Cooperative Spectrum Sensing Using Game Theory In Cognitive Radio Networks

In the last years,the growth of new wireless devices helped to increase scarcity in resources and cognitive radio networks,give a ground-breaking answer for the expanded demand for spectrum.This thesis cornerstone is the concept of spectrum sensing,that consists in the detection of primary user's presence,since the phenomena that influence on the system like shadowing,spatial diversity and multipath.The next key concept is a cooperation where all the users interact with each other in order to detect effectively,in spite all the previous phenomena.The last is the concept of game theory which presents social,economic,or any type of problem extrapolated into a game-in this case the cooperative decision based is the more suited.The research focuses on the user's demand cooperative spectrum sensing and access strategy in a cognitive radio network,with multiple SUs and multiple primary users.The methodology proposes that each user decides the channels to access and sense with other users,depending on its own traffic demand,additionally can decide which channel to sense and access.The energy can be saved for a future transmission,when the user doesn't have data to transmit.First of all,the research explores the cooperation strategy based on traffic demand,in which each user senses a channel in a determined time span.The coalition is presented with a non-transferable utility,where all users receive a value that corresponds to the coalition,that will be used in the calculations of the throughput and energy efficiency.The coalition formation in this situation is a high complexity problem,which makes it difficult to work with,and in order to overcome this complexity,it is proposed a new coalition formation algorithm.The results show that the algorithm reaches a high throughput and energy efficiency.Then the cooperation scope extends by sensing multiple channels throughout the sensing time,increasing the strategies complexity.The game sets a non-transferable utility overlapping game.For achieving a Nash stable structure,a modification of the analyzed algorithm is introduced.This algorithm is proved to stabilize at a finite number of repetitions,increasing the spectrum efficiency.The algorithm requires an optimal information exchange between users and iterations number,achieving a high throughput in comparison with the disjoint algorithm and an overlapping formation algorithm.In this dissertation,the analysis of an adaptive transmission power control scheme is described by reducing transmission energy consumption,while ensuring higher throughput.The performance evaluation considers several different factors that affect the performance of the algorithm,such as the number of users,traffic demand,and the lower bound of energy efficiency.Finally,a snapshot of the overlapping alliance structure and the number of iterations,while implementing different algorithms are given.The simulation results prove that this algorithm is better than other algorithms in terms of system throughput,affected by user payoff.