In this paper, we investigate the pertinence of the angle of arrival (AoA) as a feature for robust physical layer authentication (PLA). While most of the existing approaches to PLA focus on amplitude-dependent features of the physical layer of communication channels, such as channel frequency response, channel impulse response, or received signal strength, the use of AoA in this domain has not yet been studied in depth, particularly regarding the ability to thwart spoofing (impersonation) attacks. In this work, we demonstrate that an impersonation attack targeting AoA-based PLA is only feasible under strict conditions on the attackers location, which highlights the AoA’s role as a strong feature for unspoofable PLA, especially when 2D AoA is employed.We extend previous works considering a single-antenna attacker to the case of a multiple-antenna attacker, and we develop a theoretical characterization of the conditions under which a successful impersonation attack can be mounted. Furthermore, we have performed extensive simulations in support of theoretical analyses, to validate the robustness of AoA-based PLA.

Motion-robust, continuous, and contactless respiration monitoring has the potential to enhance clinical assessments by enabling more frequent inclusion of respiratory rate in cardiopulmonary risk evaluation. Depth sensing can be used to extract the torso surface area trend over time, which is the result of the breathing process. Movement artifacts still represent a major problem. In this paper, four subjects are asked to walk and breathe freely in a volume measurement. A commercial wearable chest strap is employed as a ground truth. Each subject pose is tracked by deep learning. Four regions of interest are evaluated for respiratory extraction: the entire torso, chest, abdomen, and chest centroid. To compensate for the effect of motion, the nose, the middle of the shoulder, and the abdomen are considered as possible movement reference points and subtracted from the signal extracted from the previously mentioned region of interest. The best results given by the chest area correspond to a Mean Absolute Error (MAE) of 0.958 bpm, a standard deviation (STD) of 2.132 bpm, and a Root Mean Square Value (RMSE) of 2.338 bpm.

Respiratory rate is a relevant parameter to be monitored to prevent adverse cardiopulmonary situations. Radar sensing represents a promising alternative to wearable sensors, since it gives the opportunity to remotely monitor a subject, while maintaining privacy and fairness. In this work, radar is used to enable a single sensor to track and extract the respiratory rate of a freely moving subject. To validate performance, the subject first walks in a straight line and then moves randomly within a room. As a reference, they are instructed to control their breathing pace using a metronome. A state of art deep learning algorithm, previously developed for person detection purposes and called mmDetect, is applied to localize and track the subject. The developed processing pipeline allows the extraction of the beat signal phase shift associated with chest displacement at each location, which is then used to reconstruct a phase signal linked to the respiratory phenomenon. A resulting measurement error of 1 breath per minute (BPM) is demonstrated.

Automotive radars could have a pivotal role in human-machine interfaces thanks to their ability to monitor human activity and physiological state in a contactless way. Nowadays, data-driven models, such as deep learning approaches, are the needed cutting-edge technology to achieve appropriate results in the classification of activities. One of the issues in the literature is the identification of walking patterns from the radar acquisition and doing so in real-time would be beneficial to give instant feedback to the user. For this reason, we propose a model composed of a Convolutional Neural Network followed by a bidirectional Long-Short Term Memory for the classification of the range-Doppler-time data obtained from radar acquisition. The approach reaches at least a 90% f1 score for the abnormal walking patterns class observing just one gait cycle or one second of acquisition and it is a perfect anomaly detector with 2.6 seconds of acquisition time. Instead, to achieve good accuracy on the classification of all the classes we need a larger window of observation with an overall accuracy of 89.1% for 8 seconds and 95.6% for the complete acquisition, lasting 12 to 16 seconds.

Cyber risk assessment is one of the top priorities of modern organizations and companies, owing to the massive amount of data they process on a daily basis and to the increasing number of successful cyber attacks. The probability of occurrence of these cyber incidents can be estimated by means of statistical tools, which exploit numerical categories to compute the probability that the organization will be breached by one or more cyber attacks. However, these approaches heavily rely on experts' estimates and/or on past data, which are not always available. In this paper we show that, by exploiting machine learning tools, cyber risk can be assessed by using some easily obtainable parameters (called maturity, complexity, attractiveness) representing the cyber posture of the organization under exam. To validate the method we propose, we apply it to three organizations in the healthcare sector having different values of maturity and complexity. The results highlight how the model can be successfully used to assign each organization a class of cyber risk, even in a crucial sector such as healthcare.

In this paper, we study the use of the angle of arrival (AoA) as a feature for performing robust, machine learning (ML)-based physical layer authentication (PLA). In fact, whereas most previous research on PLA relies on physical properties such as channel frequency/impulse response or received signal strength, the use of the AoA in this context has not yet been studied in depth as a means of providing resistance to impersonation (spoofing) attacks. In this study, we first prove that an effective impersonation attack on AoA-based PLA can only succeed under very stringent conditions on the attacker in terms of location and hardware capabilities, and thus, the AoA can in many scenarios be used as a robust feature for PLA. In addition, we exploit machine learning in our study to perform lightweight, model-free, intelligent PLA. We show the effectiveness of the proposed AoA-based PLA solutions by testing them on experimental outdoor massive multiple input multiple output data.

With the progressive aging of the population, monitoring the state of frailty of a person becomes increasingly important to prevent risk factors, which can lead to loss of autonomy and to hospitalization. Hygiene care, in particular, represents a wake-up call to detect a decline in physical and mental well-being. With the assistance of both environmental and localized sensors, measurements of hygiene-related activities can be made quickly and consistently over time. We here propose to remotely monitor these activities using a fixed camera and deep learning algorithms. In particular, three activities are considered, i.e., washing face, brushing teeth and arranging hair, together with the non-action class. Considering a dataset consisting of 11 healthy subjects of different age and sex, we show that using a Long-Short Term Memory (LSTM) neural network the selected activities can be distinguished with an accuracy of more than 92%, thus proving the validity of the proposed approach.

The use of flying Unmanned Aerial Vehicles (UAVs) for communications is becoming more and more widespread, especially in 5G and beyond networks. In such a context, detection and authentication of UAVs is assuming an increasingly important role. In this paper we show that it is possible to distinguish different drones which communicate with a fixed ground base station (BS) on the basis of their channel characteristics and of the micro-Doppler signature associated to the specific features of each UAV. An urban scenario is simulated where UAVs fly at a constant height and channels are affected by Additive White Gaussian Noise (AWGN) and fading. With the aim of helping the BS in its authentication task, we take advantage of a sparse autoencoder trained on the channel of the legitimate transmitter, while data coming from possible attackers are classified as anomalies. We prove that, with proper network training, low levels of false alarm and missed detection can be achieved, especially if the attacker has no line-of-sight link, and that the presence of micro-Doppler actually contribute to enhance the authentication performance.

Sleep quality is an index of well-being, since sleep disorders, such as sleep apnea, may constitute a health risk. A constant monitoring of subjects, especially when there are heart or respiratory diseases, is essential. The present paper aims to offer a non-invasive and comfortable sleep monitoring, by employing a BallistoCardioGraphic (BCG) signal processing. In particular, with a BCG device located below the mattress, we are able to extract the heart rate, respiratory rate and, therefore, to exploit this information to develop an automatic sleep apnea recognition algorithm. The automatic approach presented has proven to achieve accuracy and reliability and could represent a valid resource to prevent serious damages during sleep.

In the recent years the number of vehicles on the road equipped with radar sensors is increased, especially thanks to their ability to support the main Advanced Driver Assisted Systems (ADAS), such as cruise control or assisted braking. This represents a great improvement for safety, but it has a main disadvantage: all these sensors in fact transmit at the same frequency and may suffer from mutual interference, leading to incorrect detection of the targets, which causes an incorrect operation for ADAS systems. However, the determination of the effects of this mutual interference requires the definition of a theoretical model of interference, the design of which, capable of adapting to real devices, is complex. In fact, depending on how the sensors are designed, current automotive radars generate signals in different ways and the interference patterns found in the literature do not always correspond to real-life scenarios. Therefore experimental validation represents the key to understanding and mitigating the interference. In this work, an experimental analysis of radar interference based on range-Doppler maps is carried out using two automotive radars, providing also a comparison with a theoretical model.

In this article we consider authentication at the physical layer, in which the authenticator aims at distinguishing a legitimate supplicant from an attacker on the basis of the characteristics of a set of parallel wireless channels, which are affected by time-varying fading. Moreover, the attacker's channel has a spatial correlation with the supplicant's one. In this setting, we assess and compare the performance achieved by different approaches under different channel conditions. We first consider the use of two different statistical decision methods, and we prove that using a large number of references (in the form of channel estimates) affected by different levels of time-varying fading is not beneficial from a security point of view. We then consider classification methods based on machine learning. In order to face the worst case scenario of an authenticator provided with no forged messages during training, we consider one-class classifiers. When instead the training set includes some forged messages, we resort to more conventional binary classifiers, considering the cases in which such messages are either labelled or not. For the latter case, we exploit clustering algorithms to label the training set. The performance of both nearest neighbor (NN) and support vector machine (SVM) classification techniques is evaluated. Through numerical examples, we show that under the same probability of false alarm, one-class classification (OCC) algorithms achieve the lowest probability of missed detection when a small spatial correlation exists between the main channel and the adversary one, while statistical methods are advantageous when the spatial correlation between the two channels is large.

The detection of vital parameters with traditional approaches, as the electrocardiograph, requires to appropriately place electrodes in direct contact with patients’ skin, often causing irritation. On the other hand, contactless measurement of physiological parameters provides an unobtrusive and comfortable instrument for subjects’ conditions monitoring, with application to home monitoring of aging people and in particular to those suffering of heart disease. In this paper two contactless techniques are proposed, based on radar technology and on video processing from an RGB camera. In order to validate their precision, the proposed methods are compared with three wearable low cost devices, taken as a reference for the outcomes. The developed approaches prove to achieve excellent performances, with an estimated mean relative error of 0.55% with respect to a commercial cardiac strap device.

Analysis of a person's movement provides important information about his or her health status. This analysis can be performed with wearable devices or with contactless technologies. These latter in particular are of some interest, since the subject is free to move and the analysis of the movement is realistic. Despite being designed for other purposes, automotive mmWaves radars represent a powerful low-cost technology for detecting people's movements without contact which finds interesting applications as a support for home monitoring of health conditions. In this paper it is shown how to exploit commercial radars to distinguish with high precision the way of walking of a subject and the position of his hands during the activity carried out. The application of Principal Component Analysis (PCA) for feature extraction from raw data is considered, together with supervised machine learning algorithms for the actual classification of the various activities carried out during the experiments.

In this paper we assess the security performance of key-less physical layer authentication schemes in the case of time-varying fading channels, considering both partial and no channel state information (CSI) on the receiver's side. We first present a generalization of a well-known protocol previously proposed for flat fading channels and we study different statistical decision methods and the corresponding optimal attack strategies in order to improve the authentication performance in the considered scenario. We then consider the application of machine learning techniques in the same setting, exploiting different one-class nearest neighbor (OCNN) classification algorithms. We observe that, under the same probability of false alarm, one-class classification (OCC) algorithms achieve the lowest probability of missed detection when a low spatial correlation exists between the main channel and the adversary one, while statistical methods are advantageous when the spatial correlation between the two channels is higher.

We consider the problem of downloading content from a cellular network that is cached at the wireless edge while achieving privacy. In particular, we consider private information retrieval (PIR) of content from a library of files, i.e., the user wishes to download a file and does not want the network to learn any information about which file she is interested in. To reduce the backhaul usage, content is cached at the wireless edge in a number of small-cell base stations (SBSs) using maximum distance separable codes. We propose a PIR scheme based on generalized Reed-Solomon codes for this scenario that achieves privacy against a number of spy SBSs that collaborate. The proposed PIR scheme is an extension of a scheme by Kumar et al. to the case of multiple code rates, suitable for the scenario where files have different popularities. We derive the backhaul rate and optimize the content placement to minimize it. We prove that uniform content placement is optimal, i.e., all files that are cached should be stored using the same code rate. This is in contrast to the case where no PIR is required. Furthermore, we show numerically that popular content placement is optimal for some scenarios.

Transmissions over the wiretap channel have been studied for a long time from the information theory standpoint. This has allowed to assess the secrecy performance against eavesdropping while ensuring reliable transmission towards the legitimate receiver. However, most previous studies rely on a number of assumptions which are far from practical wireless communications, like infinite length codewords, random coding, discrete channels or continuous channels with Gaussian signaling. In this paper, we show how the level of security at the physical layer can be assessed from the information theoretic standpoint while taking into account the constraints of practical transmissions over realistic wireless wiretap channels, i.e., by considering practical codes with finite length, discrete modulation formats and continuous channels with fading. For this purpose, we consider the notion of mutual information security, which is provably equivalent to semantic security. Our target is to show that classical and already implemented coding and modulation schemes can be used to achieve some level of security at the physical layer, opposed to approaches resorting to completely new designs tailored to secure transmissions. To corroborate this thesis, we consider some coding and modulation schemes compliant with the IEEE 802.16e (WiMax) standard and show how they can be used to achieve some given security level.

Distributed storage systems and caching systems are becoming widespread, and this motivates the increasing interest on assessing their achievable performance in terms of reliability for legitimate users and security against malicious users. While the assessment of reliability takes benefit of the availability of well established metrics and tools, assessing security is more challenging. The classical cryptographic approach aims at estimating the computational effort for an attacker to break the system, and ensuring that it is far above any feasible amount. This has the limitation of depending on attack algorithms and advances in computing power. The information-theoretic approach instead exploits capacity measures to achieve unconditional security against attackers, but often does not provide practical recipes to reach such a condition. We propose a mixed cryptographic/information-theoretic approach with a twofold goal: estimating the levels of information-theoretic security and defining a practical scheme able to achieve them. In order to find optimal choices of the parameters of the proposed scheme, we exploit an effective probabilistic model checker, which allows us to overcome several limitations of more conventional methods.