Research On Key Technologies For Cognitive Wireless Network Performance Optimization And Evaluation

With the rapid development of the wireless communication technologies and the wireless services,the spectrum resources under 10 GHz is increasingly becoming saturated.But meanwhile,the unbalance distribution of wireless services and users results in the low utilization rate of spectrum resources.Cognitive Radio(CR)technology,which allows cognitive users to access the authorized spectrum dynamically,can effectively improve spectrum utilization efficiency and meet QoS requirements of cognitive users without affecting the communication quality of the primary users.This thesis covers relevant application scenarios of Cognitive Radio Network(CRN),and provides the intense research on the subjects such as the energy efficiency and spectrum efficiency performance optimization problems,and assessment techniques of Phase-locked loops parameters for channel switching in CRN.The ingenious solutions are proposed,and the related algorithms are verified by theoretical analysis and simulation experiments.The main contents and innovations of this research are summarized as follows:First,in order to ensure the QoS for data transmission services with different priority in the RF(Radio Frequency)energy harvesting cognitive radio network(RF-CRN)with underlay spectrum sharing mode,we introduce an innovative joint optimization algorithm of energy harvesting time and transmission power to maximize the enery efficiency of cognitive user.1)we establish the system energy efficiency model according to the working principle and QoS of RF-CRN,then the original optimization problem is decomposed into two subproblems w.r.t.the harvesting time and transmission power by using of coordinate ascent method.And we employ the Charnes-Cooper Transfermaton to equivalently convert the subproblems into convex optimization problem.An algorithm based on the joint optimization of time and power is proposed,and then we theorectically prove the proposed algorithm convergence.2)According to the assumptions about the residual energy and channel small-scale fading,we remodel the energy efficiency optimization problem for RF-CRNs,then we derive the equivalent conversion of the optimization problem based on Dinkelbach method and then a joint optimization algorithm w.r.t the time and power based on the Lambert W function and Lagrangian duality method,and the optimal condition for system resource allocation is obtained;Second,in order to maximize the capacity performance of the secondary users(SU)network for the CRN with Interweave mode,we originally do the research on the SU network system capacity optimization problem based on the supermodular game theory,and a distributed algorithm for interference compensation and power allocation is proposed in this chapter.At first,the capacity optimization problem of cognitive radio Ad hoc network(CRAHN)system is modeled based on the aspect of spectrum sensing and data transmission,and is transformed into an equivalent game model.To describe the cost of inter-user interference,a new price factor is introduced by the derivation of KKT condition.Then,we study the supermodularity condition of the equivalent game model under different SINR settings.At last,we realize the distributed power allocation algorithm design based on the Gradient Play method,and verify the network capacity and algorithm convergence performance through the simulation experiments.Finally,in order to evaluate the phase-locked loop parameters which affect the performance of subuser channel switching and receiving demodulation in the cognitive radio based telemetry network scenario,a real-time measurement technique for the loop bandwidth and loop gain of phase-locked loop(PLL)embeded in network nodes is proposed at the first time.At first,we discuss the influence factors of network performance based on shannon capacity equation,and analyze the impact effect of PLL loop bandwidth on user channel switching and signal receiving carrier synchronization.Then,a robust real-time measurement method for the loop parameters is proposed,which support the real-time loop bandwidth and gain measurement in a closed loop PLL system.Hopefully,it can be combined with adaptive technologies to optimize the network performance.The correspondence between theoretical analysis and experimental results is verified through real system and simulation experiment.This technique provides an effective solution for the performance evaluation of the control loops in the CRN user nodes.