Physical layer security, which is an alternative to traditional cryptographic methods, has emerged as a promising candidate to protect wireless transmissions from an eavesdropper. An important measure of physical layer security is the secrecy outage: the event when the instantaneous secrecy capacity, which for the considered scenario is the difference between the capacity of the Alice (the transmitter) to Bob (the legitimate receiver) channel and the capacity of the Alice to Eve (the eavesdropper) channel, is below a target secrecy rate for which the system has been designed. In this paper, we argue that design under such an outage definition on a wireless channel is inadequate in some scenarios, as it treats very different error events with equal weight. In response, we propose that two conditions are met: the instantaneous capacity between the source and destination is above a target rate and the instantaneous capacity between the source and eavesdropper is held below a target rate. The two individual target rates form a secrecy rate pair. This also naturally splits the outage event into two regions with quite different costs: the eavesdropper's ability to decode the message (which is a security breach and must be avoided), and the destination's inability to decode the message (which simply requires a retransmission). We use our formulation to consider rate pair selection in hybrid ARQ systems and provide numerical results supporting this approach.
Physical layer security, which is an alternative to traditional cryptographic methods, has emerged as a promising candidate to protect wireless transmissions from an eavesdropper. An important measure of physical layer security is the secrecy outage: the event when the instantaneous secrecy capacity, which for the considered scenario is the difference between the capacity of the Alice (the transmitter) to Bob (the legitimate receiver) channel and the capacity of the Alice to Eve (the eavesdropper) channel, is below a target secrecy rate for which the system has been designed. In this paper, we argue that design under such an outage definition on a wireless channel is inadequate in some scenarios, as it treats very different error events with equal weight. In response, we propose that two conditions are met: the instantaneous capacity between the source and destination is above a target rate and the instantaneous capacity between the source and eavesdropper is held below a target rate. The two individual target rates form a secrecy rate pair. This also naturally splits the outage event into two regions with quite different costs: the eavesdropper's ability to decode the message (which is a security breach and must be avoided), and the destination's inability to decode the message (which simply requires a retransmission). We use our formulation to consider rate pair selection in hybrid ARQ systems and provide numerical results supporting this approach.
Wireless systems are susceptible to eavesdropping as an unwelcome result of the broadcast nature of the wireless medium. Fortunately, the wireless medium also provides endowments including the temporal dispersion, the capability of signal superposition and other properties, to facilitate defending against eavesdropping. The focus of the thesis is to provide schemes for utilizing these properties at the physical layer and their performance analysis from an information-theoretic secrecy perspective. The work is based on Wyner's results on the wire-tap channel, in which the transmission between two legitimate users (Alice and Bob) is eavesdropped upon by Eve. The secrecy level is measured by the equivocation rate at Eve. The system performance is measured by the secrecy capacity. This is the maximum rate at which a message can be sent from Alice to Bob, while constraining the equivocation rate at Eve equal to the communication rate, and thus ensuring that Eve cannot deduce any useful information asymptotically. First, the scenario where a secret message needs to be delivered to the intended receiver within a given delay constraint and without the knowledge of the channel state information (CSI) at the transmitter is considered. In particular, hybrid automatic retransmission request (HARQ) protocols are revisited for a block-fading wire-tap channel via a joint consideration of channel coding, secrecy coding, and retransmission protocols. Two secure HARQ protocols: a repetition time diversity (RTD) scheme with maximal-ratio combining, and an incremental redundancy (INR) scheme based on ratecompatible Wyner secrecy codes are analyzed via an outage-based formulation. The throughput of RTD and INR protocols under probabilistic outage requirements is derived. Asymptotic analysis and numerical computations are given to demonstrate both the benefits of HARQ protocols and the tradeoff between reliability and confidentiality in delay-constrained communications. Second, the effects of interference on information-theoretic secret communication is considered. More specifically, the wiretap channel with a helping interferer (WT-HI) is proposed and studied. Here, an interferer, which does not know the confidential message, helps in ensuring the secrecy of the message by sending independent signals. Achievable secrecy rates and upper bounds on the secrecy capacity are given for both discrete memoryless and Gaussian channels. Interference has conventionally been regarded as a harmful phenomenon that leads to decreasing the rate and the reliability performance of wireless systems. In this work, it is shown that interference can be exploited to assist wireless secrecy. Lastly, the problem of secret key generation over fading channels, in which there exists an additional authenticated, public and error-free feedback channel from Bob to Alice, is considered. A key generation scheme consisting of a communication phase and a key-generation phase is presented. The communication phase is based on layered broadcast coding, which adapts the decoded rate at Bob to the actual channel state without requiring CSI available at Alice. The key-generation phase is based on the messages decoded by Bob after the communication phase and the public feedback. The achievable secrecy key rate and the optimal power distribution over coded layers for maximizing the secrecy key rate are characterized. Theoretical and numerical results show that the broadcast approach outperforms the single-level-coding based approach significantly, and hence demonstrate the role of introducing self-interference in facilitating secret-key generation over slowly fading channels when transmit CSI is not available.
In this paper, we consider physical layer security in wireless communication networks in which a source (Alice) intends to send a confidential message to a legitimate destination (Bob) with the help of a cooperative jammer (CJ), in the presence of a passive eavesdropper (Eve). Assuming that only statistical channel state information (CSI) of Eve is available, artificial-noise (AN) assisted beamforming and cooperative jamming are designed. The goal is to maximize the secrecy rate, subject to a constraint on secrecy outage probability. Numerical results validate the effectiveness of our scheme. Moreover, a higher secure energy efficiency (EE) can be achieved as compared with other schemes without a cooperative jammer.
In applications of wireless packet-oriented data networks, a special coding scheme, the hybrid automatic retransmission request (HARQ) exhibits high throughput efficiency by adapting its error correcting code redundancy to channel conditions. Motivated by the increasing importance of secure communication over wireless networks, we investigate secure packet communication based on HARQ over block-fading (BF) channels. More specifically, we consider two legitimate users communicating over a BF channel in the presence of a passive eavesdropper who intercepts the transmission through another independent BF channel. We assume that the transmitter can obtain a 1-bit ACK/NACK feedback from the receiver via a reliable public channel. Under this setting, we consider incremental redundancy (IR) and repetition time diversity (RTD) HARQ schemes based on rate-compatible Wyner secrecy codes from an information theoretic point of view. We study a good Wyner code sequence, with which the legitimate receiver can decode the message and the eavesdropper can be perfectly confused. For a given pair of reliability/secrecy outage probabilities, we derive an achievable secrecy throughput of HARQ protocols for block-fading channels. Finally, we illustrate numerically that HARQ can benefit both throughput and secrecy.
In this paper, we evaluate how much beamforming and noise injection can improve physical layer security by deriving the probability of non-zero secrecy capacity and the secrecy outage probability. Our model is a typical wiretap channel model, composed of three participants - the transmitter Alice, the legitimate receiver Bob and the eavesdropper Eve. A typical cellular scenario is considered where Alice has multiple antennas while Bob and Eve each have only one antenna (MISOSE). Beamforming and noise injection at Alice are applied to achieve a higher instantaneous signal-to-noise ratio (SNR) at Bob and a lower SNR at Eve. The probability density functions (pdf) of Bob and Eve's SNR are figured out in both noise-injection condition and injection-free condition, and we derive the closed-form expressions for the probability of non-zero secrecy capacity and the secrecy outage probability. Numerical analysis is also addressed using the two probabilities introduced above.
Relaying is a promising technique to improve wireless physical-layer security. Existing literature shows that a full-duplex relay can further improve the secrecy capacity and secrecy outage probability compared to conventional half-duplex relay, but this comes at a price of sophisticated implementation. For sake of easy implementation, two-path successive relaying has been proposed to emulate the full-duplex relay by scheduling a pair of half-duplex relays to assist the source transmission alternately. However, the performance of two-path successive relaying in secrecy communication remains unexplored. This paper proposes a secrecy two-path successive relaying protocol for a scenario with one source, one destination and two half-duplex relays. The relays operate alternately in a time division mode to forward messages continuously from source to destination in the presence of an eavesdropper. To further confuse the eavesdropper, the source and relay are scheduled to transmit jamming signals at appropriate intervals. Analytical results on the secrecy outage probability reveals that the proposed protocol is able to deliver the target secrecy rate when the SNR of the eavesdropping channels are high. In addition, the secrecy outage probability of the proposed protocol is the joint secrecy outage probability of the relay pair. Numerical simulations show that the proposed protocol achieves the highest ergodic secrecy capacity and lowest secrecy outage probability compared to the existing half duplex relaying, full duplex relaying and full duplex jamming schemes.
Physical layer security has recently been regarded as an emerging technique to complement and improve the communication security in future wireless networks. The current research and development in physical layer security are often based on the ideal assumption of perfect channel knowledge or the capability of variable-rate transmissions. In this paper, we study the secure transmission design in more practical scenarios by considering channel estimation errors at the receiver and investigating both fixed-rate and variable-rate transmissions. Assuming quasi-static fading channels, we design secure on-off transmission schemes to maximize the throughput subject to a constraint on secrecy outage probability. For systems with given and fixed encoding rates, we show how the optimal on-off transmission thresholds and the achievable throughput vary with the amount of knowledge on the eavesdropper's channel. In particular, our design covers the interesting case where the eavesdropper also uses the pilots sent from the transmitter to obtain imperfect channel estimation. An interesting observation is that using too much pilot power can harm the throughput of secure transmission if both the legitimate receiver and the eavesdropper have channel estimation errors, while the secure transmission always benefits from increasing pilot power when only the legitimate receiver has channel estimation errors but not the eavesdropper. When the encoding rates are controllable parameters to design, we further derive both a non-adaptive and an adaptive rate transmission schemes by jointly optimizing the encoding rates and the on-off transmission thresholds to maximize the throughput of secure transmissions.