An Adaptable IoT Rule Engine Framework for Dataflow Monitoring and Control Strategies

TL;DR

This paper introduces a flexible, user-friendly IoT rule engine framework with a custom DSL, designed to improve data flow monitoring and control across multiple devices, addressing limitations of existing systems.

Contribution

It proposes an adaptable, extensible rule engine framework with a new DSL, enhancing flexibility and multi-device rule formulation in IoT data management.

Findings

Prototype system validates the framework's effectiveness.

Framework is easily extensible for various IoT scenarios.

Suitable for real-time control where strict latency is not critical.

Abstract

The monitoring of data generated by a large number of devices in Internet of Things (IoT) systems is an important and complex issue. Several studies have explored the use of generic rule engine, primarily based on the RETE algorithm, for monitoring the flow of device data. In order to solve the performance problem of the RETE algorithm in IoT scenarios, some studies have also proposed improved RETE algorithms. However, implementing modifications to the general rule engine remains challenges in practical applications. The Thingsboard open-source platform introduces an IoT-specific rule engine that does not rely on the RETE algorithm. Its interactive mode attracted attention from developers and researchers. However, the close integration between its rule module and the platform, as well as the difficulty in formulating rules for multiple devices, limits its flexibility. This paper presents an adaptable and user-friendly rule engine framework for monitoring and control IoT device data flows. The framework is easily extensible and allows for the formulation of rules contain multiple devices. We designed a Domain-Specific Language (DSL) for rule description. A prototype system of this framework was implemented to verify the validity of theoretical method. The framework has potential to be adaptable to a wide range of IoT scenarios and is especially effective in where real-time control demands are not as strict.