Search

Identifying humans based on their behavioural patterns represents an attractive basis for access control as such patterns appear naturally, do not require a focused effort from the user side, and do not impose the additional burden of memorising passwords. One means of capturing behavioural patterns is through passive sensors laid out in a target environment. Thanks to the proliferation of the Internet of Things (IoT), sensing devices are already embedded in our everyday surroundings and represent a rich source of multimodal data. Nevertheless, collecting such data for authentication research purposes is challenging, as it entails management and synchronisation of a range of sensing devices, design of diverse tasks that would evoke different behaviour patterns, storage and pre-processing of data arriving from multiple sources, and the execution of long-lasting user activities. Consequently, to the best of our knowledge, no publicly available datasets suitable for behaviour-based authentication research exist. In this brief article, we describe the first multimodal dataset for behavioural authentication research collected in a sensor-enabled IoT setting. The dataset comprises of high-frequency accelerometer, gyroscope, and force sensor data collected from an office-like environment. In addition, the dataset contains 3D point clouds collected with wireless radar and electroencephalogram (EEG) readings from a wireless EEG cap worn by the study participants. Within the environment, 54 volunteers conducted 6 different tasks that were constructed to elicit different behaviours and different cognitive load levels, resulting in a total of 16 h of multimodal data. The richness of the dataset comprising 5 different sensing modalities, a variability of tasks including keyboard typing, hand gesturing, walking, and other activities, opens a range of opportunities for research in behaviour-based authentication, but also the understanding of the role of different tasks and cognitive load levels on human behaviour.

Abstract Radio Frequency Fingerprinting (RFF) has been recently investigated by the research community to enhance wireless security. This letter proposes an RFF approach based only on the complex ramp‐up transient portion of the transmitted frame. The well known issues in literature with the use of the transients for RFF is that they have very short duration, they are strongly non‐stationary and their analysis is sensitive to noise. Then, this letter proposes the application of an ensemble of transforms to the original complex time domain representation in combination with Convolution Neural Network (CNN) to address these issues. To avoid the computing demanding calculation of all the considered transforms, a pre‐processing step based on feature extraction and filter feature selection is implemented to select the most discriminating transforms to reduce the input data to the CNN. The approach is applied to two public RFF data sets, where the complex transient ramp‐up have been extracted. The result shows that the proposed approach is able to obtain a classification accuracy higher than the baseline based on the original signal or selected transforms across various values of Signal to Noise Ratio (SNR).

Cybersecurity in modern vehicles has received increased attention from the research community in recent years. Intrusion Detection Systems (IDSs) are one of the techniques used to detect and mitigate cybersecurity risks. This paper proposes a novel implementation of an IDS for in-vehicle security networks based on the concept of multi-scale histograms, which capture the frequencies of message identifiers in CAN-bus in-vehicle networks. In comparison to existing approaches in the literature based on a single histogram, the proposed approach widens the informative context used by the IDS for traffic analysis by taking into consideration sequences of two and three CAN-bus messages to create multi-scale dictionaries. The histograms are created from windows of in-vehicle network traffic. A preliminary multi-scale histogram model is created using only legitimate traffic. Against this model, the IDS performs traffic analysis to create a feature space based on the correlation of the histograms. Then, the created feature space is given in input to a Convolutional Neural Network (CNN) for the identification of the windows of traffic where the attack is present. The proposed approach has been evaluated on two different public data sets achieving a very competitive performance in comparison to the literature.

Channel identification is a useful function to support wireless telecommunication operations because the knowledge of the radio frequency propagation channel characteristics can improve communication efficiency and robustness. In recent times, the application of machine learning (ML) algorithms to the problem of channel identification has been proposed in the literature. In particular, Deep Learning (DL) has demonstrated superior performance to ’shallow’ machine learning algorithms for many wireless communication functions. Inspired by the success of DL in literature, the authors in this paper apply Convolutional Neural Networks (CNN) to the problem of channel identification, which is still an emerging research area. CNN is a deep learning algorithm that has demonstrated superior performance to ML algorithms, in particular for image processing tasks. Because the digitized RF signal is a one-dimensional time series, different algorithms are applied to convert the time series to images using various Time Frequency Transform (TFT) including the CWTs, spectrogram, and Wigner Ville distribution. The images are then provided as input to the CNN. The approach is applied to a data set based on weather radar pulse signals generated in the laboratory of the author’s facilities on which different fading models are applied. These models are inspired by the tap-delay-line 3GPP configurations defined in the standards, but they have been customized with Nakagami-m fading distribution (3GPP-like fading models). The results show the superior performance of time–frequency CNN in comparison to 1D CNN for different values of Signal to Noise Ratio (SNR) in dB. In particular, the study shows that the Continuous Wavelet Transform (CWT) has the optimal performance in this data set, but the choice of the mother wavelet remains a problem to be solved (this is a well-known problem in the research literature). Then, this study also proposes an adaptive technique for the choice of the optimal mother wavelet, which is evaluated on the mentioned data set. The results show that the adaptive proposed approach is able to obtain the optimal performance for most of the SNR conditions.

Dynamic Spectrum Access (DSA) solutions equipped with spectrum prediction can enable proactive spectrum management and tackle the increasing demand for radio frequency (RF) bandwidth. Among various prediction techniques, Long Short-Term Memory (LSTM) is a deep learning model that has demonstrated high performance in forecasting spectrum characteristics. Although well-performing, the theoretical characterization of LSTM prediction performance has not been well developed in the literature. Therefore, in this article, we examine an LSTM based temporal spectrum prediction model and characterize its prediction performance through theoretical analysis. To this end, we analyze the LSTM prediction outputs over simulated Markov-model-based spectrum data and spectrum measurements data. Our results suggest that the predicted scores of the LSTM based system model can be described using mixtures of truncated Gaussian distributions. We also estimate the performance metrics using the mixture model and compare the results with the observed prediction performance over simulated and measured datasets.

With the strong development of the Internet of Things (IoT), the definition of IoT devices’ intended behavior is key for an effective detection of potential cybersecurity attacks and threats in an increasingly connected environment. In 2019, the Manufacturer Usage Description (MUD) was standardized within the IETF as a data model and architecture for defining, obtaining and deploying MUD files, which describe the network behavioral profiles of IoT devices. While it has attracted a strong interest from academia, industry, and Standards Developing Organizations (SDOs), MUD is not yet widely deployed in real-world scenarios. In this work, we analyze the current research landscape around this standard, and describe some of the main challenges to be considered in the coming years to foster its adoption and deployment. Based on the literature analysis and our own experience in this area, we further describe potential research directions exploiting the MUD standard to encourage the development of secure IoT-enabled scenarios.

The use of mobile phones or smartphones has become so widespread that most people rely on them for many services and applications like sending e-mails, checking the bank account, accessing cloud platforms, health monitoring, buying on-line and many other applications where sharing sensitive data is required. As a consequence, security functions are important in the use of smartphones, especially because most of the applications require the identification and authentication of the device in mobility. This is usually achieved through cryptographic systems but recent research studies have also investigated alternative or complementary authentication mechanisms which can be used to strengthen cryptographic methods with multi-factor authentication. In this paper, we investigate the identification and the authentication of smartphones using the intrinsic physical properties of the mobile phones built-in microphones. The possibility to identify a microphone on the basis of features extracted from audio recordings is well known in literature but it is mostly used in forensics studies and usually relies on human voice recordings. On the contrary this paper proposes a smartphone identification and authentication approach by stimulating the built-in microphone with non-voice sounds at different frequencies. An extensive data set of 32 phones was used to evaluate experimentally the proposed approach. On the basis of the proven performance of deep learning techniques, a new Convolutional Neural Network architecture is proposed both for the identification and the authentication purposes. Its performance, in comparison to other machine learning algorithms, is demonstrated in presence of different types of noises (e.g., Gaussian White noise, Babble noise and Street noise). Satisfactory results have been obtained showing that the exploitation of a fingerprint from the microphone sensor is a good choice to assess smartphone distinctiveness.

Nowadays, images and videos have become the main modalities of information being exchanged in everyday life, and their pervasiveness has led the image forensics community to question their reliability, integrity, confidentiality, and security more and more [...]

Spectrum sensing is an important function in radio frequency spectrum management and cognitive radio networks. Spectrum sensing is used by one wireless system (e.g., a secondary user) to detect the presence of a wireless service with higher priority (e.g., a primary user) with which it has to coexist in the radio frequency spectrum. If the wireless signal is detected, the second user system releases the given frequency to maintain the principle of not interfering. This paper proposes a machine learning implementation of spectrum sensing using the entropy measure as a feature vector. In the training phase, the information about the activity of the wireless service with higher priority is gathered, and the model is formed. In the classification phase, the wireless system compares the current sensing report to the created model to calculate the posterior probability and classify the sensing report into either the presence or absence of wireless service with higher priority. This paper proposes the novel application of the Fluctuation Dispersion Entropy (FDE) measure recently introduced in the research community as a feature vector to build the model and implement the classification. An improved implementation of the FDE (IFDE) is used to enhance the robustness to noise. IFDE is further enhanced with an adaptive method (AIFDE) to automatically select the hyper-parameter introduced in IFDE. Then, this paper combines the machine learning approach with the entropy measure approach, which are both recent developments in spectrum sensing research. The approach is compared to similar approaches in literature and the classical energy detection method using a generated radar signal data set with different conditions of SNR(dB) and fading conditions. The results show that the proposed approach is able to outperform the approaches from literature based on other entropy measures or the Energy Detector (ED) in a consistent way across different levels of SNR and fading conditions.

The Intrusion Detection System (IDS) is an important tool to mitigate cybersecurity threats in an Information and Communication Technology (ICT) infrastructure. The function of the IDS is to detect an intrusion to an ICT system or network so that adequate countermeasures can be adopted. Desirable features of IDS are computing efficiency and high intrusion detection accuracy. This paper proposes a new anomaly detection algorithm for IDS, where a machine learning algorithm is applied to detect deviations from legitimate traffic, which may indicate an intrusion. To improve computing efficiency, a sliding window approach is applied where the analysis is applied on large sequences of network flows statistics. This paper proposes a novel approach based on the transformation of the network flows statistics to gray images on which Gray level Co-occurrence Matrix (GLCM) are applied together with an entropy measure recently proposed in literature: the 2D Dispersion Entropy. This approach is applied to the recently public IDS data set CIC-IDS2017. The results show that the proposed approach is competitive in comparison to other approaches proposed in literature on the same data set. The approach is applied to two attacks of the CIC-IDS2017 data set: DDoS and Port Scan achieving respectively an Error Rate of 0.0016 and 0.0048.

The detection and identification of road anomalies and obstacles in the road infrastructure has been investigated by the research community using different types of sensors. This paper evaluates the detection and identification of road anomalies/obstacles using the data collected from the Inertial Measurement Unit (IMU) installed in a vehicle and in particular from the data generated by the accelerometers’ and gyroscopes’ components. Inspired by the successes of the application of deep learning to various identification problems, this paper investigates the application of Convolutional Neural Network (CNN) to this specific problem. In particular, we propose a novel approach in this context where the time-frequency representation (i.e., spectrogram) is used as an input to the CNN rather than the original time domain data. This approach is evaluated on an experimental dataset collected using 12 different vehicles driving for more than 40 km of road. The results show that the proposed approach outperforms significantly and across different sampling rates both the application of CNN to the original time domain representation and the application of shallow machine learning algorithms. The approach achieves an identification accuracy of 97.2%. The results presented in this paper are based on an extensive optimization both of the CNN algorithm and the spectrogram implementation in terms of window size, type of window, and overlapping ratio. The accurate detection of road anomalies/obstacles could be useful to road infrastructure managers to monitor the quality of the road surface and to improve the accurate positioning of autonomous vehicles because road anomalies/obstacles could be used as landmarks.

Research findings have shown that microphones can be uniquely identified by audio recordings since physical features of the microphone components leave repeatable and distinguishable traces on the audio stream. This property can be exploited in security applications to perform the identification of a mobile phone through the built-in microphone. The problem is to determine an accurate but also efficient representation of the physical characteristics, which is not known a priori. Usually there is a trade-off between the identification accuracy and the time requested to perform the classification. Various approaches have been used in literature to deal with it, ranging from the application of handcrafted statistical features to the recent application of deep learning techniques. This paper evaluates the application of different entropy measures (Shannon Entropy, Permutation Entropy, Dispersion Entropy, Approximate Entropy, Sample Entropy, and Fuzzy Entropy) and their suitability for microphone classification. The analysis is validated against an experimental dataset of built-in microphones of 34 mobile phones, stimulated by three different audio signals. The findings show that selected entropy measures can provide a very high identification accuracy in comparison to other statistical features and that they can be robust against the presence of noise. This paper performs an extensive analysis based on filter features selection methods to identify the most discriminating entropy measures and the related hyper-parameters (e.g., embedding dimension). Results on the trade-off between accuracy and classification time are also presented.

The evolution of modern automobiles to higher levels of connectivity and automatism has also increased the need to focus on the mitigation of potential cybersecurity risks. Researchers have proven in recent years that attacks on in-vehicle networks of automotive vehicles are possible and the research community has investigated various cybersecurity mitigation techniques and intrusion detection systems which can be adopted in the automotive sector. In comparison to conventional intrusion detection systems in large fixed networks and ICT infrastructures in general, in-vehicle systems have limited computing capabilities and other constraints related to data transfer and the management of cryptographic systems. In addition, it is important that attacks are detected in a short time-frame as cybersecurity attacks in vehicles can lead to safety hazards. This paper proposes an approach for intrusion detection of cybersecurity attacks in in-vehicle networks, which takes in consideration the constraints listed above. The approach is based on the application of an information entropy-based method based on a sliding window, which is quite efficient from time point of view, it does not require the implementation of complex cryptographic systems and it still provides a very high detection accuracy. Different entropy measures are used in the evaluation: Shannon Entropy, Renyi Entropy, Sample Entropy, Approximate Entropy, Permutation Entropy, Dispersion and Fuzzy Entropy. This paper evaluates the impact of the different hyperparameters present in the definition of entropy measures on a very large public data set of CAN-bus traffic with millions of CAN-bus messages with four different types of attacks: Denial of Service, Fuzzy Attack and two spoofing attacks related to RPM and Gear information. The sliding window approach in combination with entropy measures can detect attacks in a time-efficient way and with great accuracy for specific choices of the hyperparameters and entropy measures.

A 2.3Tbps DDoS attack was recently mitigated by Amazon, which is a new record after the 2018 GitHub attack, or the famous 2016 Dyn DNS attack launched from hundreds of thousands of hijacked Internet of Things (IoT) devices. These attacks may disrupt the lives of billions of people worldwide, as we increasingly rely on the Internet. In this paper, we tackle the problem that hijacked IoT devices are often the origin of these attacks. With the goal of protecting the Internet and local networks, we propose Autopolicy: a system that automatically limits the IP traffic bandwidth—and other network resources—available to IoT devices in a particular network. We make use of the fact that devices, such as sensors, cameras, and smart home appliances, rarely need their high-speed network interfaces for normal operation. We present a simple yet flexible architecture for Autopolicy, specifying its functional blocks, message sequences, and general operation in a Software Defined Network. We present the experimental validation results, and release a prototype open source implementation.

This paper addresses the problem of authentication and identification of wireless devices using their physical properties derived from their Radio Frequency (RF) emissions. This technique is based on the concept that small differences in the physical implementation of wireless devices are significant enough and they are carried over to the RF emissions to distinguish wireless devices with high accuracy. The technique can be used both to authenticate the claimed identity of a wireless device or to identify one wireless device among others. In the literature, this technique has been implemented by feature extraction in the 1D time domain, 1D frequency domain or also in the 2D time frequency domain. This paper describes the novel application of the synchrosqueezing transform to the problem of physical layer authentication. The idea is to exploit the capability of the synchrosqueezing transform to enhance the identification and authentication accuracy of RF devices from their actual wireless emissions. An experimental dataset of 12 cellular communication devices is used to validate the approach and to perform a comparison of the different techniques. The results described in this paper show that the accuracy obtained using 2D Synchrosqueezing Transform (SST) is superior to conventional techniques from the literature based in the 1D time domain, 1D frequency domain or 2D time frequency domain.

In the transportation sector, safety risks can be significantly reduced by monitoring the behaviour of drivers and by discouraging possible misconducts that entail fatigue and can increase the possibility of accidents. The Smart Tachograph (ST), the new revision of the Digital Tachograph (DT), has been designed with this purpose: to verify that speed limits and compulsory rest periods are respected by drivers. In order to operate properly, the ST periodically checks the consistency of data from different sensors, which can be potentially manipulated to avoid the monitoring of the driver behaviour. In this respect, the ST regulation specifies a test procedure to detect motion conflicts originating from inconsistencies between Global Navigation Satellite System (GNSS) and odometry data. This paper provides an experimental evaluation of the speed verification procedure specified by the ST regulation. Several hours of data were collected using three vehicles and considering light urban and highway environments. The vehicles were equipped with an On-Board Diagnostics (OBD) data reader and a GPS/Galileo receiver. The tests prescribed by the regulation were implemented with specific focus on synchronization aspects. The experimental analysis also considered aspects such as the impact of tunnels and the presence of data gaps. The analysis shows that the metrics selected for the tests are resilient to data gaps, latencies between GNSS and odometry data and simplistic manipulations such as data scaling. The new ST forces an attacker to falsify data from both sensors at the same time and in a coherent way. This makes more difficult the implementation of frauds in comparison to the current version of the DT.

Spectrum Sensing (SS) is an important function in Cognitive Radio (CR) to detect primary users. The design of SS algorithms is one of the most challenging tasks in CR and requires innovative hardware and software solutions to enhance detection probability and minimize low false alarm probability. Although several SS algorithms have been developed in the specialized literature, limited work has been done to practically demonstrate the feasibility of this function on platforms with significant computational and hardware constraints. In this paper, SS is demonstrated using a low cost TV tuner as agile front-end for sensing a large portion of the Ultra-High Frequency (UHF) spectrum. The problems encountered and the limitations imposed by the front-end are analysed along with the solutions adopted. Finally, the spectrum sensor developed is implemented on an Android device and SS implementation is demonstrated using a smartphone.

Falsifying Global Navigation Satellite System (GNSS) data with a simulator or with a fake receiver can have a significant economic or safety impact in many transportation applications where Position, Velocity and Time (PVT) are used to enforce a regulation. In this context, the authentication of the source of the PVT data (i.e., the GNSS receiver) is a requirement since data faking can become a serious threat. Receiver fingerprinting techniques represent possible countermeasures to verify the authenticity of a GNSS receiver and of its data. Herein, the potential of clock-derived metrics for GNSS receiver fingerprinting is investigated, and a filter approach is implemented for feature selection. Novel experimental results show that three intrinsic features are sufficient to identify a receiver. Moreover, the adopted technique is time effective as data blocks of about 40 min are sufficient to produce stable features for fingerprinting.

We investigate the identification of mobile phones through their built-in magnetometers. These electronic components have started to be widely deployed in mass market phones in recent years, and they can be exploited to uniquely identify mobile phones due their physical differences, which appear in the digital output generated by them. This is similar to approaches reported in the literature for other components of the mobile phone, including the digital camera, the microphones or their RF transmission components. In this paper, the identification is performed through an inexpensive device made up of a platform that rotates the mobile phone under test and a fixed magnet positioned on the edge of the rotating platform. When the mobile phone passes in front of the fixed magnet, the built-in magnetometer is stimulated, and its digital output is recorded and analyzed. For each mobile phone, the experiment is repeated over six different days to ensure consistency in the results. A total of 10 phones of different brands and models or of the same model were used in our experiment. The digital output from the magnetometers is synchronized and correlated, and statistical features are extracted to generate a fingerprint of the built-in magnetometer and, consequently, of the mobile phone. A SVM machine learning algorithm is used to classify the mobile phones on the basis of the extracted statistical features. Our results show that inter-model classification (i.e., different models and brands classification) is possible with great accuracy, but intra-model (i.e., phones with different serial numbers and same model) classification is more challenging, the resulting accuracy being just slightly above random choice.

The correct identification of smartphones has various applications in the field of security or the fight against counterfeiting. As the level of sophistication in counterfeit electronics increases, detection procedures must become more accurate but also not destructive for the smartphone under testing. Some components of the smartphone are more likely to reveal their authenticity even without a physical inspection, since they are characterized by hardware fingerprints detectable by simply examining the data they provide. This is the case of MEMS (Micro Electro-Mechanical Systems) components like accelerometers and gyroscopes, where tiny differences and imprecisions in the manufacturing process determine unique patterns in the data output. In this paper, we present the experimental evaluation of the identification of smartphones through their built-in MEMS components. In our study, three different phones of the same model are subject to repeatable movements (composing a repeatable scenario) using an high precision robotic arm. The measurements from MEMS for each repeatable scenario are collected and analyzed. The identification algorithm is based on the extraction of the statistical features of the collected data for each scenario. The features are used in a support vector machine (SVM) classifier to identify the smartphone. The results of the evaluation are presented for different combinations of features and Inertial Measurement Unit (IMU) outputs, which show that detection accuracy of higher than 90% is achievable.