AI-PRISM

The AI-PRISM team published a scientific paper on a new optimisation mechanism to improve the usability of LPWAN networks in dynamic scenarios

Researchers David Todoli Ferrandis, Javier Silvestre Blanes, Victor Sempere Paya and Salvador Santonja Climent from partners ITI and the Polytechnical University of Valencia (UPV) published the scientific paper entitled “Adaptive Beacon Period Configurator for scalable LoRaWAN downlink applications” in the IEEE Access Journal. The publication is relevant for the AI-PRISM project, specifically for the industrial wireless sensor networks and their integration within the human-centred collaborative robotic platform.

The AI-PRISM team aims to develop the system and infrastructure that will enable the integration and control of both already existing solutions and new AI-based tools in development. New advanced and flexible communications technologies, both wired and wireless, are the basis for supporting the convergence of IT and OT traffic. On the shopfloor a flexible wireless sensing layer based on robust IoT (Industrial Internet of Things) technologies will be used to break the rigidity of current industrial architectures and communication technologies to support dynamic elements of sensorisation and digitalisation, and their seamless integration with AI-based tools and robotic devices. A network solution based on SDN, IEEE802.15.4e and Time Synchronized Channel Hopping (TSCH) will be developed to ensure robustness, real-time behaviour and security. For applications where Real-Time is not a relevant feature, but mobility and scalability are, low-power wide-area networks (LPWAN) arise as the most promising technologies to bring context and process information to collaborative robots.

This publication is related to these LPWAN networks, which are commonly used because they meet the requirements of Internet-of-Things (IoT) networks with many end devices, such as high network scalability, wide area coverage, low data rates, and delay tolerance while consuming very little energy. The LoRa wide-area network (LoRaWAN) is one of the most popular solutions, supporting three types of medium access control (MAC) options to handle distinct application demands. Class B shortens downlink frame transmission latency while maintaining low energy consumption in the end device, which makes it suitable for applications that require sending messages to devices, sensors, and actuators in the floor plant.

This paper analyses the operation of gateways with class B devices to determine the events that influence scalability and performance, presents an analytical model to describe these systems, and proposes an Adaptive Beacon Period Configurator (ABPC) optimisation mechanism. ABPC changes the time-related parameters configuration to improve the usability of these networks in dynamic scenarios. The proposed solution is then simulated and tested against an analytical model. 

During the simulation test, researchers analysed the trade-off between message waiting time, reception probability and energy consumption of an end device. The results showed how traffic density increases impact on these Key Performance Indicators (KPI) and how to try to guarantee these requirements in a network deployment.

The full paper is available here !

Do you want to learn more about AI-PRISM research and developments? Subscribe to our newsletter and follow us on LinkedIn and Twitter, so you don’t miss a thing!

Privacy Preference Center