11th International Conference on Body Area Networks

Research Article

Scalability of Kura-extended Gateways via MQTT-CoAP Integration and Hierarchical Optimizations

  • @INPROCEEDINGS{10.4108/eai.15-12-2016.2267595,
        author={Paolo Bellavista and Alessandro Zanni},
        title={Scalability of Kura-extended Gateways via MQTT-CoAP Integration and Hierarchical Optimizations},
        proceedings={11th International Conference on Body Area Networks},
        publisher={ACM},
        proceedings_a={BODYNETS},
        year={2017},
        month={4},
        keywords={internet of things (iot) scalability cloud-sensor interaction iot gateways kura californium mqtt coap},
        doi={10.4108/eai.15-12-2016.2267595}
    }
    
  • Paolo Bellavista
    Alessandro Zanni
    Year: 2017
    Scalability of Kura-extended Gateways via MQTT-CoAP Integration and Hierarchical Optimizations
    BODYNETS
    EAI
    DOI: 10.4108/eai.15-12-2016.2267595
Paolo Bellavista1,*, Alessandro Zanni1
  • 1: University of Bologna
*Contact email: paolo.bellavista@unibo.it

Abstract

The introduction of innovative Internet of Things (IoT) gateways, powered by efficient support protocols and tools, has the potential to boost IoT development and related opportunities. Some first open-source platforms and solutions have been developed along that direction, primarily with the goals of targeting device limitations in terms of available resources and of simplifying management operations in heterogeneous environments. This paper originally proposes a multi-protocol IoT gateway solution with the specific and main goals of enabling efficiency and scalability via proper MQTT-CoAP integration. To this purpose, we have extended the Kura framework, which exploits only MQTT for machine-to-machine communication, by implementing a significant CoAP support for scalable hierarchy-based coordination, capable of externally exposing any kind of resource with REST APIs. The paper also reports an extensive set of in-the-field measurements that show the scalability of the proposed prototype that, to achieve the desired performance, also exploits some original optimizations introduced by our work on both Kura and Californium frameworks.