Research On Physical Layer Security Technology Of Cognitive Network

Cognitive Radio(CR)is one of the key technologies of next-generation wireless communications for its cognitive and configurable capabilities.The technology enables wireless communication devices to dynamically adjust operating parameters based on the wireless environment and utilize scarce spectrum resources in an efficient and flexible manner.However,with the rapid development of cognitive radio technology and the continuous improvement of node communication capabilities,the security of communication data in cognitive wireless networks is facing unprecedented challenges.In addition to traditional wireless security threats,due to cognitive wireless networks introduce new cognitive and refactoring capabilities,new security risks such as cognitive information leakage,‘spectrum traps' and imitating primary user eavesdropping should be taken into consideration.As a key security transmission technology that is different from the traditional encryption technology in application layer,the physical layer security transmission technology uses the difference of channels between the legitimate transceiver and the eavesdropper to design the corresponding technology or strategy to obtain the security of the system.Artificial Noise(AN)technology is widely used in the physical layer security transmission of cognitive wireless networks because it does not need to predict the a priori information of the eavesdropping channel.However,the performance obtained by the physical layer secure transmission strategy largely depends on the primary channel and the Channel State Information(CSI)acquired by the sender.At present,the research on reliable transmission of information in cognitive wireless networks mostly assumes that the channel state information is perfect.There are few researches on the design of secure transmission strategies under imperfect CSI.In this paper,under the scenario of multi-antenna transmission in cognitive wireless networks,two physical layer security transmission strategies,namely,null space artificial noise and generalized artificial noise,are optimized under perfect and imperfect CSI,aiming at analyzing the relationship between acquisition of the channel state information and security performance,the main work includes:1.Optimal security rate design for cognitive MIMOME systemsConsidering the secure transmission in cognitive networks with Underlay,to maximize the security rate of secondary user in a Multiple-Input Multiple-Output Multiple-antenna Eavesdropper(MIMOME)system,AN-aided secure transmission strategy is adopted under the constraint of interference temperature of Primary User(PU)and the transmission power limitation of Secondary User(SU).Due to the high complexity of the MIMO channel,the security rate maximization problem is non-convex and challenging.This paper proposes a two-layer iterative algorithm,the inner layer problem can be efficiently handled by solving a sequence of semi-definite problems,and the outer layer problem can be converted to a single-variable optimization problem,which can be tackled by one-dimensional search to effectively solve.The final simulation results demonstrate the effectiveness of the artificial noise algorithm in improving the secure transmission performance.2.Optimal artificial noise-aided secure transmission strategy design with imperfect channel information conditionsThrough the research on the artificial noise safety transmission technology,we found that the artificial noise can be zero forced and non-zero forced according to the transmission direction of the artificial noise signal.From the perspective of maximizing the rate of secrecy,the artificial noise nulling scheme is not optimal because it sacrifices parts of spatial degrees of freedom to reduce the interference effects on legitimate receivers.Therefore,based on the cognitive MIMOME eavesdropping channel model,under the constraint of the maximum transmission power of the secondary-user transmitter and the primary user interference temperature,we try to maximize the system security rate.Two artificial noise transmission algorithms are designed.Firstly,the artificial noise is limited to the null space of the legal secondary-user channel,and the Semi-definite programming(SDP)is used to design the null space artificial noise safe transmission algorithm.Then,a generalized artificial noise safe transmission algorithm is designed by using the Successive Convex Approximation(SCA)method.Finally,by comparing the computational complexity and performance of the two algorithms,the results show that the generalized artificial noise safety transmission algorithm has obvious performance compared with the zero-based artificial noise algorithm,but the zero-based artificial noise transmission algorithm has lower computational complexity.3.Design on robust artificial noise secure transmission strategy under imperfect channel state informationUnder the imperfect channel state information of primary users and secondary users at the transmitter,we try to achieve the Worst Case Secrecy Rate Maximization(WCSRM).A robust artificial noise security transmission algorithm is designed.Firstly,the uncertain channel is modeled.Secondly,because of the error boundary of the channel,the joint optimization of the artificial noise covariance matrix and the transmitted signal covariance matrix is an NP-hard problem.By using S-Procedure lemma and references,we transform the constraints into linear matrix inequality(LMIs).In this way,the original non-convex problem is transformed into a semi-definite programming problem.Finally,the simulation verifies the robustness of the design algorithm.