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

Communications Security Conference (CSC)

An approach to resist blind source separation attacks of speech signals

Guyue Li, Aiqun Hu

Published Jan 1, 2014• 1 citation
Publisher

Abstract

Recently, there has been great interest in the physical layer security technique which exploits the artificial noise (AN) to enlarge the channel condition between the legitimate receiver and the eavesdropper. However, in certain communication scenery, this strategy may suffer from some attacks in the signal processing perspective. In this paper, we consider speech signals and the scenario in which the eavesdropper has the similar channel performance compared to the legitimate receiver. We design the optimal artificial noise (AN) to resist the attack of the eavesdropper who uses the blind source separation (BSS) technology to reconstruct the secret information. The Optimal AN is obtained by making a tradeoff between results of direct eavesdropping and reconstruction. The simulation results show that the AN we proposed has better performance than that of the white Gaussian AN to resist the BSS attacks effectively.

DOI
10.1049/cp.2014.0738

Related papers

Robust Artificial Noise-aided Secure Communication against ICA-based Attacks

Yun Wen, Makoto Yoshida

Physical layer security (PLS) technologies have attracted much attention in recent years for their potential to provide information-theoretically secure communications. Artificial Noise (AN)-aided transmission, which can realize secure communications against the external passive eavesdropping, is considered one of the most practicable PLS technologies. In this paper, we reveal that the conventional AN scheme has a severe vulnerability when the eavesdropper adopts Independent Component Analysis (ICA)-based attack, by which the AN-aided transmission will be cracked in a high probability even with most transmission power allocated to AN signals. In order to ensure secure communications, we propose a robust AN scheme focused on the ICA's ability of blind source separation. In our proposed scheme, AN signals are generated with the same modulation scheme and a certain correlation to information signals, thus to introduce irremovable correlation to degrade the ICA's separation performance. Furthermore, we derive the optimal correlation factors from the relation between an information-AN correlation and the ICA's performance. Numerical simulation results with the comparison to the conventional AN scheme show that, our proposed scheme can ensure secure communications by degrading eavesdropper's success probability of attack from 96% to lower than 1%.

International Conference on Communications•Jun 1, 2020•0 citations

Research on BSS in Artificial Noise aided secure communication

Xinxin Li, Cai Xiaoxia, Wei Dong

The broadcast nature of wireless channels leads to the developments and applications of physical layer security techniques. The Artificial Noise (AN) aided technique is proposed to interfere with the eavesdropper for secure communications. A suppression method called Joint Approximate Diagonalization of Eigen-matrices (JADE), which is based on blind source separation algorithm (BSS), is presented under this circumstance. The eavesdropper is able to separate the communication signal and the noise. Simulation results demonstrate that the proposed method is feasible under certain circumstances, which helps the eavesdropper to defeat the AN aided secure communications effectively.

IEEE International Conference on Communication Software and Networks•Jun 1, 2016•0 citations

Artificial-Noise-Resistant Eavesdropping in MISO Wiretap Channels: Receiver Construction and Performance Analysis

Dongyang Xu, Pinyi Ren, James A. Ritcey

We consider secure communications over MISO wiretap channels, in the presence of a passive eavesdropper with multiple antennas. In this scenario, an artificial-noise-resistant (ANR) eavesdropping behavior is introduced to invalidate the artificial noise (AN) scheme proposed by Goel et al. This novel eavesdropper, by exploiting the statistical filter derived from the collected signals, can completely eliminate the influence of AN by using only two receive antennas. In particular, the received signals are firstly syphered to eliminate the power influence of AN by a linear weight which is generated from the statistical estimation for the signal covariance matrix. In comparison with the original legitimate signals, the weighted signals are however imposed by a phase difference which can be then erased by a weight vector inferred from the available information at the eavesdropper. Based on the two filtering processes, we derive a novel expression of achievable secrecy rate and give an analytical expression for the zero-secrecy-rate distance of eavesdropper to the transmitter. Finally, we characterize the expression of maximum achievable secrecy rate (MASR) and show that the optimal power allocation strategy under an ANR eavesdropping behavior is transformed into no allocation of transmission power to AN. Numerical results are presented to illustrate the damage caused by the investigated eavesdropping behavior. Interestingly, the consequence of AN elimination is not influenced even under large number of transmit antennas.

IEEE Vehicular Technology Conference•Sep 1, 2017•3 citations

Secure Over-the-Air Computation Using Zero-Forced Artificial Noise

Luis Maßny, Antonia Wachter-Zeh

Over-the-air computation has the potential to in-crease the communication-efficiency of data-dependent distributed wireless systems, but is vulnerable to eavesdropping. We consider over-the-air computation over block-fading additive white Gaussian noise channels in the presence of a passive eaves-dropper. The goal is to design a secure over-the-air computation scheme. We propose a scheme that achieves MSE-security against the eavesdropper by employing zero-forced artificial noise, while keeping the distortion at the legitimate receiver small. In contrast to former approaches, the security does not depend on external helper nodes to jam the eavesdropper’s received signal. We thoroughly design the system parameters of the scheme, propose an artificial noise design that harnesses unused transmit power for security, and give an explicit construction rule. Our design approach is applicable in both cases, if the eavesdropper’s channel coefficients are known and if they are unknown in the signal design. Simulations demonstrate the performance, and show that our noise design outperforms other methods.

arXiv:2212.04288•IEEE Information Theory Workshop (ITW)•Apr 23, 2023•8 citations

Effective eavesdropping in the Artificial noise aided security scheme

Feilong Wu, Wenjie Wang, Bobin Yao +1 more

The Artificial noise (AN) injection scheme is recently proposed to secure wireless transmissions. By simultaneously transmitting data symbols and artificial interference, it prevents the eavesdropper from decoding the secret message intended for the legitimate receiver. In this paper, we mainly focus on weakness of the AN security scheme. First, we analyze the necessary condition for such a scheme to guarantee that the legitimate transmitter-receiver pair can achieve a positive secrecy communication rate. And then, we further propose an effective eavesdropping strategy to counteract such a security scheme once the condition does not hold up in some wireless systems. Our proposed wiretap method involves complex-valued independent component analysis (ICA) and requires the antenna number of the eavesdropper's is larger or equal to that of the transmitter's. Simulation results demonstrate the effectiveness of this wiretap method.

IEEE/CIC International Conference on Communications in China (ICCC)•Aug 1, 2013•8 citations

Whispering Lies to the Eavesdropper: Physical Layer Spoofing Against Eavesdropping Attacks

Ozan Alp Topal, Güneş Karabulut Kurt

Recent work in physical layer security focuses on the message transmission between legitimate nodes without revealing the transmitted data to the eavesdropper. However, this gives room for the eavesdropper to adapt its resources to combat with the legitimate nodes. Instead, in this article,we introduce a friendly spoofing aiming at misleading the eavesdropper with a falsifying message signal to mask the original message signal. The effect of such an attack on the legitimate receiver is analyzed. A joint power optimization algorithm is proposed to maximize the spoofing performance while maintaining the communication requirements of legitimate nodes. The numerical results indicate that the proposed system can work under favorable channel propagation cases, and we can provide 4 b/symbols with 99.9% coverage probability with both legitimate and misleading messages.

IEEE Systems Journal•Jul 13, 2022•1 citation
Recently, there has been great interest in the physical layer security technique which exploits the artificial noise (AN) to enlarge the channel condition between the legitimate receiver and the eavesdropper. However, in certain communication scenery, this strategy may suffer from some attacks in the signal processing perspective. In this paper, we consider speech signals and the scenario in which the eavesdropper has the similar channel performance compared to the legitimate receiver. We design the optimal artificial noise (AN) to resist the attack of the eavesdropper who uses the blind source separation (BSS) technology to reconstruct the secret information. The Optimal AN is obtained by making a tradeoff between results of direct eavesdropping and reconstruction. The simulation results show that the AN we proposed has better performance than that of the white Gaussian AN to resist the BSS attacks effectively.
Physical layer security (PLS) technologies have attracted much attention in recent years for their potential to provide information-theoretically secure communications. Artificial Noise (AN)-aided transmission, which can realize secure communications against the external passive eavesdropping, is considered one of the most practicable PLS technologies. In this paper, we reveal that the conventional AN scheme has a severe vulnerability when the eavesdropper adopts Independent Component Analysis (ICA)-based attack, by which the AN-aided transmission will be cracked in a high probability even with most transmission power allocated to AN signals. In order to ensure secure communications, we propose a robust AN scheme focused on the ICA's ability of blind source separation. In our proposed scheme, AN signals are generated with the same modulation scheme and a certain correlation to information signals, thus to introduce irremovable correlation to degrade the ICA's separation performance. Furthermore, we derive the optimal correlation factors from the relation between an information-AN correlation and the ICA's performance. Numerical simulation results with the comparison to the conventional AN scheme show that, our proposed scheme can ensure secure communications by degrading eavesdropper's success probability of attack from 96% to lower than 1%.
The broadcast nature of wireless channels leads to the developments and applications of physical layer security techniques. The Artificial Noise (AN) aided technique is proposed to interfere with the eavesdropper for secure communications. A suppression method called Joint Approximate Diagonalization of Eigen-matrices (JADE), which is based on blind source separation algorithm (BSS), is presented under this circumstance. The eavesdropper is able to separate the communication signal and the noise. Simulation results demonstrate that the proposed method is feasible under certain circumstances, which helps the eavesdropper to defeat the AN aided secure communications effectively.
We consider secure communications over MISO wiretap channels, in the presence of a passive eavesdropper with multiple antennas. In this scenario, an artificial-noise-resistant (ANR) eavesdropping behavior is introduced to invalidate the artificial noise (AN) scheme proposed by Goel et al. This novel eavesdropper, by exploiting the statistical filter derived from the collected signals, can completely eliminate the influence of AN by using only two receive antennas. In particular, the received signals are firstly syphered to eliminate the power influence of AN by a linear weight which is generated from the statistical estimation for the signal covariance matrix. In comparison with the original legitimate signals, the weighted signals are however imposed by a phase difference which can be then erased by a weight vector inferred from the available information at the eavesdropper. Based on the two filtering processes, we derive a novel expression of achievable secrecy rate and give an analytical expression for the zero-secrecy-rate distance of eavesdropper to the transmitter. Finally, we characterize the expression of maximum achievable secrecy rate (MASR) and show that the optimal power allocation strategy under an ANR eavesdropping behavior is transformed into no allocation of transmission power to AN. Numerical results are presented to illustrate the damage caused by the investigated eavesdropping behavior. Interestingly, the consequence of AN elimination is not influenced even under large number of transmit antennas.
Over-the-air computation has the potential to in-crease the communication-efficiency of data-dependent distributed wireless systems, but is vulnerable to eavesdropping. We consider over-the-air computation over block-fading additive white Gaussian noise channels in the presence of a passive eaves-dropper. The goal is to design a secure over-the-air computation scheme. We propose a scheme that achieves MSE-security against the eavesdropper by employing zero-forced artificial noise, while keeping the distortion at the legitimate receiver small. In contrast to former approaches, the security does not depend on external helper nodes to jam the eavesdropper’s received signal. We thoroughly design the system parameters of the scheme, propose an artificial noise design that harnesses unused transmit power for security, and give an explicit construction rule. Our design approach is applicable in both cases, if the eavesdropper’s channel coefficients are known and if they are unknown in the signal design. Simulations demonstrate the performance, and show that our noise design outperforms other methods.
The Artificial noise (AN) injection scheme is recently proposed to secure wireless transmissions. By simultaneously transmitting data symbols and artificial interference, it prevents the eavesdropper from decoding the secret message intended for the legitimate receiver. In this paper, we mainly focus on weakness of the AN security scheme. First, we analyze the necessary condition for such a scheme to guarantee that the legitimate transmitter-receiver pair can achieve a positive secrecy communication rate. And then, we further propose an effective eavesdropping strategy to counteract such a security scheme once the condition does not hold up in some wireless systems. Our proposed wiretap method involves complex-valued independent component analysis (ICA) and requires the antenna number of the eavesdropper's is larger or equal to that of the transmitter's. Simulation results demonstrate the effectiveness of this wiretap method.
Recent work in physical layer security focuses on the message transmission between legitimate nodes without revealing the transmitted data to the eavesdropper. However, this gives room for the eavesdropper to adapt its resources to combat with the legitimate nodes. Instead, in this article,we introduce a friendly spoofing aiming at misleading the eavesdropper with a falsifying message signal to mask the original message signal. The effect of such an attack on the legitimate receiver is analyzed. A joint power optimization algorithm is proposed to maximize the spoofing performance while maintaining the communication requirements of legitimate nodes. The numerical results indicate that the proposed system can work under favorable channel propagation cases, and we can provide 4 b/symbols with 99.9% coverage probability with both legitimate and misleading messages.