| The exponential growth of wireless devices and the huge demand of wireless network services have led to severe radio frequency spectrum scarcity and unprecedented energy consumption issues.To solve the problems of spectrum loss and energy consumption at the same time,combining full-duplex technology with simultaneously information and power transfer technology is the new direction for green and efficient communications.This thesis studies the resource allocation algorithms and optimal transfer schemes in FD-SWIPT systems,and the main work of the thesis is as follow:1.Aiming at the existing problems of simultaneously wireless information and power transfer scheme in full duplex wireless system,this thesis proposes an optimal scheme based on weighted sum rate maximization in multiuser full-duplex system.Differing from traditional scheme,the source node and destination nodes in the scheme both work on FD mode.Assuming that the channel state information is known,the downlink channel implements SWIPT by radio frequency and the uplink channel transmits information through TDMA technology.Besides,the receivers are equipt with power splitting construction and harvest energy from interference signal.Therefore,this thesis designs a joint optimization scheme of time switching factor and power splitting coefficient to maximize the system weighted sum rate.Then the optimal solution of the time switching factor is derived by Lagrangian multiplier method,and an iteration algorithm is proposed based on golden section to optimize the power splitting coefficient,meanwhile the thesis verifies its convergence and analyzes its complexity.Simulation results demonstrate that the proposed algorithm and scheme in this thesis achieve better performance.2.Differing from the existing work of FD-SWIPT systems,this thesis optimizes the sum rates of multi-antenna FD-SWIPT system and considers its robustness and energy efficiency,also,this thesis proposes robust precoding design and energy-efficient timeslot allocation scheme for FD-SWIPT system.Firstly,in order to reduce the influence of self-interference,the self-interference is cancelled based on singular value decomposition and mean squareerror algorithm.Then,with the fixed timeslot allocation factor,on condition that there is evaluated error for self-interference channel,this thesis proposes the robust precoding design for the original problem based on first-order Taylor's series approximation,and the corresponding iterative algorithm is given.Furthermore,the timeslot allocation schemes under different constraints are proposed,and the optimal solution of timeslot allocation factors is deduced.Finally,the performance advantages of the proposed scheme are verified by comparing the simulation experiment with the existing results.3.In order to overcome the problem that the traditional linear energy harvesting model is not practical,this thesis presents a practical nonlinear energy harvesting model which considers the sensitivity and saturation characteristics of the circuit.Furthermore,for purpose of solving the problem of spectrum requirement and energy consumption of wireless relays,the proposed model in this thesis is used for studying the full duplex decode-and-forward relay system.The expression of the outage probability of the system is deduced and the outage capacity of the system is calculated.According to different system QoS requirements,this thesis further discusses the optimal scheme of non-linear energy harvesting based on outage performance,and gives the closed expression of the optimal solution to minimize the outage probability of the full-duplex system.And the optimal solution to maximize the outage capacity of the system is obtained based on the Golden Section method.The simulation results verify the correctness of the theoretical work,and the simulation results also show that the performance of the full-duplex system under the proposed model is quite different from that of the traditional linear model. |
PDF Full Text Request