Efficient Traffic Monitoring in IoT Networks for Attack Detection at the Edge

Abstract: The Internet of Things has rapidly expanded over the years, and with this comes significant security risks, with attacks increasing at an alarming rate. A way to detect attacks in the network is by having each device send traffic monitoring information to an edge device that can investigate if any devices have been exploited. Transmitting data in such a manner introduces additional data processing and network resources. In this thesis, I show that using a general-purpose compression module is a more efficient method for compressing the traffic monitoring information data to create a more efficient data transmission in the network than using a special-purpose compression module. In the investigation, I use the Arithmetic Coding algorithm for the general-purpose compression module and Concise Binary Object Representation encoding for the special-purpose compression module. This is done by comparing the two methods on artificially generated traffic monitoring data. The evaluation is based on the metrics of compression ratio, execution times, and storage requirements. In all metrics, the general-purpose compression module provided better results than the special-purpose compression module, making it the preferable method.