DATA MINING FOR INTERFERENCE AVOIDANCE IN SMART CITIES IOT NETWORKS

Abstract

A rapid growth of the wireless communications and heavily occupied spectrum lead to an inevitable interference between the heterogenous systems operating in the same frequency band. Having in mind the development of the Internet of Things (IoT) services and networks and widely present WiFi networks on the one hand, and the fact that these two systems occupy the same 2.4 GHz frequency band on the other hand, it is clear that the control of the interference and the spectrum coordination are of the highest importance. The first step in the interference control is to acquire its properties. Since the simulation of a large IoT network is not entirely possible, due to the numerous factors not known in advance, the interference assessment is performed on the SmartSantander, an IoT testbed, located in Santander, Spain. This paper presents a statistical analysis of the sensor data and describes the interference properties and its influence. These results may be used for the spectrum coordination, together with the neural networks and semantic technologies.