MQTT-CB: Cloud Based Intelligent MQTT Protocol

Erol, Muhammed Raşit; Bilen, Tuğçe; Özdem, Mehmet; Canberk, Berk

Abstract

The MQTT protocol, which stands for Message Queuing Telemetry Transport, is widely recognized within the IoT community as one of the most frequently utilized communication protocols. Conventional MQTT protocols described in the literature could improve their capacity to support distributed environments and scalability. In this manner , MQTT-ST is an advanced MQTT protocol that provides bridging capabilities within a distributed environment, making it a preferred option for IoT systems. In this study, we introduce a new, intelligent, scalable, and distributed MQTT-ST-based protocol named MQTT-CB. Our approach leverages containers to improve portability, and our protocol is designed with a cloud-based architecture that streamlines deployment. The primary contribution of our research is the integration of intelligent capabilities into the MQTT-ST protocol, utilizing an LSTM (Long Short-Term Memory) network, which is the leading deep learning model. Specifically, our protocol employs predictive algorithms to foresee retransmitted packets, dynamically adjusts the number of brokers in real-time, and reduces the number of brokers when clients are inactive. Our experiments demonstrate that our protocol significantly outperforms conventional MQTT-ST protocol regarding latency between subscribers and publishers. Furthermore, our protocol adapts seamlessly to changes in the publication rate. In summary , we present a cloud-based intelligent MQTT protocol that offers significant advantages over traditional MQTT-ST protocol.

Citation

Erol, M. R., Bilen, T., Özdem, M., & Canberk, B. (2023, August). MQTT-CB: Cloud Based Intelligent MQTT Protocol. Presented at 9th EAI International Conference, INISCOM 2023, Ho Chi Minh City, Vietnam

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 9th EAI International Conference, INISCOM 2023
• Start Date: Aug 2, 2023
• End Date: Aug 3, 2023
• Online Publication Date: Oct 31, 2023
• Publication Date: 2023
• Deposit Date: Oct 11, 2024
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• Pages: 236-254
• Series Title: Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering
• Series ISSN: 1867-8211
• Book Title: Industrial Networks and Intelligent Systems
• ISBN: 978-3-031-47358-6
• DOI: https://doi.org/10.1007/978-3-031-47359-3_18