Network administration tool for the Internet of Things

This thesis aims at developing a monitoring tool for administrators of the Internet of Things (IoT) networks. Internet of Things become more prevalent nowadays. It enhances our productivity by automating diverse aspects of our life without needing humans to intervene by disposing of small embedded devices that communicate between themselves. One such application is the Wireless Sensor Networks in which sensor nodes periodically sense the environment by means of sensors and send those results to a central server which will analyze the received data. This could be used to retrieve the humidity of a forest. However, as of today no solution exists to monitor the traffic generated by such networks. Administrators do not dispose of a solution to see how much the devices composing the IoT (Internet of Things) networks send packets and also to whom. In this thesis, we propose a solution to that issue. Our solution consists of two parts: the retrieval of information from the IoT network and the displaying of the information retrieved via a Graphical User Interface (GUI). To retrieve the information, we propose to use IPFIX, more precisely TInyIPFIX which is a lighter version. Nodes register flows which are the destination to whom they sent data, plus the amount of bytes and packets to that particular destination. After a timeout occurred, the node exports the flows using the TinyIPFIX format to the collector server. The gateway node converts each TinyIPFIX format to IPFIX format. This assures that our solution follows a common well used standard. To display graphically the information received, we developed a web solution using Node.js that shows the current topology of the nodes. It also displays the amount of bytes generated by the network with some meaningful statistics like the top nodes that sent the most bytes. The tool is modular permitting users to add plugins easily. The results showed that our solution induced a cost in term of energy for nodes composing the IoT network, but that by using aggregators, which merge multiple TinyIPFIX packets into one, the average energy consumption could be reduced quite a lot.