This letter considers a data-driven approach to a secret key rate analysis of physical layer secret key generation based on mutual information neural estimator (MINE), which achieves the state-of-the-art estimation accuracy. Since MINE does not require any statistical knowledge of randomness sources, the proposed approach is applicable to any source model and even real data which cannot be fully expressed by a simple model. We first demonstrate by simulations that the proposed estimation provides good approximations to the theoretical secret key rate for a simple model, and then show that the achievable secret key rate highly depends on sources of randomness and employed channel estimation schemes for a realistic wireless channel model. As an application of our analysis, we present a method to train quantization through backpropagation that maximizes the estimated secret key rate.
This letter considers a data-driven approach to a secret key rate analysis of physical layer secret key generation based on mutual information neural estimator (MINE), which achieves the state-of-the-art estimation accuracy. Since MINE does not require any statistical knowledge of randomness sources, the proposed approach is applicable to any source model and even real data which cannot be fully expressed by a simple model. We first demonstrate by simulations that the proposed estimation provides good approximations to the theoretical secret key rate for a simple model, and then show that the achievable secret key rate highly depends on sources of randomness and employed channel estimation schemes for a realistic wireless channel model. As an application of our analysis, we present a method to train quantization through backpropagation that maximizes the estimated secret key rate.
As an alternative and appealing approach to providing information security in wireless communication systems, secret key generation at physical layer has demonstrated its potential in terms of efficiency and reliability over traditional cryptographic methods. Without the necessity of a management centre for key distribution or reliance on computational complexity, physical layer key generation protocols enable two wireless entities to extract identical and dynamic keys from the randomness of the wireless channels associated with them.\nIn this thesis, the reliability of secret key generation at the physical layer is examined in practical wireless channels with imperfect channel state information (CSI). Theoretical analyses are provided to relate key match rate with channel's signal-to-noise ratio (SNR), degrees of channel reciprocity, and iterations of information reconciliation.\nIn order to increase key match rate of physical layer secret key generation, improved schemes in the steps of channel estimation and sample quantization are proposed respectively. In the channel estimation step, multiple observations of the wireless channels are integrated with a linear processor to provide a synthesized and more accurate estimation of the wireless channel. In the sample quantization step, a magnitude based quantization method with two thresholds is proposed to quantize partial samples, where specific quantization areas are selected to reduce cross-over errors. Significant improvements in key match rate are proven for both schemes in theoretical analysis and numerical simulations. Key match rate can even achieve 100% in both schemes with the assistance of information reconciliation process.\nIn the end, a practical application of physical layer secret key generation is presented, where dynamic keys extracted from the wireless channels are utilized for securing secret data transmission and providing efficient access control.
The reciprocity of wireless channels is a prerequisite for physical layer secret key generation (SKG). However, inherent factors, such as noise, asynchronous observations, and hardware impairments, disrupt the ideal reciprocity in the channel state information (CSI) observed by the two legitimate parties. To address this issue, we propose a mutual information-driven autoencoder (MIAE) architecture to extract reciprocal channel features from the non-ideal channel observations of legitimate parties. MIAE is constructed with an AutoEncoder Network (AENet) and a mutual information neural estimator (MINE). Specifically, AENet employs a structure with dual encoders and a shared decoder. The two encoders, integrated with convolutional block attention modules (CBAMs), are designed to focus on reciprocal features within CSI observations and to compensate for temporal variations caused by asynchronous observations. The shared encoder and the correspondingly designed loss function further concentrate the autoencoder’s reciprocity enhancement capability into the encoders. MINE is integrated into the proposed MIAE to estimate the mutual information between the channel features of the two parties. This estimation is then used to formulate a mutual information loss, which guides the encoders to learn channel features that closely match the optimal distribution, thereby boosting the key generation rate. Furthermore, a complete SKG scheme is designed based on the proposed channel feature extractor, MIAE. Simulation results show that our proposed MIAE can extract channel features with strong reciprocity and thus achieve excellent SKG performance based on the extracted features. The generalization performance of the proposed MIAE architecture for communication scenarios of different scales has also been examined.
Secret key generation at physical layer has attracted more and more attentions as an emerging cryptography method. Compared to traditional security approaches, secret key generation at physical layer not only avoids the problem of key distribution, but also holds high efficiency and low complexity in application. Based on the principle of channel reciprocity, Alice and Bob both estimate the channel conditions and extract shared keys from this source of common randomness. In this paper, we first analyze the basic steps (channel estimation, sample quantization and key reconciliation) of secret key generation, key match rate with different quantization levels and key reconciliation times are also simulated. While in practical situations, different non-reciprocity factors affect the channel estimation step, key match rate can be greatly decreased and hardly meet real time cryptography requirements. In order to increase the key match rate, the unified framework of physical layer key generation has been extended to utilizing multiple observations of wireless channels to generate secret keys. An improved key generation approach with multiple observations can well deal with discrepancies between transceivers and keep increasing the key match rate. Theoretical analysis and simulation results both validate the significant improvement due to multiple observations. With an increased number of observations on both sides, the desired key match rate can be achieved much greater than with a single observation, and also the probability of key recovery by Eve can be decreased.
By exploiting multipath fading channels as a source of common randomness, physical layer (PHY) based key generation protocols allow two terminals with correlated observations to generate secret keys with information-theoretical security. The state of the art, however, still suffers from major limitations,e.g., low key generation rate, lower entropy of key bits and a high reliance on node mobility. In this paper, a novel cooperative key generation protocol is developed to facilitate high-rate key generation in narrowband fading channels, where two keying nodes extract the phase randomness of the fading channel with the aid of relay node(s). For the first time, we explicitly consider the effect of estimation methods on the extraction of secret key bits from the underlying fading channels and focus on a popular statistical method - maximum likelihood estimation (MLE). The performance of the cooperative key generation scheme is extensively evaluated theoretically. We successfully establish both a theoretical upper bound on the maximum secret key rate from mutual information of correlated random sources and a more practical upper bound from Cramer-Rao bound (CRB) in estimation theory. Numerical examples and simulation studies are also presented to demonstrate the performance of the cooperative key generation system. The results show that the key rate can be improved by a couple of orders of magnitude compared to the existing approaches.
In this work we investigate the problem of secret key generation (SKG) between two communicating wireless devices, according to the physical-layer security paradigm. We propose a general framework for any communication channel or waveform. In details, we study a filterbank-based method, which allows the generation of secret security keys from a wideband radio channel, independently from the baseband modem implementation. We believe that channel awareness is of utmost importance to understand the applicability of SKG methods, i.e., knowing their secret bit-rate under different channel scenarios. For that purpose, we investigate the SKG performance by means of Monte Carlo simulations that collect radio channel statistics and obtain the SKG performance through mutual information numerical estimation.
An emerging area of interest in wireless security is the generation of secret encryption keys by leveraging the features of the physical propagation channel. The shared randomness of the wireless channel between two legitimate nodes in a communication network is exploited to provide strong secrecy. The secret keys are generated based on fluctuations or evolution of the wireless channel state. However, to the best of our knowledge, little work has been performed on algorithms to exploit the increased randomness available to wireless nodes equipped with MIMO antennas. In this paper, we present a practical key generation scheme for MIMO channels based on quantization performed on blind channel estimates. Our schemes generate random keys for transmitter and receiver by simultaneously quantizing the blind channel estimates. Our schemes are simulated for a realistic fading channel model, and performance metrics for error rate and efficiency are provided.