Game theoretic approach in optimal power allocation and anti jamming support in cognitive radio networks

Abstract

With the growing demand in the wireless spectrum in recent years, the efficiency of spectrum utilization need to be improved to cater the everincreasing data rate requirements. To be explicit, the optimization of spectrum usage is in general, a multi-objective optimization problem, which is incredibly difficult to analyze and solve. This thesis deals with two major issue in CRN i.e., power allocation and combating jamming. Effective resource allocation in cognitive radio network can be implemented by applying game theory in power allocation based on pricing model. Earlier research focuses on the impact of game models such as Cournot, and Bertrand in formulating the utility function under the constraint power and service quality. In this thesis first, the condition for optimal power allocation in CRN using Stackelberg game where the previous user decides its output and then the erstwhile user does so, knowing the output characteristics by the former user, is formulated. Based on the profit of PU (as a leader) and SU (as a follower), the optimized solution is formulated for power and interference price and it is modeled as a convex function of transmission power achieved by Nash equilibrium which involves backward induction. The proposed Resource Allocation using Stackelberg Game (RASG) algorithm tries to optimize uniform pricing and power allocation among SUs such that maximizes throughput and fairness. The simulation results show a significant enhancement in throughput and fairness compared to power optimization based on Bertrand and Cournot game theory. newline