IEEE COMMUNICATIONS LETTERS, ACCEPTED FOR PUBLICATION1Secure Communication via Sending Artificial Noise by the Receiver:Outage Secrecy Capacity/Region AnalysisWei Li, Mounir Ghogho, Senior Member, IEEE, Bin Chen, Chunlin XiongAbstract—A novel approach for ensuring confidential wirelesscommunication is proposed and analyzed from an information-theoretic standpoint. In this approach, the legitimate receivergenerates artificial noise (AN) to impair the intruder’s channel.This method is robust because it does not need the feedback ofchannel state information (CSI) to the transmitter and does notassume that the number of Eve’s antennas should be smallerthan that of Bob. Furthermore, we propose a new concept ofoutage secrecy region to evaluate the secrecy performance froma geometrical perspective. This should be useful if we need toknow what zone should be protected (or militarized). Analysisand simulation results in practical environments show that theproposed method has a good performance.Index Terms—Artificial noise, privacy, outage secrecy region,secrecy capacity, physical layer security.I. INTRODUCTIONWpassive eavesdropper in an unknown location within “earshot”of a wireless transmission taps information about the trans-mitted signal without risk of detection. A natural frameworkfor information security at the physical layer is the so-calledwiretap channel introduced by Wyner [1] and associatednotion of secrecy capacity. Secrecy problems involve threenodes: the transmitter (Alice), the legitimate receiver (Bob)and the eavesdropper (Eve). Alice wants to communicate withBob while leaving Eve unable to decode the secret message.It is shown that perfect secrecy can be achieved without anykey, provided that Bob has a better channel than Eve. Thesecrecy capacity is defined as the maximum achievable ratefrom Alice to Bob while keeping Eve completely ignorant ofthe transmitted message. Later, Wyner’s work was extended tonondegraded discrete memoryless broadcast channels in [2],and then to the Gaussian channel in [3], recently to MIMOchannels in [4] and to fading channels in [5].In order to increase the secrecy capacity, artificial noise(AN) based method was suggested in [6]. In this method,AN is generated through multiple transmit antennas or thecooperating nodes, and is injected into the null-subspace ofBob’s MIMO channel [7], [8]. AN is utilized to impair Eve’sIRELESS communication is inherently insecure owingto the broadcast nature of the wireless medium. AManuscript received June 19, 2012. The associate editor coordinating thereview of this letter and approving it for publication was M. Tao.This work was supported in part by the NSFC under Grants 61101096 and61101098, and the NSF of Hunan Province under Grant 11jj4055.W. Li, B. Chen, and C. Xiong are with the School of Electronic Scienceand Engineering, National University of Defense Technology, Changsha,410073 P. R. China (e-mail: liwei.nudt.cn@gmail.com; lierbency@126.com;xchlzju@nudt.edu.cn.).M. Ghogho is with the University of Leeds, Leeds LS2 9JT, U.K., andalso with the International University of Rabat, 11100, Morocco (e-mail:m.ghogho@leeds.ac.uk).Digital Object Identifier 10.1109/LCOMM.2012.12.121344channel, while not affecting Bob’s channel. The work in [9],[10] jointly optimizes the beamforming vector and the ANcovariance matrix to achieve diverse signal to interference plusnoise ratio (SINR) constraints for Bob and Eve. An outageprobability-based approach to design the beamformers wasproposed in [11]. However, these schemes have to face thefollowing challenges: a) The channel state information (CSI)or at least partial CSI of Bob is needed at the transmitter;feeding back the CSI to the transmitter occupies some channelresource; b) If there is an uncertainty on the CSI at thetransmitter, the AN may leak to Bob and thus reduces hisSNR; this problem is even worse when Eve tries to personateBob and feeds back her own CSI to Alice; considering theimperfect CSI, a robust Bayesian approach for multiuserMIMO wiretap channels was presented in [12]; c) if there arecolluding Eves, or Eve has multiple antennas and the numberof antennas exceeds the number of Alice’s antennas, the ANcan be calculated and eliminated if the CSI is perfectly known.In this paper, we propose a novel AN based method toovercome the above problems. Different from the existingworks where AN is added at the transmitter, the AN in ourmethod is generated by the intended receiver, Bob, as shownin Fig. 1. The AN impairs the intruder’s channel while itcan be counteracted by Bob. This method has the followingadvantages: a) The CSI is not needed by Alice, so there is nofeedback channel and thus the bandwidth resource is saved;uncertainty on the CSI is not an issue which implies robustnessof the method; however, in order to use a wire-tap code, thechannel capacity of Bob should be known at Alice. b) theAN can be generated by either multiple antennas or a singleantenna, which is more practical than the existing AN methodswhich need multiple antennas at the transmitter; c) this methoddoes not assume that the number of Eve’s antennas shouldbe smaller than that of Bob; indeed, even if there is a largenumber of antennas at Eve or there are colluding Eves, the ANis still hard to be totally eliminated because the CSI betweenBob and Eve is not known to Eve; d) the proposed methodcan be combined with the masked beamforming scheme wherean AN is generated at the transmitter to improve secrecyperformance; e) it is particularly useful when the receiver hasa stronger ability than the transmitter (e.g. the receiver is abase station); f) it is efficient if Eves are located around Bob;this is generally the case in several situations.Another contribution of this paper is the introduction of theconcepts of outage secrecy region (OSR) using a geometricalperspective. In practical communication systems, when theeavesdroppers are passive, it is impossible to calculate thesecrecy capacity or Eve’s bit error rate. In [8] a probabilisticframework using stochastic geometry is presented to quantify1089-7798/12$31.00 © 2012 IEEE
