Research On Spectrum Allocation Under Non-Cooperative Game Conditions

With the development of distributed cognitive communication network and technology,it is impossible for users to implement a unified spectrum allocation strategy like a centralized cognitive communication network.At present,the spectrum strategy based on non-cooperative game is becoming a resear-ch hotspot.This thesis firstly combines the system environment parameters with the utility function of cognitive users,and gives the spectrum reconstruct-ion process and the required modules.By constructing a non-cooperative game model,the multiple spectrum resources allocation equilibrium strategy in this scenario is studied.The corresponding utility function is designed between users and spectrum agents who hold multiple spectrum resources.The competitive behavior in the game model is analyzed and summarized.It is proved that the proposed game model has a unique Nash equilibrium solution.Further,a distributed dynamic iterative algorithm is applied to dynamically adjust the unit selling price of each spectrum resource.The simulation results verify the effectiveness of the game model proposed in this thesis.Secondly,this thesis introduces the penalty factor to study,discuss and solve the problem of fairness among cognitive users.As cognitive users continue to increase their demand for spectrum,the higher the cost and expense they pay,resulting in less revenue for themselves,which makes it impossible for a single cognitive user to purchase spectrum resources indefinitely in order to maximize their own interests.Finally,this thesis studies a heuristic stochastic search algorithm-Fireworks algorithm to solve the Nash equilibrium solution of the proposed game model.Considering the slow convergence of the original fireworks algorithm,this thesis proposes a dynamic search fireworks algorithm based on simulated annealing.The simulation results show that the proposed improved algorithm can quickly converge to the Nash equilibrium solution,and obtains an optimal spectrum allocation strategy among cognitive users,which also ensures the fairness between cognitive users.