Cooperative Set Management Research Based On Multi-connectivity

With the development of next-generation wireless communication technology,a variety of unprecedented services and applications have emerged-autonomous driving,telemedicine,smart factories,etc.,having put forward more stringent requirements on network performance.Millimeter wave,heterogeneous network and multi-connectivity(MC)technologies have become research hotspots in academia and industry.However,the heterogeneous MC network under dense networking not only improves the network performance,but also complicates the network structure.Each user equipment can maintain connections with more than one access points.Especially,the network management in traditional single-connectivity network is no longer applicable for the cell handover and power control in the MC network.The network management needs to consider the characteristics of the MC network to design a new solution.From the perspective of the user,multiple access points serving the same user equipment in MC network are viewed as a cooperative set,and efficient management of the cooperative set becomes the key to improve network performance.Therefore,this thesis proposes two kinds of MC network based cooperative set management schemes for cell handover in mobile scenarios and power control of access points under backhaul links with limited capacity.The main work and innovations are as follows:A cooperative set update scheme based on prediction of received signal strength is proposed.By introducing a cooperative set update scheme,this scheme solves the problem of frequent cell handover in MC networks under dense networking.In the scheme,the cooperative set update is divided into three steps:preparation,decision making,and execution.In the preparation phase,the low-band base station collects the signal strength of the millimeter-wave access points(mmAPs)for processing and generates a candidate set.In the decision-making phase,cooperative set update triggering and cell selection are performed.Using the advantages of the MC network,the signal-to-interference-and-noise ratio(SINR)based update trigger mechanism is designed firstly,then based on the recursive least squares method,the received signal strength of mmAPs in the cooperative set and candidate set are predicted,and the target cells for the cooperative set update are selected;in the execution phase,based on the above process of cooperative set update,the complete signaling interaction process is designed.Simulation results show that the proposed scheme can effectively reduce the number of unnecessary handovers in a MC network and improve network throughput.A cooperative control power control scheme based on game theory is proposed.The scheme introduces constraints on the capacity of the backhaul links.The utility functions of the mmAP and the user equipments(UEs)are designed respectively,and a multi-leader multi-follower Stackelberg game is built by maximizing the respective utility functions of both player,and backward induction is adopted to analyze the best strategy.First,given the strategy set of the mmAPs,the best response strategy of the UE is analyzed;then it is substituted back to the utility function of the mmAPs to solve the Nash equilibrium problem among mmAPs.The iterative algorithm based on sub-gradient method is proposed to obtain the game equilibrium point,and the transmit power of the cooperative set is allocated according to the equilibrium point.Simulation results show that the proposed scheme can effectively reduce the load on the backhaul links while meeting the user's requirement on services.