Reliable, trustworthy sensor data collection for event detection in Cyber-Physical Systems using Genetic Algorithm and Differential Privacy

In Cyber-Physical Systems (CPS), sensors are commonly integrated to collect diverse characteristics and transmit the data to nodes located at upstream for further processing. However, the quality of data in CPS is frequently compromised by significant limitations in resources, environmental consequences, user privacy concerns and security issues. Moreover, the process of event detection by data in CPS can be complex and unreliable if the data collected by the sensor is not verified during data collection, prior to transmission, and before aggregation.This paper introduces the trustworthy and reliable data collection architecture for event detection in CPS. In the domain of CPS, this framework enables the collection of reliable data. The primary concept is to allow a sensor’s module to locally verify the reliability of data collected during an event before sending it to the upstream nodes. Additionally, it verifies the trustworthiness of incoming data prior to collection at the sink node. It aids in the detection of defective sensors.Collaborative IoT strategies, Genetic Algorithm (GA), gate-level modeling using Verilog user defined primitive, and programmable logic device are utilized to achieve reliable event detection in the proposed method. Gray coding serves to guarantee the reliability of the obtained data and aids in identifying a defective sensor. Differential Privacy is employed for protecting the data before it is transmitted to the upstream nodes.This framework proves to be reliable in data collection and is applicable in most CPS applications by performance analysis and simulations.