Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)

Cooperative Relay Based on Machine Learning for Enhancing Physical Layer Security

Jintao Xing, Tiejun Lv, Xuewei Zhang

Published Sep 1, 2019• 16 citations
Publisher

Abstract

Physical layer security is a crucial technology to achieve secure information transmission. However, emerging technologies, such as machine learning, impose significant challenges on existing security systems. In this paper, we propose a new relay system to achieve physical layer secure transmission and construct a deep learning model on the relay to design secure beamforming vector. This design is able to adapt to the current channel of the legitimate user and resist the eavesdropping from eavesdroppers. The model takes the channel’s statistical characteristic into account and it can learn the channel state information (CSI) to design secure beamforming in order to against eavesdropping effectively. The performance of bit error rate (BER) is close to the Shannon limit when the receiver demodulates the received signals, and we also ensure that the eavesdroppers cannot demodulate the transmitted signals. In addition, we design a power allocation algorithm, which considers the eavesdropper’s CSI on the relay. Simulation results show that the proposed model is able to reach the pre-defined BER requirements. The model can maximize the system security rate by optimizing power allocation at the relay.

DOI
10.1109/pimrc.2019.8904319

Related papers

Robust Cooperative Relay Beamforming Design for Security

Xiangwu Gong, Feihong Dong, Hongjun Li +1 more

In this paper, we investigate a security transmission scheme at the physical layer for cooperative wireless relay networks in the presence of a passive eavesdropper.While the security scheme has been previously investigated with perfect channel state information(CSI) in the presence of a passive eavesdropper, this paper focuses on researching the robust cooperative relay beamforming mechanism for wireless relay networks which makes use of artificial noise (AN) to confuse the eavesdropper and increase its uncertainty about the source message.The transmit power used for AN is maximized to degrade the signal-to-interference-plus-noise-ratio (SINR) level at the eavesdropper, while satisfying the individual power constraint of each relay node and worst-case SINR constraint at the desired receiver under a bounded spherical region for the norm of the CSI error vector from the relays to the destination.Cooperative beamforming weight vector in the security scheme can be obtained by using S-Procedure and rank relaxation techniques.The benefit of the proposed scheme is showed in simulation results.

KSII Transactions on Internet and Information Systems•Nov 30, 2015•0 citations

RAV: Relay Aided Vectorized Secure Transmission in Physical Layer Security for Internet of Things Under Active Attacks

Ning Zhang, Renyong Wu, Shenglan Yuan +2 more

Internet of Things (IoT) security becomes of great importance, as IoT is the foundation for many emerging services. To safeguard IoT security, cryptosystems at upper layer relying on sophisticated key management alone can face many challenges due to the massive deployment of resource constrained machine-type communication (MTC) devices. Physical layer (PHY) security can complement and enhance IoT security, by exploiting the characteristics of the bottom layer. In PHY security, channel state information (CSI) estimated through reverse pilot training is essential for the sender to select appropriate beamforming/precoder, which however is also vulnerable to adversaries. An adversary can actively launch pilot contamination attacks to affect the channel estimation and improve its signal reception quality. In this paper, we propose a relay-aided vectorized (RAV) secure transmission scheme, to safeguard the downlink communication in IoT networks under potential pilot contamination attacks. The proposed scheme does not distinguish the pilot sequences sent from an adversary and the receiver; and the sender utilizes what it receives to estimate the CSI for beamforming/precoder design. Then, a set of data symbols are presuperposed using a random complex matrix to form signal vectors to send. Through cooperation with a relay, the signal vectors can be recovered by the intended receiver whereas the adversary or the relay cannot, as proved through security analysis. The simulation results also demonstrate that the bit error rate (BER) of the adversary is 0.5 regardless of its channel quality, indicating perfect secrecy is achieved.

IEEE Internet of Things Journal•May 28, 2019•69 citations

Optimal Relay Selection for Wireless Relay Channel with External Eavesdropper: a NN-based Approach

Zhixiang Deng, Qian Sang, Yuan Gao +1 more

In this paper, we exploit the potential benefits of machine learning in enhancing physical layer security in cooperative wireless networks. We focus on the case where multiple relays adopt amplify-and-forward (AF) relaying to forward information from the source to the destination. It is assumed that the global channel state information (CSI) of the legitimate links and wiretap links is available to the source. The optimal relay is selected to improve physical layer security against eavesdropping. By modeling the problem of the relay selection as a multi-class classification problem, a neural network (NN) based scheme is proposed to select the optimal relay which guarantees the perfect secrecy performance of the relay cooperative communication system. Compared with the conventional relay selection scheme, the simulation results show that our proposed scheme not only achieves almost the same secrecy performance, but also has advantage of relatively small feedback overhead. The work presented here provides insights into the new design of relay selection based on machine learning.

IEEE/CIC International Conference on Communications in China (ICCC)•Aug 1, 2018•6 citations

Improving Wireless Physical Layer Security via Cooperating Relays

Lun Dong, Zhu Han, Athina P. Petropulu +1 more

Physical (PHY) layer security approaches for wireless communications can prevent eavesdropping without upper layer data encryption. However, they are hampered by wireless channel conditions: absent feedback, they are typically feasible only when the source-destination channel is better than the source-eavesdropper channel. Node cooperation is a means to overcome this challenge and improve the performance of secure wireless communications. This paper addresses secure communications of one source-destination pair with the help of multiple cooperating relays in the presence of one or more eavesdroppers. Three cooperative schemes are considered: decode-and-forward (DF), amplify-and-forward (AF), and cooperative jamming (CJ). For these schemes, the relays transmit a weighted version of a reencoded noise-free message signal (for DF), a received noisy source signal (for AF), or a common jamming signal (for CJ). Novel system designs are proposed, consisting of the determination of relay weights and the allocation of transmit power, that maximize the achievable secrecy rate subject to a transmit power constraint, or, minimize the transmit power subject to a secrecy rate constraint. For DF in the presence of one eavesdropper, closed-form optimal solutions are derived for the relay weights. For other problems, since the optimal relay weights are difficult to obtain, several criteria are considered leading to suboptimal but simple solutions, i.e., the complete nulling of the message signals at all eavesdroppers (for DF and AF), or the complete nulling of jamming signal at the destination (for CJ). Based on the designed relay weights, for DF in the presence of multiple eavesdroppers, and for CJ in the presence of one eavesdropper, the optimal power allocation is obtained in closed-form; in all other cases the optimal power allocation is obtained via iterative algorithms. Numerical evaluation of the obtained secrecy rate and transmit power results show that the proposed design can significantly improve the performance of secure wireless communications.

IEEE Transactions on Signal Processing•Dec 11, 2009•1,355 citations

Relay beamforming design for physical layer secure communication via line search algorithm

Siqi Chen, Cong Sun

This paper considers a physical layer security model in wireless communications. Two legitimate users communicate through several relays with the presence of an eavesdropper. We jointly design the relay beamforming weights and minimize the total relay transmit power, while ensuring users' Quality of Services and preventing the information being eavesdropped at the same time. The problem is a robust optimization problem, because of the imperfect channel state information from users and relays to the eavesdropper. First the original problem is simplified, where the high order robust terms are omitted. Then we design an iterative algorithm based on line search, by solving two Quadratically Constrained Quadratic Programming subproblems and a one-dimensional subproblem. Simulation results indicate that the proposed algorithm outperforms the state of the arts.

China Communications•Dec 1, 2021•1 citation

Artificial Noise-Aided Secure Relay Communication With Unknown Channel Knowledge of Eavesdropper

Bin Li, Meiying Zhang, Yue Rong +1 more

In this article, a new relay-aided secure communication system is investigated, where a transmitter sends signals to a destination via an amplify-and-forward (AF) relay in the presence of an eavesdropper. We consider a general system configuration, where the source, relay, destination, and eavesdropper are all equipped with multiple antennas. In the practical scenarios of unknown eavesdropper's channel state information (CSI) and uncertainty of the eavesdropper's location, we aim to maximize the expected value of the system secrecy rate over the presumed distribution of the eavesdropper's channels, by exploiting the artificial noise (AN) transmitted by the source and relay nodes. The system design issue is formulated as a nonconvex stochastic optimization problem with a source transmission power constraint and a nonconvex relay transmission power constraint. A novel computational method is proposed to solve this challenging problem. The new method is developed based on an exact penalty function method together with a parallel stochastic decomposition algorithm. Numerical simulations are performed to study the effectiveness of the proposed scheme at various locations of the eavesdropper. Simulation results show that for most cases, secure communication can be achieved without the CSI knowledge of eavesdropper's channels, and the achievable secrecy rate follows the trend of a benchmark system where the eavesdropper's full CSI is available. In particular, the achievable system secrecy rate increases with the number of antennas at the legitimate users. Moreover, the optimal power allocated for the transmission of the AN increases with the system signal-to-noise ratio. The proposed computational method achieves a higher system secrecy rate than a conventional penalty function based approach.

IEEE Transactions on Wireless Communications•Jan 10, 2021•39 citations
Physical layer security is a crucial technology to achieve secure information transmission. However, emerging technologies, such as machine learning, impose significant challenges on existing security systems. In this paper, we propose a new relay system to achieve physical layer secure transmission and construct a deep learning model on the relay to design secure beamforming vector. This design is able to adapt to the current channel of the legitimate user and resist the eavesdropping from eavesdroppers. The model takes the channel’s statistical characteristic into account and it can learn the channel state information (CSI) to design secure beamforming in order to against eavesdropping effectively. The performance of bit error rate (BER) is close to the Shannon limit when the receiver demodulates the received signals, and we also ensure that the eavesdroppers cannot demodulate the transmitted signals. In addition, we design a power allocation algorithm, which considers the eavesdropper’s CSI on the relay. Simulation results show that the proposed model is able to reach the pre-defined BER requirements. The model can maximize the system security rate by optimizing power allocation at the relay.
In this paper, we investigate a security transmission scheme at the physical layer for cooperative wireless relay networks in the presence of a passive eavesdropper.While the security scheme has been previously investigated with perfect channel state information(CSI) in the presence of a passive eavesdropper, this paper focuses on researching the robust cooperative relay beamforming mechanism for wireless relay networks which makes use of artificial noise (AN) to confuse the eavesdropper and increase its uncertainty about the source message.The transmit power used for AN is maximized to degrade the signal-to-interference-plus-noise-ratio (SINR) level at the eavesdropper, while satisfying the individual power constraint of each relay node and worst-case SINR constraint at the desired receiver under a bounded spherical region for the norm of the CSI error vector from the relays to the destination.Cooperative beamforming weight vector in the security scheme can be obtained by using S-Procedure and rank relaxation techniques.The benefit of the proposed scheme is showed in simulation results.
Internet of Things (IoT) security becomes of great importance, as IoT is the foundation for many emerging services. To safeguard IoT security, cryptosystems at upper layer relying on sophisticated key management alone can face many challenges due to the massive deployment of resource constrained machine-type communication (MTC) devices. Physical layer (PHY) security can complement and enhance IoT security, by exploiting the characteristics of the bottom layer. In PHY security, channel state information (CSI) estimated through reverse pilot training is essential for the sender to select appropriate beamforming/precoder, which however is also vulnerable to adversaries. An adversary can actively launch pilot contamination attacks to affect the channel estimation and improve its signal reception quality. In this paper, we propose a relay-aided vectorized (RAV) secure transmission scheme, to safeguard the downlink communication in IoT networks under potential pilot contamination attacks. The proposed scheme does not distinguish the pilot sequences sent from an adversary and the receiver; and the sender utilizes what it receives to estimate the CSI for beamforming/precoder design. Then, a set of data symbols are presuperposed using a random complex matrix to form signal vectors to send. Through cooperation with a relay, the signal vectors can be recovered by the intended receiver whereas the adversary or the relay cannot, as proved through security analysis. The simulation results also demonstrate that the bit error rate (BER) of the adversary is 0.5 regardless of its channel quality, indicating perfect secrecy is achieved.
In this paper, we exploit the potential benefits of machine learning in enhancing physical layer security in cooperative wireless networks. We focus on the case where multiple relays adopt amplify-and-forward (AF) relaying to forward information from the source to the destination. It is assumed that the global channel state information (CSI) of the legitimate links and wiretap links is available to the source. The optimal relay is selected to improve physical layer security against eavesdropping. By modeling the problem of the relay selection as a multi-class classification problem, a neural network (NN) based scheme is proposed to select the optimal relay which guarantees the perfect secrecy performance of the relay cooperative communication system. Compared with the conventional relay selection scheme, the simulation results show that our proposed scheme not only achieves almost the same secrecy performance, but also has advantage of relatively small feedback overhead. The work presented here provides insights into the new design of relay selection based on machine learning.
Physical (PHY) layer security approaches for wireless communications can prevent eavesdropping without upper layer data encryption. However, they are hampered by wireless channel conditions: absent feedback, they are typically feasible only when the source-destination channel is better than the source-eavesdropper channel. Node cooperation is a means to overcome this challenge and improve the performance of secure wireless communications. This paper addresses secure communications of one source-destination pair with the help of multiple cooperating relays in the presence of one or more eavesdroppers. Three cooperative schemes are considered: decode-and-forward (DF), amplify-and-forward (AF), and cooperative jamming (CJ). For these schemes, the relays transmit a weighted version of a reencoded noise-free message signal (for DF), a received noisy source signal (for AF), or a common jamming signal (for CJ). Novel system designs are proposed, consisting of the determination of relay weights and the allocation of transmit power, that maximize the achievable secrecy rate subject to a transmit power constraint, or, minimize the transmit power subject to a secrecy rate constraint. For DF in the presence of one eavesdropper, closed-form optimal solutions are derived for the relay weights. For other problems, since the optimal relay weights are difficult to obtain, several criteria are considered leading to suboptimal but simple solutions, i.e., the complete nulling of the message signals at all eavesdroppers (for DF and AF), or the complete nulling of jamming signal at the destination (for CJ). Based on the designed relay weights, for DF in the presence of multiple eavesdroppers, and for CJ in the presence of one eavesdropper, the optimal power allocation is obtained in closed-form; in all other cases the optimal power allocation is obtained via iterative algorithms. Numerical evaluation of the obtained secrecy rate and transmit power results show that the proposed design can significantly improve the performance of secure wireless communications.
This paper considers a physical layer security model in wireless communications. Two legitimate users communicate through several relays with the presence of an eavesdropper. We jointly design the relay beamforming weights and minimize the total relay transmit power, while ensuring users' Quality of Services and preventing the information being eavesdropped at the same time. The problem is a robust optimization problem, because of the imperfect channel state information from users and relays to the eavesdropper. First the original problem is simplified, where the high order robust terms are omitted. Then we design an iterative algorithm based on line search, by solving two Quadratically Constrained Quadratic Programming subproblems and a one-dimensional subproblem. Simulation results indicate that the proposed algorithm outperforms the state of the arts.
In this article, a new relay-aided secure communication system is investigated, where a transmitter sends signals to a destination via an amplify-and-forward (AF) relay in the presence of an eavesdropper. We consider a general system configuration, where the source, relay, destination, and eavesdropper are all equipped with multiple antennas. In the practical scenarios of unknown eavesdropper's channel state information (CSI) and uncertainty of the eavesdropper's location, we aim to maximize the expected value of the system secrecy rate over the presumed distribution of the eavesdropper's channels, by exploiting the artificial noise (AN) transmitted by the source and relay nodes. The system design issue is formulated as a nonconvex stochastic optimization problem with a source transmission power constraint and a nonconvex relay transmission power constraint. A novel computational method is proposed to solve this challenging problem. The new method is developed based on an exact penalty function method together with a parallel stochastic decomposition algorithm. Numerical simulations are performed to study the effectiveness of the proposed scheme at various locations of the eavesdropper. Simulation results show that for most cases, secure communication can be achieved without the CSI knowledge of eavesdropper's channels, and the achievable secrecy rate follows the trend of a benchmark system where the eavesdropper's full CSI is available. In particular, the achievable system secrecy rate increases with the number of antennas at the legitimate users. Moreover, the optimal power allocated for the transmission of the AN increases with the system signal-to-noise ratio. The proposed computational method achieves a higher system secrecy rate than a conventional penalty function based approach.