Search

In this research, biometric authentication, which has been widely used for different purposes in the last quarter-century, is studied. Dorsal hand veins are used for biometric authentication. “Deep learning” (DL) and “generative adversarial networks” (GANs) are used together as keys in the study. A DL-GAN is obtained by combining deep learning and GAN. The developed DL-GAN method is tested on two separate databases. The adversarial network (DL-GAN) method is developed to increase the authentication process’s proportional value. For identity verification, dorsal hand veins with biometric physical properties are used. A multistep approach is used for selecting hand dorsal features, including preimage processing and effectively identifying individuals. The deep learning productive antinetwork method is used to effectively identify individuals based on the information obtained from the dorsal hand vein images. For the test in the study, two open access databases are used. These databases are the Jilin University - dorsal hand vein database and the 11K hands database. The results of the experiments performed on the dataset related to the dorsal hand vessels show that the DL-GAN method reaches an identity accuracy level of 98.36% and has an error rate of 2.47% and a standard accuracy of 0.19%. The accuracy of the experimental results in the second dataset is 96.43%, the equal error rate is 3.55% and the standard accuracy is 0.21%. The improved DL-GAN method obtains better results than physical biometric methods such as LBP, LPQ, GABOR, FGM, BGM and SIFT.

Immunohistochemical staining of tissue sections is a vital technique in pathological diagnostics and theranostics. Several kinds of detection systems are available—each of them with their advantages and disadvantages. Here we present the results of a study assessing a prototype immunohistochemical detection technology (PIDT) for visualization of antigens in tissue sections. Different tumor tissues (n = 11) were stained with selected antibodies (n = 30) and a subset of these under different fixation conditions. The staining properties were assessed according to six staining quality parameters (signal distribution, intensity, tissue and slide background, acutance, clarity of details, and subcellular morphological details), and the results were compared with those of a well-established detection system (EnVision FLEX). Overall, both detection methods revealed good to optimal results regarding the evaluated parameters even under unfavorable fixation conditions. However, with the prototype detection technology a quicker turnaround time was reached primarily due to shorter primary antibody incubation times. Moreover, PIDT-stained tissues showed higher signal intensity and a uniform signal distribution over the tissue slide, still, with well-preserved tissue morphology and without impairing the gradation of staining intensity of different cell types. In particular, the prototype detection technology performed better in poorly or delayed fixed tissue. In situations where rapid and profound results are in demand, and particularly in the context of a small laboratory setting, this prototype detection technology could be a useful addition to the established detection systems. Electronic supplementary material The online version of this article (10.1007/s00418-020-01906-5) contains supplementary material, which is available to authorized users.

The novel coronavirus (COVID-19) has become widespread around the world. It started in Wuhan, China, and has since spread rapidly among people living in other countries. Hence, the World Health Organization has considered COVID-19 as a pandemic that threatens millions of people’s lives. Due to the high number of infected people, many hospitals have been facing critical issues in providing the required medical services. For instance, some clinical centers have been unable to provide one of the most important medical services, namely blood tests to determine whether an individual is infected with COVID-19. Therefore, it is important to present an alternative diagnosis option to prevent the further spread of COVID-19. In this paper, a proposed intelligent detection communication system (IDCS) is configured for distributed mobile clinical centers to control the pandemic. In addition, the intelligent system is integrated with the Zigbee communication protocol to build a mobile COVID-19 detection system. The proposed system was trained on X-ray COVID-19 lung images used to identify infected people. The Zigbee protocol and decision tree algorithm were used to design the IDCS. The results of the proposed system show high accuracy 94.69% and accept results according to the performance measurements.

Visible light communications (VLC) technology is emerging as a candidate to meet the demand for interconnected devices’ communications. However, the costs of incorporating specific hardware into end-user devices slow down its market entry. Optical camera communication (OCC) technology paves the way by reusing cameras as receivers. These systems have generally been evaluated under static conditions, in which transmitting sources are recognized using computationally expensive discovery algorithms. In vehicle-to-vehicle networks and wearable devices, tracking algorithms, as proposed in this work, allow one to reduce the time required to locate a moving source and hence the latency of these systems, increasing the data rate by up to 2100%. The proposed receiver architecture combines discovery and tracking algorithms that analyze spatial features of a custom RGB LED transmitter matrix, highlighted in the scene by varying the cameras’ exposure time. By using an anchor LED and changing the intensity of the green LED, the receiver can track the light source with a slow temporal deterioration. Moreover, data bits sent over the red and blue channels do not significantly affect detection, hence transmission occurs uninterrupted. Finally, a novel experimental methodology to evaluate the evolution of the detection’s performance is proposed. With the analysis of the mean and standard deviation of novel K parameters, it is possible to evaluate the detected region-of-interest scale and centrality against the transmitter source’s ideal location.

The emergence of the Internet of Things (IOT) as a result of the development of the communications system has made the study of cyber security more important. Day after day, attacks evolve and new attacks are emerged. Hence, network anomaly-based intrusion detection system is become very important, which plays an important role in protecting the network through early detection of attacks. Because of the development in machine learning and the emergence of deep learning field, and its ability to extract high-level features with high accuracy, made these systems involved to be worked with real network traffic CSE-CIC-IDS2018 with a wide range of intrusions and normal behavior is an ideal way for testing and evaluation . In this paper , we test and evaluate our deep model (DNN) which achieved good detection accuracy about 90% .

IoT networks are increasingly popular nowadays to monitor critical environments of different nature, significantly increasing the amount of data exchanged. Due to the huge number of connected IoT devices, security of such networks and devices is therefore a critical issue. Detection systems assume a crucial role in the cyber-security field: based on innovative algorithms such as machine learning, they are able to identify or predict cyber-attacks, hence to protect the underlying system. Nevertheless, specific datasets are required to train detection models. In this work we present MQTTset, a dataset focused on the MQTT protocol, widely adopted in IoT networks. We present the creation of the dataset, also validating it through the definition of a hypothetical detection system, by combining the legitimate dataset with cyber-attacks against the MQTT network. Obtained results demonstrate how MQTTset can be used to train machine learning models to implement detection systems able to protect IoT contexts.

Spy app is a class of malware for mobile devices that allows an adversary to steal sensitive information. Detecting spy apps is challenging because they do not rely on classic malware techniques, for instance, they use standard services to store stolen data, and do not perform privileges escalation on the victim phone. Thus, their behavior is generally closer to the benign apps and poses new challenges for their detection. In this paper, we propose ASAINT: A Spy App Identification System based on Network Traffic. To the best of our knowledge, ASAINT is the first system capable of detecting spy apps in a network without any physical or software control of the victim mobile device. Core of our approach is a wide range of non-intrusive network detection methods designed by studying several popular spy apps. We test ASAINT on a self-collected dataset containing network traffic from both spy and benign applications, either on Android and iOS. Our result is an F1-score of 0.85 on average, that confirms the effectiveness of ASAINT. Moreover, our analysis provides a methodological classification of the exfiltration strategies used by spy apps in different operating systems. In sum, our work gives new and practical insights about the detection of modern spy apps, paving the way for future research in detecting this class of malware.

The rapid increase in network traffic has recently led to the importance of flow-based intrusion detection systems processing a small amount of traffic data. Furthermore, anomaly-based methods, which can identify unknown attacks are also integrated into these systems. In this study, the focus is concentrated on the detection of anomalous network traffic (or intrusions) from flow-based data using unsupervised deep learning methods with semi-supervised learning approach. More specifically, Autoencoder and Variational Autoencoder methods were employed to identify unknown attacks using flow features. In the experiments carried out, the flow-based features extracted out of network traffic data, including typical and different types of attacks, were used. The Receiver Operating Characteristics (ROC) and the area under ROC curve, resulting from these methods were calculated and compared with One-Class Support Vector Machine. The ROC curves were examined in detail to analyze the performance of the methods in various threshold values. The experimental results show that Variational Autoencoder performs, for the most part, better than Autoencoder and One-Class Support Vector Machine. Scientific and Technological Research Council of Turkey (TUBITAK)Turkiye Bilimsel ve Teknolojik Arastirma Kurumu (TUBITAK) [2211] The work of Sultan Zavrak was supported by the Scienti~c and Technological Research Council of Turkey (TUBITAK) through the 2211/C Ph.D. Scholarship Programme for Priority Areas. WOS:000544044400003 2-s2.0-85086985318

Vehicle delay is one of the most important performance measures of effectiveness (PMOE) in intersection traffic operations. It allows traffic engineers to evaluate the performance of a traffic system component or the effectiveness of the system-wide control strategy. It is often utilized for real-time applications such as adaptive signal control, congestion management, and dynamic traffic assignment. However, obtaining intersection performance data in real time such as intersection control delay can be very time consuming and labor intensive. This paper develops and tests a reliable method called the Automated Vehicle Delay Estimation Technique (AVDET), which automatically estimates delay at a signalized intersection using detector data and signal timing information from the existing traffic signal controller. Results from the delay estimation algorithm were compared with those from simulation, followed by statistical tests under varying traffic operation conditions. The findings showed that AVDET was able to provide effective results under different traffic and signal control scenarios. Future work of field implementation for the proposed algorithms is recommended to investigate the model reliability and effectiveness in real-time traffic conditions.

International audience; Today, industrial control systems (ICS) are connected to the Internet. The addition of computer capabilities to ICSs has improved performance, reduced costs, and many other benefits. However, ICSs are not designed with security in mind. These systems, therefore, present security risks related to their IT vulnerabilities. Applying IT's security solutions to the ICS are not enough due to the difference between these two systems represented by the interaction between the cyber and physical system in ICS. Hence, it seems relevant to rely on a physical model of the cyber-physical system to obtain an intrusion detection system (IDS) for those systems. Most of the IDSs are based on rules to define how the IDS would detect attacks on the system. These rules are generally used either to describe possible attacks scenarios on the systems or the normal system behavior. However, creating and maintaining handmade rules for a complex system could be very difficult. In this article, we propose a model-based approach that consists of generating IDS rules from a physical system model and the authorized behavior specification. Then we illustrate the proposed algorithm in a case study.

In the 500 kV hybrid high-voltage direct current (HVDC) circuit breaker, there are five 110 kV fast mechanical switches (FMSs) in the nominal current path. When the detection system detects the failure of the hybrid circuit breaker, in order to protect the disconnectors or semiconductors, the FMSs are required to close quickly. The closing drive technology can be divided into two categories, based on the closing speed of the Thomson-coil actuator. One is based on the conventional closing drive, the open−close operating time of which is 20.0 ms. The other type is based on the fast closing drive, which is used under fault conditions and the total time is ∼7.5 ms. A 110 kV/3 kA prototype FMS employing an interrupter with a 30 mm maximum open gap and 4 kg total moving mass has been modelled based on the equivalent circuit method. Through the simulation, the effect of discharging time sequence on open−close operation has been studied, and the general timing control design principle has been provided.

The openness of the World Wide Web (Web) has become more exposed to cyber-attacks. An attacker performs the cyber-attacks on Web using malware Uniform Resource Locators (URLs) since it widely used by internet users. Therefore, a significant approach is required to detect malicious URLs and identify their nature attack. This study aims to assess the efficiency of the machine learning approach to detect and identify malicious URLs. In this study, we applied features optimization approaches by using a bio-inspired algorithm for selecting significant URL features which able to detect malicious URLs applications. By using machine learning approach with static analysis technique is used for detecting malicious URLs applications. Based on this combination as well as significant features, this paper shows promising results with higher detection accuracy. The bio-inspired algorithm: particle swarm optimization (PSO) is used to optimized URLs features. In detecting malicious URLs, it shows that naïve Bayes and support vector machine (SVM) are able to achieve high detection accuracy with rate value of 99%, using URL as a feature.

With the purpose of identifying cyber threats and possible incidents, intrusion detection systems (IDSs) are widely deployed in various computer networks. In order to enhance the detection capability of a single IDS, collaborative intrusion detection networks (or collaborative IDSs) have been developed, which allow IDS nodes to exchange data with each other. However, data and trust management still remain two challenges for current detection architectures, which may degrade the effectiveness of such detection systems. In recent years, blockchain technology has shown its adaptability in many fields, such as supply chain management, international payment, interbanking, and so on. As blockchain can protect the integrity of data storage and ensure process transparency, it has a potential to be applied to intrusion detection domain. Motivated by this, this paper provides a review regarding the intersection of IDSs and blockchains. In particular, we introduce the background of intrusion detection and blockchain, discuss the applicability of blockchain to intrusion detection, and identify open challenges in this direction.

Fuel cell technology is considered to be the 21st century of choice for clean and efficient power generation technology, due to a fuel cell itself is a very complex physical and chemical process, the input output are different kinds of physical quantities, so a practical fuel cell must have accurate monitoring and control the performance of these physical quantities. Are briefly introduced in this paper the characteristic and principle of the fuel cell, the function of fuel cell system and the fuel cell test system are analyzed, and made a specific fuel cell test system implementation scheme. The system Atmega32 is the core, can finish the collection work of various data quantity very well, theapplication fuel cell inspection meter can complete to the battery voltage and the current measurement. This system has great flexibility, except for the detection of PEMFC, it can also be adapted to different types of fuel cell detection.

Introduction. The article deals with the optimization of a moving target signals detection systems against clutter on probabilistic criteria. The object of research is the detection system implementing coherent rejection of clutter followed by coherent or non-coherent accumulation of rejection residues. The purpose is to optimize rejector filter weight coefficients depending on the correlation properties of clutter on probabilistic criteria. Probabilistic characteristics of the detection systems. The expressions for the probability characteristics of the detection systems with coherent rejection clutter and subsequent coherent or non-coherent accumulation of rejection residues are obtained respectively. These expressions establish a functional relationship between the averaged on Doppler signal phase of the correct detection probability and correlation clutter parameters and detection system characteristics. The optimization procedure. The criteria to optimize the weight vector of the rejector filter is listed. These criteria make it possible to communicate the optimal weight vector with clutter parameters on the basis of non-linear programming methods. The quasi-Newton iterative procedure for finding the optimal vector is shown. The restrictions are introduced on samples rejector filter frequency response to achieve unimodal extremum. The numerical results of optimization. We consider the results of numerical optimization system of coherent rejection and subsequent non-coherent accumulation on probabilistic criteria. A comparison of their results with the same optimization of the rejector filter on energy criteria is conducted. Conclusion. The proposed optimization method of detection systems on probabilistic criteria allows to obtain significant gains in the efficiency of detection signals compared with the optimization on energy criteria and to implement the limit efficiency for this class of systems., В статье рассмотрена оптимизация систем обнаружения сигналов движущихся целей на фоне пассивных помех по вероятностному критерию. Получены выражения для вероятностных характеристик соответствующих систем обнаружения. Приведены критерии оптимизации весового вектора режекторного фильтра, позволяющие установить связь оптимального весового вектора с параметрами помехи на основе методов нелинейного программирования. Приведена квазиньютоновская итерационная процедура поиска оптимального вектора. Получены числовые результаты оптимизации системы обнаружения. Установлены существенные выигрыши в эффективности обнаружения сигналов по сравнению с оптимизацией систем обнаружения по энергетическому критерию., У статті розглянута оптимізація систем виявлення сигналів рухомих цілей на тлі пасивних перешкод за ймовірнісним критерієм. Отримані вирази ймовірнісних характеристик систем виявлення. Наведено критерії оптимізації вагового вектора режекторного фільтра, що дозволяють встановити зв'язок між оптимальним ваговим вектором та параметрами перешкоди на основі методів нелінійного програмування. Наведено приклад квазіньютонівської ітераційної процедури пошуку оптимального вектора. Встановлено істотні виграші в ефективності виявлення сигналів у порівнянні з оптимізацією систем виявлення за енергетичним критерієм.

ISBN: 978-1-4503-4490-6 ; Conference Code:130944; International audience; Stance detection systems often integrate social clues in their algorithms. While the influence of social groups on stance is known, there is no evaluation of how well state-of-the-art community detection algorithms perform in terms of detecting like-minded communities, i.e. communities that share the same stance on a given subject. We used Twitter's social interactions to compare the results of community detection algorithms on datasets on the Scottish Independence Referendum and US Midterm Elections. Our results show that algorithms relying on information diffusion perform better for this task and confirm previous observations about retweets being better vectors of stance than mentions.

For elderly people fall incidents are life-changing events that lead to degradation or even loss of autonomy. Current fall detection systems are not integrated and often associated with undetected falls and/or false alarms.In this paper, a social- and context-aware multi-sensor platform is presented, which integrates information gathered by a plethora of fall detection systems and sensors at the home of the elderly, by using a cloud-based solution, making use of an ontology. Within the ontology, both static and dynamic information is captured to model the situation of a specific patient and his/her (in)formal caregivers. This integrated contextual information allows to automatically and continuously assess the fall risk of the elderly, to more accurately detect falls and identify false alarms and to automatically notify the appropriate caregiver, e.g., based on location or their current task.The main advantage of the proposed platform is that multiple fall detection systems and sensors can be integrated, as they can be easily plugged in, this can be done based on the specific needs of the patient. The combination of several systems and sensors leads to a more reliable system, with better accuracy. The proof of concept was tested with the use of the visualizer, which enables a better way to analyze the data flow within the back-end and with the use of the portable testbed, which is equipped with several different sensors. HighlightsWe propose a new extensible and adaptable cloud-based fall detection framework.An ontology-based approach is used to model the situation of patient and caregivers.The captured information from the sensors is combined with contextual information.Existing fall detection systems can easily be integrated to improve accuracy.Different combinations of fall detection systems and sensors can be used.

The Detection and Prevention system against many attacks has been formulated in Mobile ad hoc networks to secure the data and to provide the uninterrupted service to the legitimate clients. The formulation of opinion of neighbors or belief value or Trust value plays vital role in the detection system to avoid attacks. The attack detection system always extracts the behaviors of nodes to identify the attack patterns and prediction of future attacks. The False positives and false negatives plays vital role on identification of attackers accurately without any false positives and negatives .Our system uses the Aumann agreement theorem for convergence of Truth on opinion based on the bound of confidence value, such that truth consensus will maintained, The accuracy of system will be enhanced through this methodology

Automated detection of snow avalanches is an important tool for avalanche forecasting and for assessing the effectiveness of avalanche control measures at bad visibility. Avalanche detection systems are usually based on infrasound, seismic, or radar signals. Within this study, we compared three different types of avalanche detection systems: one avalanche radar, one infrasound array system consisting of four infrasound sensors, and a newly developed single sensor infrasound system. A special focus is given to the new single sensor system, which is a low cost, easy to install system, originally designed for the detection of debris flows and debris floods. Within this work, we analysed how this single sensor system could be adapted to detect also snow avalanches. All three systems were installed close to a road near Ischgl (Tyrol, Austria) at the avalanche-exposed Paznaun Valley. The valley is endangered by two avalanche paths which are controlled by several avalanche towers. The radar system detected avalanches accurately and reliably but was limited to the particular avalanche path towards which the radar beam was directed. The infrasound array could detect avalanches from all surrounding avalanche paths, however, with a higher effort for installation. The newly tested single infrasound sensor system was significantly cheaper and easier to install than the other two systems. It could also detect avalanches form all directions, although without information about the direction. In summary, each of the three different systems was able to successfully detect avalanches and had its particular strengths and weaknesses, which should be considered according to the specific requirements of a particular practical application.