본문 바로가기
  • Home

Network-Aware QoS Optimization for MQTT Broker-Based Message Delivery

  • Journal of Internet of Things and Convergence
  • Abbr : JKIOTS
  • 2025, 11(4), pp.91~98
  • Publisher : The Korea Internet of Things Society
  • Research Area : Engineering > Computer Science > Internet Information Processing
  • Received : July 15, 2025
  • Accepted : August 8, 2025
  • Published : August 31, 2025

Jueun Jeong 1 PARK SEOYEON 1 Yeunwoong Kyung 2

1국립공주대학교
2서울과학기술대학교

Accredited

ABSTRACT

In the Internet of Things(IoT) environment, Message Queuing Telemetry Transport(MQTT) is a widely used message transmission protocol based on its lightweight and efficiency. However, in the standard MQTT architecture, subscribers statically configure their Quality of Service(QoS) levels without considering network conditions, which can lead to unnecessary overhead or data loss under depending on changes in transmission environments. In this paper, we propose a new QoS-optimized transmission method where the broker evaluates network conditions in real-time, then dynamically selects and adjusts to the lowest Quality of Service (QoS) level that satisfies the allowable latency and loss rate required by the subscriber's application class before delivering the message. The proposed approach allows the broker to recommend an appropriate QoS level at the initial subscription stage to prevent misconfiguration and adaptively adjust the QoS level at the time of message delivery, thereby improving both efficiency and reliability. Equation-based performance analysis demonstrates that the proposed method adapts flexibly to changing network conditions, especially in environments with irregular or long publishing intervals, and effectively reduces packet loss by increasing the QoS level when the network becomes unstable. Future work will focus on designing and experimentally verifying a method to more actively control QoS according to subscriber requirements by considering the trade-off between reliability and latency.

Citation status

* References for papers published after 2024 are currently being built.