Optimal transmission strategy for CPS based on Stackelberg gaming theory
Date of Issue2018-09-11
School of Electrical and Electronic Engineering
According to definition, the Cyber Physical System(CPS) is a mechanism where component or system is controlled or monitored by computer or controller, which is tightly integrated with the network and users. Embedded computers often use feedback loops to monitor and control physical processes, and in turn, these physical processes affect calculations and vice versa. However, there is always the threat of eavesdropper during the transmission process of CPS, where wiretapping from the sensor can lead to serious consequences. Previous works covering this issue all introduce a ‘friendly’ jammer to interfere with the wiretapping behavior of eavesdropper. However, introducing the jammer influences sensor utility, making the jammer and sensor compete for channel capacity. Most researchers attempt to use gaming theory to remedy this circumstance. To this, there are multiple gaming theories we can select depending on model, such as Nash equilibrium or Cournot Competition, etc. In this dissertation project, we come up with an algorithm based on the Stackelberg Gaming theory to optimize the power utility of sensor, named the Stochastic Algorithm with Feedback (SAF) algorithm. In the previous works, most researchers only consider the utility function of one or two models for the sensor and jammer. Here, we study the different behavior exhibited in 4 cases and compare the result. Additionally, the Bit Error Rate (BER) is also taken into consideration within this report to evaluate the algorithm. With the simulation result about utility function and BER, we discover that the SAF algorithm can optimize sensor utility and stabiliz BER even the channel environment gets worsens.
DRNTU::Engineering::Electrical and electronic engineering