An architecture for the analysis and detection of anomalies in smart city WSNs

In the last few years, Wireless Sensor Networks (WSN) are gaining importance as a data collection mechanism for smart city systems. The development, deployment and operation of these networks involve a wide and heterogeneous set of technologies and participants. In many cases, city councils have outsourced the implementations of their WSNs to different external providers. This has resulted in a loss of control and visibility over the security of each individual WSN and, as well, over the entire system as a whole. In this article, we first describe the security problems related to the present model of WSN implementation within smart city systems. Then, we propose a non-intrusive architecture to recover part of the lost visibility, detect attacks on the WSNs operated by third parties, increase control over the providers and, in general, improve the security of the smart city from a holistic perspective.