ABSTRACT
Cyber physical system (CPS) can monitor the environmental conditions using wireless sensor networks (WSNs). WSNs comprise of large number of sensor nodes that are densely deployed in a geographical area where specific events have to be monitored or detected. These sensor nodes are low power devices with limitations on computing capability and storage capacity. These sensor nodes sense the occurrence of events and transmit the sensed data to the base station. Sensor nodes produce huge amount of data which needs to be aggregated for effective summarization. False data injected by the compromised sensor nodes will distort the summarized results. Most of the existing security mechanisms in WSNs employ trusted aggregator nodes for eliminating false data injected by the compromised sensor nodes exploiting the spatiotemporal correlation. They do not address the issue of compromised aggregator nodes. In this work, a secure framework for environment monitoring using CPSs is presented. CPS controls and monitors the system using security algorithms, integrated with the internet and its users. This employs an outlier analysis technique 76to eliminate the false data injected by the compromised nodes including the aggregator nodes ensuring secure data aggregation. Further, this scheme provides confidential data transmission and effectively handles the other inside attacks. Simulation results show that about 95% of the injected false data are filtered from the network even if 25–30% of the sensor nodes are compromised.
