Study Of Physical Layer Security In Hybrid Power Line-Wireless Communication Network

With the ubiquitous access of massive terminals and the expansion of power grid big data services,only power line communication（PLC）can not meet the needs of the rapid growth communication for the future power Internet of Thing.Therefore,hybrid power line-wireless communication（HPWC）will become the development trend of the future power communication.However,HPWC will face more severe security problems due to the influence of mixed dual communication media.Meanwhile,facing the dual challenges of key management explosion and quantum computing cracking,the traditional key encryption technology may not ensure absolute security any more.Then,Physical Layer Security（PLS）technology is widely concerned.It utilizes the uniqueness and mutuality of the transmission medium to achieve information encryption and provide a secure physical channel for the data of upper layer.Complementing with the traditional key encryption technology,PLS technology can greatly improve the communication security performance.A lot of researches of PLS have been done in wireless communication（WLC）network while there are few related studies in HPWC network so far.Motivated by the above,considering multiple-input multiple-output（MIMO）-HPWC wiretap channel,the PLC coding schemes are studied in this paper as follows:（1）For the communication security of HPWC,the MIMO-HPWC wiretap channel model is investigated.Firstly,for the PLC channel,weighing the complexity against accuracy of channel model,MK（Zimmermann M and Dostert K）model is employed to characterize the Single-Input Single-Output（SISO）-PLC channel.Accordingly,a MIMO-PLC channel model with elements of non-independently identical distribution is established based on the correlation coefficient matrix,and the strong impulsive noise is considered.Next,for the WLC channel,based on large-scale and small-scale fading characteristics,the MIMO-WLC channel with elements of independently identical distribution is constructed.Then,considering the concealment of eavesdropper,a deterministic uncertainty model is utilized to describe the channel statement information（CSI）of the eavesdropper.Besides,non-collusion and collusion wiretapping channel models are constructed for the different cooperation modes of eavesdroppers.Finally,adding the above wiretapping channel model into a mixed PLC and WLC network,a MIMO-HPWC wiretapping channel model is established.（2）For the single-user MIMO-HPWC cooperative relay network,a robust strategy of artificial noise aided secrecy precoding and decode-and-forward relay is conceived.Specially,it is divided into the following steps: 1)Describe the secrecy outage probability（SOP）based on the CSI estimation error.2)Accordingly,the optimization problem of security outage capacity maximization is built by jointly optimizing the AN-aided secrecy precoding of both transmitter and relay under the SOP constraint.3)Resorting to the semi defined program,the quadratic constrained quadratic programming based on SOCM can be transformed into a primary programming.4)Employing Bernstein Type Inequality and Charnes-Cooper transformation,the SOP constraints is decoupled and linearized.5)The above block convex problem can be iteratively solved by utilizing the block coordinated descend（BCD）algorithm,and its convergence and complexity is analyzed.Simulation results show that our proposed SOCM algorithm can obtain the better security performance than zero forced and singular value decomposition algorithm.（3）For the multi-user MIMO-HPWC interference network,a secrecy precoding strategy based on distributed interference management is proposed.More specially,it can be divided into the following steps: 1)resorting to the asynchronous distributed pricing mechanism,the utility maximum（UM）optimization problem based on distributed interference management is constructed.Furthermore,it’s mathematically proved that the UM optimization is the equal to the concentrated sum secrecy rate maximization optimization from the perspective of the Karush-Kuhn-Tucker（KKT）solution.2)The objective function of UM optimization problem is convex approximated by weighted minimized mean square error.3)Resorting to the Schur complement and sign-definiteness lemma,he constraints of channel estimation error are decoupled and linearized.4)The optimal artificial noise and secrecy precoding can be iteratively solved by BCD algorithm,and its convergence and complexity is discussed.Simulation results illustrates that in the case of perfect CSI,our proposed UM algorithm outperforms the interference alignment algorithm and the minimized total mean square error algorithm in the low and medium transmit power domain.In the case of imperfect CSI,our proposed UM algorithm always outperforms the artificial noise aided interference alignment algorithm.In conclusion,for the universal HPWC multi-hop（relay）network and interference network,this paper conducts the security theoretical research and simulation verification of cooperative relay and distributed interference management respectively,and solves the underlying security coding problem for the actual HPWC network.The proposed strategy has certain theoretical and practical significance for the security research of ubiquitous Internet of Things network such as the power Internet of Things.