The synergy of complex event processing and tiny machine learning in industrial IoT

Abstract

Focusing on comprehensive networking, the Industrial Internetof-Things (IIoT) facilitates efficiency and robustness in factory operations. Various intelligent sensors play a central role, as they generate a vast amount of real-time data that can provide insights into manufacturing. Complex event processing (CEP) and machine learning (ML) have been developed actively in the last years in IIoT to identify patterns in heterogeneous data streams and fuse raw data into tangible facts. In a traditional compute-centric paradigm, the raw field data are continuously sent to the cloud and processed centrally. As IIoT devices become increasingly pervasive, concerns are raised since transmitting such an amount of data is energy-intensive, vulnerable to be intercepted, and subjected to high latency. Decentralized on-device ML and CEP provide a solution where data is processed primarily on edge devices. Thus communications can be minimized. However, this is no mean feat because most IIoT edge devices are resource-constrained with low power consumption. This paper proposes a framework that exploits ML and CEP's synergy at the edge in distributed sensor networks. By leveraging tiny ML and µCEP, we now shift the computation from the cloud to the resource-constrained IIoT devices and allow users to adapt on-device ML models and CEP reasoning rules flexibly on the fly. Lastly, we demonstrate the proposed solution and show its effectiveness and feasibility using an industrial use case of machine safety monitoring. CCS CONCEPTS• Computing methodologies → Machine learning approaches; • Computer systems organization → Real-time systems; Sensor networks; • Information systems → Data streaming.

An automatic unsupervised complex event processing rules generation architecture for real-time IoT attacks detection

“…This group is the least common as it requires unsupervised training and the existence of rules capable of detecting items of interest. However, we can find works such as that of Ren et al [14]. This one focuses on optimizing performance in IoT environments, which is the main differentiating factor with respect to other proposals.…”

“…To this end, network packets are defined as simple events and the detected attacks are the resulting complex events. The successful deployment of CEP engines in IoT environments has been widely demonstrated [13][14][15][16]. Although CEP is very advantageous for real-time attack detection, it has one limitation, namely the need for a domain expert who is able to define the rules that must be followed to carry out such detection.…”

Incremental Online Learning Algorithms Comparison for Gesture and Visual Smart Sensors

“…We used an SMT32 Nucleo F401-RE equipped with an accelerometer shield as a test platform. This case study can be extended to other real-life applications such as industrial condition monitoring or anomaly detection [5], [6]. The second case study compares different CL algorithms to classify instances from the MNIST dataset [7].…”

Artificial intelligence and internet of things in manufacturing decision processes

“…Next, the planning and optimization of the AI domain are considered together with the IoT domain's cloud/fog/edge architecture in [49], [70]. Cloud architecture centralizes data storage and processing in remote servers, fog architecture distributes processing to the network edge, and edge architecture pushes computation directly to IoT devices.…”

“…With the planning and optimization of the AI domain, each architecture offers distinct advantages in terms of scalability, real-time processing, and resource utilization, allowing for tailored solutions based on the specific use case. Ren et al [70] proposed a data-centric approach to perform decentralized tiny ML and micro complex event processing (CEP) computation on IoT edge devices.…”