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 Transactions on Communications

Deep Reinforcement Learning-Enabled Secure Visible Light Communication Against Eavesdropping

Liang Xiao, Geyi Sheng, Sicong Liu, Huaiyu Dai, Mugen Peng, Jian Song

Published Jul 22, 2019• 109 citations
Publisher

Abstract

The inherent broadcast characteristics of the visible light communication (VLC) channel makes VLC downlinks susceptible to unauthorized terminals in many actual VLC scenarios, such as offices and shopping centers. This paper considers a multiple-input-single-output (MISO) VLC scenario with multiple light fixtures acting as the transmitter, a VLC receiver as the legitimate user, and an eavesdropper attempting to intercept the undisclosed information. To improve the confidentiality of VLC links, a physical-layer anti-eavesdropping framework is proposed to obscure the unauthorized eavesdroppers and diminishes their capability of inferring the information through smart beamforming over the MISO VLC wiretap channel. To cope with the intractable problem of finding the theoretically optimal solution of the secrecy rate and utility for the MISO VLC wiretapping channel, a reinforcement learning (RL)-based VLC beamforming control scheme is proposed to achieve the optimal beamforming policy against the eavesdropper. Furthermore, a deep RL-based VLC beamforming control scheme is proposed to handle the curse of dimensionality for both observation space and action space and avoid the quantization error of the RL-based algorithm. Simulation results show that the proposed learning-based VLC beamforming control schemes can significantly decrease the bit error rate of the legitimate receiver and increase the secrecy rate and utility of the anti-eavesdropping MISO VLC system, compared with the benchmark strategy.

DOI
10.1109/tcomm.2019.2930247

Related papers

Smart Jamming for Secrecy: Deep Reinforcement Learning Enabled Secure Visible Light Communication

Sicong Liu, Xianbin Liu, Xiaojiang Du +1 more

As one of the indoor communication technologies, visible light communication (VLC) has drawn great attention for its advantages such as ultra-wide unlicensed spectrum, power saving and low complexity. The nature of the visible light propagation is an open channel, which is vulnerable to wiretapping. This paper investigates a secure VLC mechanism enabled by multiple light fixtures acting as friendly jammers. The goal of the friendly jammers is to diminish the capability of the eavesdropper to infer the undisclosed information, on the premise of causing minimal impact on the legitimate receiver. For this reason, an algorithm based on reinforcement learning is proposed to dynamically optimize the friendly jamming policy in realistic nonstationary environments. In order to resolve the difficult problem of the dimensional curse and to effectively represent the continuous state and action spaces, an algorithm based on deep reinforcement learning is devised, which utilizes deep convolutional neural networks to accelerate the convergence rate of the learning process. A differentiable neural dictionary is introduced to make full use of the experiences in similar anti-eavesdropping scenarios to improve the learning capability. Simulation results demonstrate that, the proposed schemes can achieve a higher secrecy rate and a lower bit error rate than some state-of-the-art schemes.

IEEE Transactions on Wireless Communications•Sep 18, 2024•12 citations

Physical-Layer Security for MISO Visible Light Communication Channels

Ayman Mostafa, Lutz Lampe

This paper considers improving the confidentiality of visible light communication (VLC) links within the framework of physical-layer security. We study a VLC scenario with one transmitter, one legitimate receiver, and one eavesdropper. The transmitter has multiple light sources, while the legitimate and unauthorized receivers have a single photodetector, each. We characterize secrecy rates achievable via transmit beamforming over the multiple-input, single-output (MISO) VLC wiretap channel. For VLC systems, intensity modulation (IM) via light-emitting diodes (LEDs) is the most practical transmission scheme. Because of the limited dynamic range of typical LEDs, the modulating signal must satisfy certain amplitude constraints. Hence, we begin with deriving lower and upper bounds on the secrecy capacity of the scalar Gaussian wiretap channel subject to amplitude constraints. Then, we utilize beamforming to obtain a closed-form secrecy rate expression for the MISO wiretap channel. Finally, we propose a robust beamforming scheme to consider the scenario wherein information about the eavesdropper's channel is imperfect due to location uncertainty. A typical application of the proposed scheme is to secure the communication link when the eavesdropper is expected to exist within a specified area. The performance is measured in terms of the worst-case secrecy rate guaranteed under all admissible realizations of the eavesdropper's channel.

IEEE Journal on Selected Areas in Communications•May 13, 2015•254 citations

Jamming for Secrecy: Reinforcement Learning Based Anti-Eavesdropping Visible Light Communication

Xianbin Liu, Sicong Liu

Due to the broadcast nature of visible light communication (VLC), the secrecy protection is a crucial issue. In this paper, aiming at the communication scenario of a point-to-point VLC transmission link, an anti-eavesdropping model utilizing friendly jamming is constructed, which is composed of a light-emitting diode (LED) transmitter, a legitimate receiver equipped with a photodiode (PD), and a potential eavesdropper trying to obtain the undisclosed information transmitted. In addition, the system is equipped with a friendly jammer composed of multiple LEDs. A friendly jamming scheme based on reinforcement learning (RL) is proposed, which adopts different friendly jamming actions with the dynamic change of the environment. The simulation results show that compared with the benchmark state-of-the-art method, the proposed scheme can significantly reduce the bit error rate (BER) of the legitimate receiver and improve the overall performance of the anti-eavesdropping VLC system.

International Conference on Communications•May 16, 2022•3 citations

Physical-layer security for indoor visible light communications

Ayman Mostafa, Lutz Lampe

This paper considers secure transmission over the visible light communication (VLC) channel by the means of physical-layer security techniques. In particular, we consider achievable secrecy rates of the multiple-input, single-output (MISO) wiretap VLC channel. The VLC channel is modeled as a deterministic and real-valued Gaussian channel subject to amplitude constraints. We utilize null-steering and artificial noise strategies to achieve positive secrecy rates when the eavesdropper's channel state information (CSI) is perfectly known and entirely unknown to the transmitter, respectively. In both scenarios, the legitimate receiver's CSI is available to the transmitter. We numerically evaluate achievable secrecy rates under typical VLC scenarios and show that simple precoding techniques can significantly improve the confidentiality of VLC links.

International Conference on Communications•Jun 1, 2014•130 citations

Discrete Input Signaling for Secure MISO VLC Systems with Randomly Located Eavesdroppers

Mohamed Amine Arfaoui, Ali Ghrayeb, Chadi M. Assi +1 more

We study in this paper the secrecy performance of the multiple-input single-output (MISO) visible light communication (VLC) channel in the presence of randomly located eavesdroppers. We consider a system model comprising a transmitter, equipped with multiple fixtures of LEDs, one legitimate receiver equipped with a single photo-diode (PD) and a group of randomly located (and colluding) eavesdroppers, each equipped with a single PD. We derive a closed-form expression for the average achievable secrecy rate as a function of the beamforming vector, the input distribution, and the location density of the eavesdroppers using stochastic geometry. We then investigate the optimal beamformer that maximizes the average achievable secrecy rate. We corroborate the analytical results through Monte Carlo simulations and we demonstrate substantial improvements provided by the proposed scheme over existing ones.

IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)•Sep 1, 2018•3 citations

Enhancing the Secrecy Performance of Gaussian MISO VLC Wiretap Channels with Randomly Located Eavesdroppers

Mohamed Amine Arfaoui, Ali Ghrayeb, Chadi M. Assi

We study in this paper the achievable secrecy rate of the Gaussian multiple-input single-output (MISO) visible light communication (VLC) in the presence of randomly located eavesdroppers. We consider a system model comprising a transmitter (Alice) equipped with multiple fixtures of LEDs, one legitimate receiver (Bob) equipped with a single photo-diode (PD) and a group of randomly located eavesdroppers, each equipped with a single PD. The channel is modeled as deterministic, real- valued and subject to an amplitude constraint, through which Alice wants to communicate privately with Bob. We consider the case where the eavesdroppers are colluding together, i.e., they act jointly in eavesdropping on the communication between Alice and Bob. We adopt the truncated generalized normal (TGN) as input signaling and beamforming as transmission strategy. At first, we derive a closed-form expression of the average achievable secrecy rate as a function of the density of eavesdroppers using stochastic geometry. Then, we investigate the optimal beamformer that maximizes the average achievable secrecy rate of the system. Finally, the analysis is verified by Monte Carlo simulations.

International Conference on Communications•May 1, 2018•4 citations
The inherent broadcast characteristics of the visible light communication (VLC) channel makes VLC downlinks susceptible to unauthorized terminals in many actual VLC scenarios, such as offices and shopping centers. This paper considers a multiple-input-single-output (MISO) VLC scenario with multiple light fixtures acting as the transmitter, a VLC receiver as the legitimate user, and an eavesdropper attempting to intercept the undisclosed information. To improve the confidentiality of VLC links, a physical-layer anti-eavesdropping framework is proposed to obscure the unauthorized eavesdroppers and diminishes their capability of inferring the information through smart beamforming over the MISO VLC wiretap channel. To cope with the intractable problem of finding the theoretically optimal solution of the secrecy rate and utility for the MISO VLC wiretapping channel, a reinforcement learning (RL)-based VLC beamforming control scheme is proposed to achieve the optimal beamforming policy against the eavesdropper. Furthermore, a deep RL-based VLC beamforming control scheme is proposed to handle the curse of dimensionality for both observation space and action space and avoid the quantization error of the RL-based algorithm. Simulation results show that the proposed learning-based VLC beamforming control schemes can significantly decrease the bit error rate of the legitimate receiver and increase the secrecy rate and utility of the anti-eavesdropping MISO VLC system, compared with the benchmark strategy.
As one of the indoor communication technologies, visible light communication (VLC) has drawn great attention for its advantages such as ultra-wide unlicensed spectrum, power saving and low complexity. The nature of the visible light propagation is an open channel, which is vulnerable to wiretapping. This paper investigates a secure VLC mechanism enabled by multiple light fixtures acting as friendly jammers. The goal of the friendly jammers is to diminish the capability of the eavesdropper to infer the undisclosed information, on the premise of causing minimal impact on the legitimate receiver. For this reason, an algorithm based on reinforcement learning is proposed to dynamically optimize the friendly jamming policy in realistic nonstationary environments. In order to resolve the difficult problem of the dimensional curse and to effectively represent the continuous state and action spaces, an algorithm based on deep reinforcement learning is devised, which utilizes deep convolutional neural networks to accelerate the convergence rate of the learning process. A differentiable neural dictionary is introduced to make full use of the experiences in similar anti-eavesdropping scenarios to improve the learning capability. Simulation results demonstrate that, the proposed schemes can achieve a higher secrecy rate and a lower bit error rate than some state-of-the-art schemes.
This paper considers improving the confidentiality of visible light communication (VLC) links within the framework of physical-layer security. We study a VLC scenario with one transmitter, one legitimate receiver, and one eavesdropper. The transmitter has multiple light sources, while the legitimate and unauthorized receivers have a single photodetector, each. We characterize secrecy rates achievable via transmit beamforming over the multiple-input, single-output (MISO) VLC wiretap channel. For VLC systems, intensity modulation (IM) via light-emitting diodes (LEDs) is the most practical transmission scheme. Because of the limited dynamic range of typical LEDs, the modulating signal must satisfy certain amplitude constraints. Hence, we begin with deriving lower and upper bounds on the secrecy capacity of the scalar Gaussian wiretap channel subject to amplitude constraints. Then, we utilize beamforming to obtain a closed-form secrecy rate expression for the MISO wiretap channel. Finally, we propose a robust beamforming scheme to consider the scenario wherein information about the eavesdropper's channel is imperfect due to location uncertainty. A typical application of the proposed scheme is to secure the communication link when the eavesdropper is expected to exist within a specified area. The performance is measured in terms of the worst-case secrecy rate guaranteed under all admissible realizations of the eavesdropper's channel.
Due to the broadcast nature of visible light communication (VLC), the secrecy protection is a crucial issue. In this paper, aiming at the communication scenario of a point-to-point VLC transmission link, an anti-eavesdropping model utilizing friendly jamming is constructed, which is composed of a light-emitting diode (LED) transmitter, a legitimate receiver equipped with a photodiode (PD), and a potential eavesdropper trying to obtain the undisclosed information transmitted. In addition, the system is equipped with a friendly jammer composed of multiple LEDs. A friendly jamming scheme based on reinforcement learning (RL) is proposed, which adopts different friendly jamming actions with the dynamic change of the environment. The simulation results show that compared with the benchmark state-of-the-art method, the proposed scheme can significantly reduce the bit error rate (BER) of the legitimate receiver and improve the overall performance of the anti-eavesdropping VLC system.
This paper considers secure transmission over the visible light communication (VLC) channel by the means of physical-layer security techniques. In particular, we consider achievable secrecy rates of the multiple-input, single-output (MISO) wiretap VLC channel. The VLC channel is modeled as a deterministic and real-valued Gaussian channel subject to amplitude constraints. We utilize null-steering and artificial noise strategies to achieve positive secrecy rates when the eavesdropper's channel state information (CSI) is perfectly known and entirely unknown to the transmitter, respectively. In both scenarios, the legitimate receiver's CSI is available to the transmitter. We numerically evaluate achievable secrecy rates under typical VLC scenarios and show that simple precoding techniques can significantly improve the confidentiality of VLC links.
We study in this paper the secrecy performance of the multiple-input single-output (MISO) visible light communication (VLC) channel in the presence of randomly located eavesdroppers. We consider a system model comprising a transmitter, equipped with multiple fixtures of LEDs, one legitimate receiver equipped with a single photo-diode (PD) and a group of randomly located (and colluding) eavesdroppers, each equipped with a single PD. We derive a closed-form expression for the average achievable secrecy rate as a function of the beamforming vector, the input distribution, and the location density of the eavesdroppers using stochastic geometry. We then investigate the optimal beamformer that maximizes the average achievable secrecy rate. We corroborate the analytical results through Monte Carlo simulations and we demonstrate substantial improvements provided by the proposed scheme over existing ones.
We study in this paper the achievable secrecy rate of the Gaussian multiple-input single-output (MISO) visible light communication (VLC) in the presence of randomly located eavesdroppers. We consider a system model comprising a transmitter (Alice) equipped with multiple fixtures of LEDs, one legitimate receiver (Bob) equipped with a single photo-diode (PD) and a group of randomly located eavesdroppers, each equipped with a single PD. The channel is modeled as deterministic, real- valued and subject to an amplitude constraint, through which Alice wants to communicate privately with Bob. We consider the case where the eavesdroppers are colluding together, i.e., they act jointly in eavesdropping on the communication between Alice and Bob. We adopt the truncated generalized normal (TGN) as input signaling and beamforming as transmission strategy. At first, we derive a closed-form expression of the average achievable secrecy rate as a function of the density of eavesdroppers using stochastic geometry. Then, we investigate the optimal beamformer that maximizes the average achievable secrecy rate of the system. Finally, the analysis is verified by Monte Carlo simulations.