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

Frontiers of Information Technology & Electronic Engineering

Reconfigurable intelligent surface-aided secret key generation using an autoencoder and K-means quantization

Zeyu Li, Panpan Xu, Qiangqiang Gao, Chunguo Li, Weijie Tan

Published Aug 1, 2025• 3 citations
Publisher

Abstract

In quasi-static wireless channel scenarios, the generation of physical layer keys faces the challenge of invariant spatial and temporal channel characteristics, resulting in a high key disagreement rate (KDR) and low key generation rate (KGR). To address these issues, we propose a novel reconfigurable intelligent surface (RIS)-aided secret key generation approach using an autoencoder and K -means quantization algorithm. The proposed method uses channel state information (CSI) for channel estimation and dynamically adjusts the reflection coefficients of the RIS to create a rapidly fluctuating channel. This strategy enables the extraction of dynamic channel parameters, thereby enhancing channel randomness. Additionally, by integrating the autoencoder with the K -means clustering quantization algorithm, the method efficiently extracts random bits from complex, ambiguous, and high-dimensional channel parameters, significantly reducing KDR. Simulations demonstrate that, under various signal-to-noise ratios (SNRs), the proposed method performs excellently in terms of KGR and KDR. Furthermore, the randomness of the generated keys is validated through the National Institute of Standards and Technology (NIST) test suite.

DOI
10.1631/fitee.2400799

Related papers

Machine Learning-Assisted Wireless PHY Key Generation with Reconfigurable Intelligent Surfaces

Long Jiao, Guohua Sun, Junqing Le +1 more

The key generation rate (KGR) performance of wireless physical layer (PHY) key generation can be limited by the quasi-static slow fading environment. In this work, we aim to exploit the radio environment reconfiguration ability enabled by reconfigurable intelligent surface (RIS) to improve KGR of PHY key generation. By rapidly changing the RIS configurations, the randomness or entropy rate of the wireless channel can be significantly increased, thus improving the KGR. To achieve high KGR while keeping low bit disagreement ratio (BDR), for the first time, we propose a machine learning (ML) based adaptive quantization level prediction scheme to decide an optimal quantization level based on channel state information (CSI). Simulation results show that with a prediction accuracy as high as 98.2%, the proposed ML-based prediction model tends to assign high quantization levels in the high SNR regime to reduce BDR, while adopting low quantization levels under low SNRs to maintain a low BDR.

ACM Workshop on Wireless Security and Machine Learning (WiseML)•Jun 21, 2021•29 citations

RIS-Assisted Integrated Communication and Secret Key Generation in Quasi-Static Environments

Zheyuan Deng, Xiaoyan Hu, Keming Ma +3 more

Wireless Channel-based Secret Key Generation (WC-SKG) offers a promising alternative for wireless communication security, yet suffers from an extremely low key generation rate (KGR) in quasi-static environments (e.g., indoor Internet of Things), where key refresh cycles can extend for hours, creating a security vulnerability. Reconfigurable Intelligent Surfaces (RIS) can boost KGR by introducing artificial randomness, yet existing schemes only randomize the RIS phase shifts without using environmental priors, failing to maximize the KGR and meet the communication signal-to-noise ratio (SNR) requirement, causing service outages. To overcome these limitations, we propose the RIS-assisted Integrated Communication and WC-SKG (RICK) scheme. First, we treat the RIS-adjusted channel as a designable random variable and derive its optimal probability density function (PDF) using a novel geometric-algebraic framework to maximize the KGR under the communication SNR constraint. Second, we design a constrained-clustering-based quantization region division scheme tailored to this optimal non-uniform PDF, guaranteeing uniformly distributed secret keys. Simulation results show RICK achieves a KGR approximately 3.5-5 times higher than the state-of-the-art scheme while saving at least 20 dB of transmit power for the same KGR, confirming its effectiveness in quasi-static scenarios.

IEEE Transactions on Information Forensics and Security•Jan 1, 2026•0 citations

Reconfigurable-Intelligent-Surface-Assisted Secret Key Generation Under Spatially Correlated Channels in Quasi-Static Environments

Vahid Shahiri, Hamid Behroozi, Ali Kuhestani +1 more

Physical-layer key generation (PLKG) can significantly enhance the security of classic encryption schemes by efficiently providing secret keys in resource-limited network like the Internet of Things (IoT). However, reaching a high key generation rate (KGR) is challenging in applications like smart home or remote area sensing with quasi-static channels. Recently, exploiting reconfigurable intelligent surface (RIS) to induce randomness in quasi-static wireless channels has received significant research interest. However, the inherent spatial correlation among the RIS elements is rarely studied, which can alter the optimum physical-layer key generation (PLKG) approach in terms of KGR and randomness in the key sequence. Specifically, for the first time, in this contribution, we take into account a spatially correlated RIS, which intends to enhance the KGR in a quasi-static medium. Novel closed-form analytical expressions for KGR are derived for the two cases of random phase shift (RPS) and our proposed equal phase shift (EPS) in the RIS elements. We also analyze the correlation between the channel samples to ensure the randomness of the generated secret key sequence. It is shown that the EPS scheme can effectively exploit the inherent spatial correlation between the RIS elements and it leads to a higher KGR compared to the widely used RPS strategy. We further formulate an optimization problem in which we determine the optimal portion of time dedicated to direct and indirect channel estimation, which has never been addressed in previous studies. We show the accuracy and the fast convergence of our sequential convex programming (SCP)-based algorithm and discuss the various parameters affecting spatially correlated RIS-assisted PLKG.

IEEE Internet of Things Journal•Jan 2, 2024•24 citations

Enhancing Secret Key Generation in Block Fading Channels using Reconfigurable Intelligent Surfaces

Hibatallah Alwazani, Anas Chaaban

Physical-layer security (PLS) is superior to classical cryptography techniques due to its notion of perfect secrecy and independence to an eavesdropper's computational power. One form of PLS arises when Alice and Bob (the legitimate users) exchange signals to extract a common key from the random common channels. The drawback of extracting keys from wireless channels is the ample dependence on the dynamicity and fluctuations of the radio channel. However, some radio channels are constant such as line-of-sight (LoS) and can be estimated by Eve (an illegitimate user), or can be quite static in behavior due to the presence low-mobility users thus restricting the amount of randomness. This in turn lowers the secret key rate (SKR) defined as the number of bits of key generated per channel use. In this work, we aim to address this challenge by using a reconfigurable intelligent surface (RIS) to produce random phases at certain carefully curated intervals such that it disrupts the channel in low-entropy environments. We propose an RIS assisted key generation method, study its performance, and compare with benchmarks to observe the benefit of using an RIS while considering various important metrics such as key mismatch rate and average secret key throughput. Simulations are made to validate our theoretical findings showing an improvement in performance when an RIS is deployed.

arXiv:2308.04726•IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)•Sep 5, 2023•0 citations

Adaptive physical-layer key generation via RIS phase-shift matrix design in quasi-static environment

Qiangqiang Gao, Weijie Tan, Xin Cao +3 more

Abstract In quasi-static environment, the performance of physical-layer key generation (PKG) is limited by low entropy. Deploying reconfigurable intelligent surface (RIS) is a mainstream method to solve this problem. However, the secret key performance (SKP) is generally insufficient at low signal-to-noise ratio (SNR). In this paper, we propose an RIS-assisted adaptive PKG scheme. Specifically, we design phase based uniform quantization to enhance the randomness of the RIS channel while simplifying computations. Then, we present an RIS phase shift matrix design based on rejection sampling to adapt the quantization algorithm, which can reduce bit disagreement rate (BDR) and improve key generation rate (KGR). Furthermore, to balance the system’s requirements for low BDR and high KGR, we derive a theoretical model for estimating the BDR probability. By computing the theoretical BDR, we determine the maximum quantization order for different SNRs. Combined with a discard strategy, this enables adaptive selection of the quantization order. Simulation results demonstrate that the proposed scheme achieves lower BDR and higher KGR than existing approaches. Moreover, it does not require complex algorithms, making it suitable for resource-constrained internet of things scenarios. (RIS_Assist_PKG: https://github.com/GQQ1206/RIS_Assist_PKG )

Journal of King Saud University - Computer and Information Sciences•Aug 20, 2026•0 citations

Physical Layer Group Key Generation With the Aid of Reconfigurable Intelligent Surfaces

Vahid Shahiri, Guyue Li, Hamid Behroozi

Reconfigurable intelligent surfaces (RIS) have the ability to alter the wireless environment by making changes in the impinging signal. While RIS has been extensively studied for enhancing wireless communications, its potential for facilitating group key generation (GKG) remains unexplored. In this study, we exploit the RIS to make the aggregate reflecting channels of different user terminals (UTs) as similar as possible to be able to extract common group secret keys from their channels. Specifically, the RIS will adjust its parameters to pave the way for GKG based on the physical channels of the UTs. Our method exploits the already gathered channel state information (CSI) in the RIS to beneficially design the phase shifts and does not impose additional probing burden on the network. We consider both passive RIS (PRIS) and active RIS (ARIS) to generate the group keys. The PRIS is widely adopted in physical layer key generation (PLKG) studies due to its use of passive elements, whereas the ARIS demonstrates superior capability in aligning the aggregate reflected channels among nodes in the GKG scenario, as demonstrated in this study. We will exploit various optimization methods like successive convex approximation (SCA) and semidefinite relaxation with Gaussian randomization (SDR-GR) to address the raised optimization problems. Unlike most of the studies in the literature, our scheme can achieve a high GKG rate in static environments as well. Finally, we will examine the performance of the proposed method by normalized mean squared error (NMSE), key error rate (KER), key generation rate (KGR) and key randomness metrics. Our numerical results verify that for the equal available power budget, the ARIS significantly outperforms PRIS in NMSE and KER, achieving more than four times higher KGR.

arXiv:2507.00714•Jul 1, 2025•0 citations
In quasi-static wireless channel scenarios, the generation of physical layer keys faces the challenge of invariant spatial and temporal channel characteristics, resulting in a high key disagreement rate (KDR) and low key generation rate (KGR). To address these issues, we propose a novel reconfigurable intelligent surface (RIS)-aided secret key generation approach using an autoencoder and K -means quantization algorithm. The proposed method uses channel state information (CSI) for channel estimation and dynamically adjusts the reflection coefficients of the RIS to create a rapidly fluctuating channel. This strategy enables the extraction of dynamic channel parameters, thereby enhancing channel randomness. Additionally, by integrating the autoencoder with the K -means clustering quantization algorithm, the method efficiently extracts random bits from complex, ambiguous, and high-dimensional channel parameters, significantly reducing KDR. Simulations demonstrate that, under various signal-to-noise ratios (SNRs), the proposed method performs excellently in terms of KGR and KDR. Furthermore, the randomness of the generated keys is validated through the National Institute of Standards and Technology (NIST) test suite.
The key generation rate (KGR) performance of wireless physical layer (PHY) key generation can be limited by the quasi-static slow fading environment. In this work, we aim to exploit the radio environment reconfiguration ability enabled by reconfigurable intelligent surface (RIS) to improve KGR of PHY key generation. By rapidly changing the RIS configurations, the randomness or entropy rate of the wireless channel can be significantly increased, thus improving the KGR. To achieve high KGR while keeping low bit disagreement ratio (BDR), for the first time, we propose a machine learning (ML) based adaptive quantization level prediction scheme to decide an optimal quantization level based on channel state information (CSI). Simulation results show that with a prediction accuracy as high as 98.2%, the proposed ML-based prediction model tends to assign high quantization levels in the high SNR regime to reduce BDR, while adopting low quantization levels under low SNRs to maintain a low BDR.
Wireless Channel-based Secret Key Generation (WC-SKG) offers a promising alternative for wireless communication security, yet suffers from an extremely low key generation rate (KGR) in quasi-static environments (e.g., indoor Internet of Things), where key refresh cycles can extend for hours, creating a security vulnerability. Reconfigurable Intelligent Surfaces (RIS) can boost KGR by introducing artificial randomness, yet existing schemes only randomize the RIS phase shifts without using environmental priors, failing to maximize the KGR and meet the communication signal-to-noise ratio (SNR) requirement, causing service outages. To overcome these limitations, we propose the RIS-assisted Integrated Communication and WC-SKG (RICK) scheme. First, we treat the RIS-adjusted channel as a designable random variable and derive its optimal probability density function (PDF) using a novel geometric-algebraic framework to maximize the KGR under the communication SNR constraint. Second, we design a constrained-clustering-based quantization region division scheme tailored to this optimal non-uniform PDF, guaranteeing uniformly distributed secret keys. Simulation results show RICK achieves a KGR approximately 3.5-5 times higher than the state-of-the-art scheme while saving at least 20 dB of transmit power for the same KGR, confirming its effectiveness in quasi-static scenarios.
Physical-layer key generation (PLKG) can significantly enhance the security of classic encryption schemes by efficiently providing secret keys in resource-limited network like the Internet of Things (IoT). However, reaching a high key generation rate (KGR) is challenging in applications like smart home or remote area sensing with quasi-static channels. Recently, exploiting reconfigurable intelligent surface (RIS) to induce randomness in quasi-static wireless channels has received significant research interest. However, the inherent spatial correlation among the RIS elements is rarely studied, which can alter the optimum physical-layer key generation (PLKG) approach in terms of KGR and randomness in the key sequence. Specifically, for the first time, in this contribution, we take into account a spatially correlated RIS, which intends to enhance the KGR in a quasi-static medium. Novel closed-form analytical expressions for KGR are derived for the two cases of random phase shift (RPS) and our proposed equal phase shift (EPS) in the RIS elements. We also analyze the correlation between the channel samples to ensure the randomness of the generated secret key sequence. It is shown that the EPS scheme can effectively exploit the inherent spatial correlation between the RIS elements and it leads to a higher KGR compared to the widely used RPS strategy. We further formulate an optimization problem in which we determine the optimal portion of time dedicated to direct and indirect channel estimation, which has never been addressed in previous studies. We show the accuracy and the fast convergence of our sequential convex programming (SCP)-based algorithm and discuss the various parameters affecting spatially correlated RIS-assisted PLKG.
Physical-layer security (PLS) is superior to classical cryptography techniques due to its notion of perfect secrecy and independence to an eavesdropper's computational power. One form of PLS arises when Alice and Bob (the legitimate users) exchange signals to extract a common key from the random common channels. The drawback of extracting keys from wireless channels is the ample dependence on the dynamicity and fluctuations of the radio channel. However, some radio channels are constant such as line-of-sight (LoS) and can be estimated by Eve (an illegitimate user), or can be quite static in behavior due to the presence low-mobility users thus restricting the amount of randomness. This in turn lowers the secret key rate (SKR) defined as the number of bits of key generated per channel use. In this work, we aim to address this challenge by using a reconfigurable intelligent surface (RIS) to produce random phases at certain carefully curated intervals such that it disrupts the channel in low-entropy environments. We propose an RIS assisted key generation method, study its performance, and compare with benchmarks to observe the benefit of using an RIS while considering various important metrics such as key mismatch rate and average secret key throughput. Simulations are made to validate our theoretical findings showing an improvement in performance when an RIS is deployed.
Abstract In quasi-static environment, the performance of physical-layer key generation (PKG) is limited by low entropy. Deploying reconfigurable intelligent surface (RIS) is a mainstream method to solve this problem. However, the secret key performance (SKP) is generally insufficient at low signal-to-noise ratio (SNR). In this paper, we propose an RIS-assisted adaptive PKG scheme. Specifically, we design phase based uniform quantization to enhance the randomness of the RIS channel while simplifying computations. Then, we present an RIS phase shift matrix design based on rejection sampling to adapt the quantization algorithm, which can reduce bit disagreement rate (BDR) and improve key generation rate (KGR). Furthermore, to balance the system’s requirements for low BDR and high KGR, we derive a theoretical model for estimating the BDR probability. By computing the theoretical BDR, we determine the maximum quantization order for different SNRs. Combined with a discard strategy, this enables adaptive selection of the quantization order. Simulation results demonstrate that the proposed scheme achieves lower BDR and higher KGR than existing approaches. Moreover, it does not require complex algorithms, making it suitable for resource-constrained internet of things scenarios. (RIS_Assist_PKG: https://github.com/GQQ1206/RIS_Assist_PKG )
Reconfigurable intelligent surfaces (RIS) have the ability to alter the wireless environment by making changes in the impinging signal. While RIS has been extensively studied for enhancing wireless communications, its potential for facilitating group key generation (GKG) remains unexplored. In this study, we exploit the RIS to make the aggregate reflecting channels of different user terminals (UTs) as similar as possible to be able to extract common group secret keys from their channels. Specifically, the RIS will adjust its parameters to pave the way for GKG based on the physical channels of the UTs. Our method exploits the already gathered channel state information (CSI) in the RIS to beneficially design the phase shifts and does not impose additional probing burden on the network. We consider both passive RIS (PRIS) and active RIS (ARIS) to generate the group keys. The PRIS is widely adopted in physical layer key generation (PLKG) studies due to its use of passive elements, whereas the ARIS demonstrates superior capability in aligning the aggregate reflected channels among nodes in the GKG scenario, as demonstrated in this study. We will exploit various optimization methods like successive convex approximation (SCA) and semidefinite relaxation with Gaussian randomization (SDR-GR) to address the raised optimization problems. Unlike most of the studies in the literature, our scheme can achieve a high GKG rate in static environments as well. Finally, we will examine the performance of the proposed method by normalized mean squared error (NMSE), key error rate (KER), key generation rate (KGR) and key randomness metrics. Our numerical results verify that for the equal available power budget, the ARIS significantly outperforms PRIS in NMSE and KER, achieving more than four times higher KGR.