Exploratory approach for network behavior clustering in LoRaWAN

Abstract The interest in the Internet of Things (IoT) is increasing both as for research and market perspectives. Worldwide, we are witnessing deployment several IoT networks different applications, spanning from home automation to smart cities. majority these deployments were quickly set up with aim providing connectivity without deeply engineering infrastructure optimize network efficiency scalability. now moving towards analysis behavior such systems order characterize improve their functionality. In systems, many data related device human interactions stored databases, well information level (wireless or wired) gathered by operators. this paper, provide a systematic approach process wide area wireless platform based on LoRaWAN (Long Range Wide Area Network). Our study can be used profiling devices, group them according characteristics, detecting anomalies. Specifically, use k -means algorithm packets radio behavior. We tested our real where entire captured traffic proprietary database. Quite important fact that captures, via interface, multiple Indeed was performed 997, 183 2169 devices involved only subset known considered operator, meaning an operator cannot control whole system but contrary has observe it. able analyze clusters’ contents, revealing results line current alerts malfunctioning remarking reliability proposed approach.