Research On Key Techniques Of Secure And Energy-Efficient Beamforming For Wireless Communications

In the past two decades,wireless communication technologies have grown very rapidly.However,due to the broadcast nature of wireless transmission,there always exists the risk of security,that is,legitimate users in the wireless communication network always have the risk of being wiretapped by eavesdropping users,which seriously threatens the secure transmission of the wireless communication network.At the same time,with the rapid growth of mobile data traffic services in recent years,the energy consumption problems faced by wireless communication networks and the entire communication industry have become more serious.Beamforming(BF),as a directional transmission signal processing technology,is considered to be an effective solution to the problem of secure and energy-efficient transmission in wireless communication networks.Therefore,this thesis conducts research on the energy-efficient and secure transmission technology in the downlink wireless communication networks,with an objective to analyze and improve the security and energy efficiency(EE)of the wireless communication network by designing the base station to transmit BF vectors.The specific research contents and main contributions are as follows:1.Considering the scenario of multiple cognitive user(CU)and eavesdropping user(EU)pairs in the cognitive network,where the cognitive base station(CBS)is equipped with multiple antennas,and a multi-antenna EU is surrounding each single-antenna CU.The problem of maximizing the EE of all CUs in the cognitive radio network(CRN)subject to secrecy rate is studied,and a two-layer iterative algorithm is proposed to solve the BF vector of EE maximization problem.Since the original optimization problem is non-convex and cannot be solved directly.We first convert the objective function in the original fractional form to an equivalent subtraction form,and then apply the Difference of Two-Convex Functions(D.C.)method to convert the equivalent optimization problem into an approximate convex optimization problem.Finally,numerical results showing the effectiveness of the proposed energy efficiency maximization scheme are provided.2.Aiming at the scenario where the primary base station(PBS)interferes with the cognitive network and the cognitive network has multiple EUs,the problem of secrecy energy efficiency(SEE)maximization for CU in the CRN based on artificial noise is studied,and a two-layer algorithm is proposed to solve the corresponding optimal BF vector of the SEE maximization problem.First,based on the instantaneous channel state information(CSI),a BF scheme with maximum SEE is proposed.Since the original optimization problem is non-convex,an auxiliary variable greater or equal to one is introduced to convert the original optimization problem into an equivalent optimization problem with only the objective function being non-convex,which is then converted into another equivalent subtraction-form optimization problem.The subtraction-form problem is again converted to an approximate convex problem by using the D.C.method.In addition,considering a practical situation where instantaneous CSI may not be available in real time,another SEE maximization BF scheme is proposed based on statistical CSI.Finally,numerical simulation results demonstrating the advantages of the two proposed SEE maximization schemes in terms of SEE performance,as well as the feasibility of the SEE maximization scheme under statistical CSI are provided.3.The problem of maximizing SEE in NOMA-based CRNs is studied,and a two-layer iterative algorithm is given to calculate the corresponding optimal BF vector.It is considered that the two primary users(PUs)are a pair of NOMA users,out of which,the strong user is a legitimate receiver,and the weak user is a potential EU.In order to improve the secrecy performance of the legitimate users in primary network,the secondary user(SU)of the cognitive network is designed to work in full duplex(FD)mode,that is,when the SU receives the signals from the primary network and the cognitive network,it also sends artificial noise signals to interfere with the PUs.Since the objective function of the original optimization problem is non-convex,it is first converted into an equivalent optimization problem in the form of subtraction.To facilitate further processing,the optimization variable is changed from a BF vector to a matrix,with rank-one constraint.Then an approximate convex optimization problem is obtained through the first-order Taylor expansion.Finally,numerical simulations are employed to evaluated the SEE performance of the CRN using NOMA and FD technology.4.The secrecy performance analysis of the amplify and forward(AF)relay network using BF is studied,where a multi-antenna EU attempts to wiretap the data message sent from a multi-antenna sources to a multi-antenna destination through a single antenna relay,and the EU is close to the destination.It is assumed that all communication links obey Rayleigh fading,and due to the obstacle or distance,direct links between the source and destination and between the source and EU are not available.With the help of the second-type modified Bessel function and hypergeometric function,we first derive the approximate analytical expressions of the safety outage probability(SOP)and the average safety rate(ASR)of relay communication networks.Next,in order to further study the diversity gain and array gain of the relay communication network,by using the asymptotic relationship of the hypergeometric function,we derive the asymptotic expression of the secrecy outage probability and average safety rate under the high main-to-eavesdropper signal-noise ratio(MER).Finally,numerical simulation results are given to verify the correctness of the theoretical security analysis and disclose the influence of the number of antennas on the security performance of the considered relay network.