Cognitive Radio Spectrum Allocation Algorithm Based On No-cooperative Game Theory

The Spectrum sharing technology is one of the key technologies in the cognitive radio systems. It is well known that the cognitive radio.system is consisted of primary users (PU) and second user (SU). The primary users compete with each other for more profit and the second users aim at getting more spectrum resource during sharing the spectrum. Game theory is an effective tool for the primary user and the second user to analyze their profit and spectrum recourse, so a lot of spectrum sharing algorithms based on game theory were proposed in recent years but the algorithms got some limitations. For example, during sharing the spectrum, the algorithms didn’t consider the impact on the profit and positivity of the primary users which was caused by the primary users and the second users when making use of the spectrum. The difference of requirements for each second user and the difference of the spectrum for release and the impact on the cognitive users when the primary users using the spectrum were not considered in the current algorithms. Directing at these limitations of the algorithms, the improved algorithms was proposed in this thesis.First of all, the latest development and research of the spectrum sharing based on game theory in the cognitive radio systems was reviewed and the feasibility of the spectrum sharing using the game theory was discussed in this thesis.Secondly, the proper Bertrand model was adopted by primary users to study the price game strategy among the different primary users. After the simulation finished, the simulation result showed that the Bertrand model was fit for the primary users, and the price converged to a proper level which called the Nash equilibrium price, simultaneously, the simulation result also proved that when the Nash equilibrium price was achieved, the primary users got the best profit and the strategy was the best. Aimed at the problem that the current algorithms didn’t consider the impact on the profit and positivity of the primary users, which was caused by the primary users and the second users when making use of the spectrum, an enhance factor which was named enhancer was proposed in this thesis, besides considered the spectrum usage rate of the primary users, the inverse demand function was improved. The effectiveness of the improved algorithms was proved by the theoretical analysis and simulation and the Nash equilibrium price was achieved.At last, the spectrum sharing between the second users was studied by adopting the Cournot model. Based on the analysis about the current algorithms of spectrum sharing using the game theory for the second users and specific to the situation that most of the pricing function in the current spectrum sharing algorithms didn’t consider the difference of the spectrum, an improved pricing function for the primary user was proposed to reflect the fairness better and make sure that the spectrum the second user got could satisfy the needs of themselves in the best way. Furthermore, considered the difference of spectrum requirements of each second user and the impact which was caused by the usage rate of the primary users on the profit and spectrum selection strategy of the second user, an improved pricing function which was called the pricing function for second user was proposed in the way that the static spectrum requirements parameter was replaced by a spectrum requirement function which was volatile. The improved pricing function for the second user was studied in theory and simulated in the MATLAB and identified that the proposed pricing function for the second user was effective and the price can reach the Nash equilibrium price.