This paper considers the physical-layer secrecy design for full-duplex (FD) bidirectional communications in the presence of an eavesdropper (Eve). The goal of this work is to maximize the sum secrecy rate (SSR) of the bidirectional transmissions via appropriately designing the transmit covariance matrices at the legitimate nodes. To this end, we propose an alternating difference-of-concave (ADC) approach to iteratively optimizing the transmit covariance matrices. We show that each ADC iteration can be carried out efficiently with a semi-closed- form solution, and that every limit point of ADC iterations is a stationary solution of the SSR maximization problem. Besides the SSR maximization, this paper also deals with a robust SSR maximization problem to account for imperfect CSI of Eve. Assuming a moment-based random CSI error model (i.e., only mean and covariance of the error are known, but the exact distribution is not known), robust transmit designs based on Markov's inequality and the robust conic reformulation are developed. The efficacy of the proposed designs is demonstrated through simulations.
This paper considers the physical-layer secrecy design for full-duplex (FD) bidirectional communications in the presence of an eavesdropper (Eve). The goal of this work is to maximize the sum secrecy rate (SSR) of the bidirectional transmissions via appropriately designing the transmit covariance matrices at the legitimate nodes. To this end, we propose an alternating difference-of-concave (ADC) approach to iteratively optimizing the transmit covariance matrices. We show that each ADC iteration can be carried out efficiently with a semi-closed- form solution, and that every limit point of ADC iterations is a stationary solution of the SSR maximization problem. Besides the SSR maximization, this paper also deals with a robust SSR maximization problem to account for imperfect CSI of Eve. Assuming a moment-based random CSI error model (i.e., only mean and covariance of the error are known, but the exact distribution is not known), robust transmit designs based on Markov's inequality and the robust conic reformulation are developed. The efficacy of the proposed designs is demonstrated through simulations.
Consider a full-duplex (FD) bidirectional secure communication system, where two communication nodes, named Alice and Bob, simultaneously transmit and receive confidential information from each other, and an eavesdropper, named Eve, overhears the transmissions. Our goal is to maximize the sum secrecy rate (SSR) of the bidirectional transmissions by optimizing the transmit covariance matrices at Alice and Bob. To tackle this SSR maximization (SSRM) problem, we develop an alternating difference-of-concave (ADC) programming approach to alternately optimize the transmit covariance matrices at Alice and Bob. We show that the ADC iteration has a semiclosed form beamforming solution, and is guaranteed to converge to a stationary solution of the SSRM problem. Besides the SSRM design, this paper also deals with a robust SSRM transmit design under a moment-based random channel state information (CSI) model, where only some roughly estimated first- and second-order statistics of Eve's CSI are available, but the exact distribution or other high-order statistics is not known. This moment-based error model is different from the widely used bounded-sphere error model and the Gaussian random error model. Under the considered CSI error model, the robust SSRM is formulated as an outage probability-constrained SSRM problem. By leveraging the Lagrangian duality theory and difference-of-concave (DC) programming, a tractable safe solution to the robust SSRM problem is derived. The effectiveness and the robustness of the proposed designs are demonstrated through simulations.
This paper considers multiple-input multiple-output (MIMO) full-duplex (FD) two-way secrecy systems. Specifically, both multi-antenna FD legitimate nodes exchange their own confidential message in the presence of an eavesdropper. Taking into account the imperfect channel state information (CSI) of the eavesdropper, we formulate a robust sum secrecy rate maximization (RSSRM) problem subject to the outage probability constraint of the achievable sum secrecy rate and the transmit power constraint. Unlike other existing channel uncertainty models, e.g., norm- bounded and Gaussian-distribution, we exploit a moment-based random distributed CSI uncertainty model to recast our formulate RSSRM problem into convex optimization frameworks based on a Markov's inequality and robust conic reformulation, i.e., semidefinite programming (SDP). In addition, difference-of-concave (DC) approximation is employed to iteratively tackle the transmit covariance matrices of these legitimate nodes. Simulation results are provided to validate our proposed FD approaches.
This paper deals with the physical-layer (PHY) security problem for a full-duplex (FD) two-way relay system, where two FD sources exchange confidential information through an FD relay. Assuming imperfect channel state information (CSI) of the eavesdropper (Eve), we aim for a robust amplify-andforward (AF) relaying scheme that maximizes the sum secrecy rate of the two-way transmissions. Different from the widely used norm-bounded deterministic error model and the Gaussian random error model, we adopt a moment-based random error model to capture the imperfection of Eve's CSI. Specifically, the CSI error is assumed to be randomly distributed with given mean and covariance, but the exact distribution or other highorder statistics are not known. Under this moment-based error model, we formulate an outage-constrained sum secrecy rate maximization (OC-SRM) problem, which guarantees the outage sum secrecy rate is kept below certain threshold for arbitrary distributions fulfilling the mean and covariance. To tackle this OC-SRM problem, we first reexpress the outage probability into a more explicit form, and then develop a semidefinite relaxation (SDR)-based alternating difference-of-concave (DC) programming approach to computing a safe solution. Simulation results corroborate the robustness of the proposed design when varying the Eve's CSI error distribution.
Consider a full-duplex two-way relay network, where two legitimate nodes simultaneously transmit and receive confidential information through a full-duplex multiantenna relay, in the presence of an eavesdropper. To secure the communications, an artificial-noise (AN)-aided amplify-and-forward (AF) strategy is employed at the relay, with a goal of maximizing the sum secrecy rate of the two-way transmissions. This sum secrecy rate maximization (SSRM) problem is nonconvex by nature, but can be converted into the form of the difference-of-concave (DC) functions after the semidefinite relaxation (SDR). Thus, the classical DC programming naturally applies. We prove that the SDR is tight and give a specific way to recover a stationary solution of the SSRM problem from the relaxed DC problem. Moreover, to reduce the iteration complexity of DC, we proposed an inexact DC framework, which uses an approximate solution to iterate, rather than a globally optimal one. The convergence of the inexact DC to a stationary solution of the SSRM problem is also established.
We discuss the physical layer security scheme in the Full-Duplex (FD) MIMO point-to-point two-way communication system with residual self-interference, in which legitimate nodes send confidential information and null space Artificial Noise (AN) while receiving information. Because the Channel State Information (CSI) of the eavesdropper is unavailable, we optimize the covariance matrices of the information signal as well as the allocation of the antenna for transmitting and receiving to minimize the signal power consumption under the target rate constraint. As a result, the power of AN is maximized within the limit of total power, so the interception capability of the eavesdropper is suppressed as much as possible. Since self-interference cannot be completely eliminated, the optimization process of one legitimate node depends on the optimization result of the other. By substituting self-interference power in the secrecy rate formula with its average value, the joint optimization process at the two nodes is transformed into two separate and solvable optimization processes. Then, the Water-Filling Algorithm (WFA) and bisection algorithm are used to get the optimal covariance matrices of the signal. Furthermore, we derive the theoretical lower bound of ergodic achievable secrecy rate under rayleigh channels to evaluate the performance of the scheme. The simulation results show that the theoretical derivation is correct, and the actual achievable rate is very close to the target rate, which means that the approximation in the optimization is feasible. The results also show that secrecy transmission can be realized because a considerable secrecy rate can be achieved.
We propose a robust resource allocation framework to improve the physical layer security in the presence of an active eavesdropper. In the considered system, we assume that both legitimate receiver and eavesdropper are full-duplex (FD) while most works in the literature concentrate on passive eavesdroppers and half-duplex (HD) legitimate receivers. In this paper, the adversary intends to optimize its transmit and jamming signal parameters so as to minimize the secrecy data rate of the legitimate transmission. In the literature, assuming that the receiver operates in HD mode, secrecy data rate maximization problems subject to the power transmission constraint have been considered in which cooperating nodes act as jammers to confound the eavesdropper. This paper investigates an alternative solution in which we take advantage of FD capability of the receiver to send jamming signals against the eavesdroppers. The proposed self-protection scheme eliminates the need for external helpers. Moreover, we consider the channel state information uncertainty on the links between the active eavesdropper and other legitimate nodes of the network. Optimal power allocation is then obtained based on the worst-case secrecy data rate maximization, under a legitimate transmitter power constraint in the presence of the active eavesdropper. Numerical results confirm the advantage of the proposed secrecy design and in certain conditions, demonstrate substantial performance gain over the conventional approaches.