powerful tools to address complex big data analytics[5,6].Therefore,the adaptation of AI-based methods is highly demanded for achieving their full potential in smart IoT applications.However,the IoT devices,which run in an unstable environment and have poor computing capabilities,are often confronted with a range of application requirements,such as quick response,secure communications,and privacy protection[7,8].Currently,lightweight security and privacy solutions specifically designed for the devices and servers operating in the IoT environment are still lacking.
展开▼
