Coalition Game For Information Sharing In Flying Ad-hoc Network

Flying Ad-Hoc Network(FANET)is ad-hoc network formed by the Unmanned Aerial Vehicles(UAVs),which provides an effective solution for the UAV communication.Coalition game is the theory of the strategic decision making of mutual cooperative nodes.In the thesis,the coalition game is applied to the FANET to realize the information sharing between the UAVs.Two kinds of coalition game are used to solve the problem of information sharing.Firstly,coalition formation game is applied to study the information sharing problem.It is divided into two phases.In the first information exchanging phase,reconnaissance UAVs exchange the different information among themselves.In the second information distribution phase,task UAVs obtain the information from the reconnaissance UAVs.Both of the two phases are based on the coalition formation game.In the information exchanging phase,the coalition formation algorithm can reduce the delay of information sharing.In the information distribution phase,task UAVs which are out the range of reconnaissance UAVs can achieve the information with the help of other members of the coalitions.Simulation results show that the proposed algorithm performs short information sharing delay and improves reliability compared with the non-cooperative approach.Secondly,based on the coalitional graph game,the optimal route of UAVs information sharing is studied in order to reduce the time required for information sharing.Due to the limitation of the UAVs' resources,some nodes may be dynamically transformed into non-cooperative nodes.Belief update mechanism is proposed to analyze the type of the UAVs to choose well behaved ones.Based on coalitional graph game,the cooperative UAVs form into the coalition which has the largest value,and share the information according to the route of the coalition.The simulation results prove that compared with the non-cooperative method,the coalitional graph game can reduce the time of information sharing.