An Experimental Evaluation of Application Layer Protocols for the Internet of Things

Abstract

Abstract:The Internet of Things is envisaged to interconnect billions of devices, from sensors and actuators to smart objects, computers and vehicles. The main obstacle that arises is that technology should allow physical objects, usually constrained devices, to interact with applications. Therefore, already existing technologies and patterns should be adapted to the new requirements, or novel ones must be created. In this context, the application layer protocols play an important part in orchestrating an IoT network. This paper focuses on the comparison of different protocols by carrying out experiments and prototyping a real-world interaction between IoT network actors.

Analysis on functionalities and security features of Internet of Things related protocols

“…The authors of [58] state that the AMQP protocol is not suitable for use in IoT systems composed of devices that are particularly limited in terms of energy. To demonstrate this, in [58], different simulations have been carried out comparing different protocols; for each protocol, a test environment has been set up and the number of bytes sent, the time needed to obtain a response and the number of packets sent have been measured. The results showed that AMQP is the protocol that sent the most data over the network with a longer send time than the other protocols.…”

“…It must also be considered that the AMQP protocol exploits the TLS protocol to guarantee security. In conclusion, as described in [59] and [58], the use of less expensive protocols is recommended in such scenarios.…”

An IoT Environmental and Motion Assistance System for the Smart Mobility Context

“…The Internet of Things is envisaged to interconnect billions of devices, from sensors and actuators to smart objects, computers and vehicles (Năstase, Sandu & Popescu, 2017). Smart mobility represents an important aspect embedded in the larger multidimensional smart city concept, characterized by personalized environments capable of providing continuous remote monitoring via sensor networks (Coardoș & Marinescu, 2020).…”

On the rewards of self-adaptive IoT honeypots

“…If they do not match, the attack will not be launched therefore no valuable data will be collected. Such elements make the data collected to be of low quality and quantity [37,38].…”