Complex Event Recognition Notification Methodology for Uncertain IoT Systems Based on Micro-Service Architecture

Abstract

One of the main purposes of the Internet of Things (IoT) systems is to provide information on the observed physical system and to notify when a certain observed complex event occurs in the system. The biggest challenge of the IoT systems is that the systems collect a massive amount of uncertain data from diverse IoT devices, where the devices are connected through the internet. In addition to that, some events are inferred from other events and uncertainty is propagated from parent events to the inferred events, which additionally contributes to overall system uncertainty. The observed complex events are a complex relationship of primitive events that are produced by IoT devices and collected in IoT system. There were a couple of attempts to quantify the system uncertainty by estimating the likelihood of the occurrence of events of interest while taking into account the uncertain data and uncertain parameters of the model of the physical system with rule-based complex event processing and/or Bayesian network. These methods were implemented as stand-alone software solutions that are unable to scale under heavy loads of incoming events. This paper proposes a micro-service based notification methodology that uses complex event recognition to handle the IoT system uncertainty.