Performance assessment of management protocols and platforms for internet of things

Abstract

Internet of Things (IoT) is a paradigm where devices are connected to the Internet and can interact with each other, making a scenario of machine to machine communication. IoT environments need scalability, standardized communication, and context-aware requirements for managing securely and accurately connected devices in real-world environments. Efficient management of IoT networks requires consideration to the constraints of their low-power devices and the complexity of deploying the underlying communication infrastructure. An IoT management scenario is represented by a growing number of connected devices, characterized by their heterogeneity more network management protocols and platforms, and IoT devices. One of the IoT network challenges is related to devices heterogeneity where each object has different processing capabilities, memory capacity, and communication technologies. However, IoT device networks are generally characterized by devices with restricted resources and sometimes deployed in hard areas to reach. In addition, an IoT network may include a large number of heterogeneous devices that need to be managed. All these factors determine the need for management solutions tailored for devices and the IoT network. To ensure better use of IoT networks, effective and automatic management is required. This dissertation studies different solutions for network management and devices in an IoT context, identifying their characteristics and technical challenges. With respect to network management protocols and IoT devices, most of the solutions try to solve the problem of energy saving and simplicity due to the basic hardware features of the devices. There are been development of other solutions tailored for an IoT environment, but Simple Network Management Protocol (SNMP) and Network Configuration Protocol (NETCONF) continue being used on a large scale by commercial equipment. Among the IoT management platforms, OpenDayLight has gained advantage over other platforms because it is modular and capable for managing both the network and devices to support native protocols, namely, SNMP, NETCONF, Constrained Application Protocol (CoAP), and OpenFlow. However, currently, in the literature there are no management platforms that address all IoT-related issues. These technologies have been studied and discussed in detail. Based on the comparative study of the available solutions, a new IoT management platform with a friendly user interface, called M4DN.IoT (Management for Device and Network in Internet of Things), was proposed and created. This modular platform integrates and manages the individual functionalities of the devices in an IoT network as well as the state and characteristics of that network. In addition, the platform provides generic services, such as node discovery, data storage, and authentication, which are basic requirements for creating IoT applications. The solution has been evaluated, demonstrated, validated, and is ready for use in the middleware platform for IoT, the In.IoT.