The Research On Interference Management Technology In Simultaneous Wireless Information And Power Transfer System

Multiple-Input Multiple-Output(MIMO)technology can provide multi-user diversity gain and significantly improve the spectral efficiency without increasing bandwidth and total antenna transmission power.However,due to the increase of the number of connected devices,the interference in multi-user communication systems has become one of the harmful factors affecting the system performance.In order to reduce or avoid interference,technologies such as multiple access and interference alignment(IA)are adopted for interference suppression.However,as a wireless signal,interference carries not only information,but also energy.When users have energy demand,interference can be used as an energy source.Simultaneous Wireless Information and Power Transfer(SWIPT)is a technology that transmits information and energy simultaneously through radio frequency signals.If the interference is used for wireless energy harvesting,green communication can be realized,so it is necessary to effectively manage and use the interference in the system.This thesis combines multi-user MIMO technology with SWIPT technology to make more in-depth research on multi-user interference management in the next generation wireless communication network.The main research contents are as follows:Firstly,IA technology is used to effectively manage interference in SWIPT system.Most of the previous research on interference management using IA technology in SWIPT systems carry out IA first,and then optimize SWIPT information or energy performance.This method needs IA processing for all users,so the complexity is high,and it is not necessary to align the energy harvesting users.At the same time,because the energy harvesting users fail to obtain energy from the interference signal,it causes energy waste.Aiming at the problems of high computational complexity,no alignment of energy harvesting users and energy waste in IA processing for all users.This thesis proposes two design schemes of SWIPT and IA based on user selection mode from the transmitting side and the receiving side respectively.On one hand,these schemes determine the user pairs of each time slot as energy harvesting(EH)user pairs or information decoding(ID)user pairs according to the real-time requirements of the system,on the other hand,they only implement IA for ID user pairs,and EH user pairs harvest the energy sent from ID user pairs.In the process of optimization design,the first design scheme takes the harvested power of EH user pairs and the signal to interference plus noise ratio(SINR)of ID user pairs as the maximization objectives,and performs IA processing on ID user pairs.This multi-objective optimization problem is transformed into a scalar optimization problem by scalar method.Since the problem is non convex,we transform it into two sub-problems by iterative optimization method,which can be solved by using the standard orthogonal basis of zero space and Rayleigh quotient theory.The second design scheme takes the transmission power of ID user pairs as the optimization goal.The constraints include the minimum harvested power demand of EH user pairs,the minimum SINR demand of ID user pairs,and IA of ID user pairs.The semi-positive definite relaxation(SDR)method is used in the solution process.The results show that the proposed scheme only implements IA for ID user pairs,which can effectively reduce the computational complexity,while EH user pairs harvest energy from interference,which reduces the energy waste.Secondly,under the linear EH model,the multi-objective beamforming technology is used to effectively manage the interference in SWIPT system.At present,most of the beamforming schemes of SWIPT system are single objective optimization schemes.These schemes take one of the common objectives of system optimization as the only objective and the other objectives as constraints.Because single objective optimization can not achieve the design expectation of the next generation wireless communication network and can not solve the problem of compatibility of multiple conflicting targets.Therefore,in this thesis,two multi-objective beamforming design schemes based on power splitting structure are proposed under different scenarios considering whether IA constraints are added.The two proposed schemes jointly design the power splitting ratio,beamforming vector and decoding vector on the receiving side to maximize the total SINR of ID signal and the total harvested power by EH signal.The scalarization method is utilized to solve the multi-objective problem.In the design process,the first design scheme does not consider the constraints of IA,and is solved by maximizing the generalized Rayleigh quotient and the vectorization theory of Hermitian matrix.The second design scheme optimizes the performance of the system under IA constraints,and uses the standard orthogonal basis of zero space to equivalent the optimization variables.The results show that in the process of optimizing the two conflicting objectives,the performance of the two conflicting objectives is improved due to the appropriate selection of weighting coefficients.At the same time,the achievable sum rate and harvested sum power of the proposed algorithm in the first scheme are better than the existing scheme of aligning interference first.When the second scheme has high signal-to-noise ratio,its sum rate performance is better than the first scheme due to IA.Finally,under the non-linear EH model,beamforming technology is used to effectively manage the interference in SWIPT system.At present,most of the energy harvesting circuits adopt the linear EH model,in which the harvested power by the terminal equipment increases in proportion to the input RF power.Because the linear model is inconsistent with the non-linear characteristics of actual devices,and a small number of existing studies on the non-linear EH model rarely consider the problem of decoding optimization on the receiving side,a beamforming design scheme considering the influence of decoding vector on the receiving side under the non-linear EH model of power splitting structure is proposed in this thesis.The scheme takes the maximization of the harvested power of EH terminal as the optimization goal and the minimum required SINR of ID terminal as the constraint.The optimization problem is non-convex since the objective function has the form of multi-variable fractional function sum.Hence,the double-layer nested iteration method is utilized to deal with it.In the outer iteration,the original problem is decomposed into three sub-problems.The solution of two optimization variables can be processed by the conventional convex optimization method.When solving the transmission precoding vector,the transformed objective function is still in the form of non-convex fractional function sum.In this thesis,the inner iteration method is used to further convert it into two sub-problems.SDR theory,non-linear proportional optimization theory and modified Newton algorithm are used to solve the problem,and the system performance is compared with that under linear EH model.The results show that the proposed scheme has good convergence.With different transmission power and minimum required SINR,the harvested power by the non-linear EH model is significantly higher than that of the linear EH model.At the same time,the information decoding performance of the system is effectively improved due to the optimization of decoding vector on the receiving side.