The massive streams of Internet of Things (IoT) data require a timely analysis to retain data usefulness. Stream processing systems (SPSs) enable this task, deriving knowledge from the IoT data in real-time. Such real-time analytics benefits many applications but can also be used to violate user privacy, as the IoT data collected from users or their vicinity is inherently sensitive. In this paper, we present our systematic look into privacy issues arising from the intersection of SPSs and IoT, identifying key research challenges towards achieving holistic privacy protection in SPSs and proposing the solutions.

翻译：物联网（IoT）产生的海量数据流需要实时分析以保持其数据价值。流处理系统（SPS）通过从物联网数据中实时提取知识来实现这一任务。这种实时分析虽能为众多应用带来益处，却也可能被用于侵犯用户隐私——因为从用户或其周边环境采集的物联网数据本身就具有敏感性。本文系统性地审视了流处理系统与物联网交叉领域产生的隐私问题，揭示了在流处理系统中实现全栈隐私保护的关键研究挑战，并提出了相应解决方案。