Optimal Control Policy Based on Game Theory in Wireless Sensor Networks

This paper proposes an admission control policy based on game theory in Wireless Sensor Networks (WSNs). Each sensor node is regarded as a random service server containing two states (sleep and active). A two-dimensional Markov model is established, the probability of each steady state of the system is calculated through the transition equations. To achieve optimum revenue, we considered a reward when a data packet is accepted and a holding cost per unit time when waiting for service. By analyzing net income with cooperative game theory, we can find a suitable threshold value for the sensor to receive data packets into the system. In addition, we also calculate the influence of arrival rates and service rates with different parameters, and sleep or active process on the net benefit, providing a multi-faceted solution to increase the net revenue.