Biblio

Cyber security is a crucial factor for modern power system as many applications are heavily relied on the result of state estimation. Therefore, it is necessary to assess and enhance cyber security for new applications in power system. As an emerging technology, smart grid topology control has been investigated in stability and reliability perspectives while the associated cyber security issue is not studied before. In successful false data injection attack (FDIA) against AC state estimation, attacker could alter online stability check result by decreasing real power flow measurement on the switching target line to undermine physical system stability in topology control. The physical impact of FDIA on system control operation and stability are illustrated. The vulnerability is discussed on perfect FDIA and imperfect FDIA against residue based bad data detection and corresponding countermeasure is proposed to secure critical substations in the system. The vulnerability and countermeasure are demonstrated on IEEE 24 bus reliability test system (RTS).

Vulnerability being the buzz word in the modern time is the most important jargon related to software and operating system. Since every now and then, software is developed some loopholes and incompleteness lie in the development phase, so there always remains a vulnerability of abruptness in it which can come into picture anytime. Detecting vulnerability is one thing and predicting its occurrence in the due course of time is another thing. If we get to know the vulnerability of any software in the due course of time then it acts as an active alarm for the developers to again develop sound and improvised software the second time. The proposal talks about the implementation of the idea using the artificial neural network, where different data sets are being given as input for being used for further analysis for successful results. As of now, there are models for studying the vulnerabilities in the software and networks, this paper proposal in addition to the current work, will throw light on the predictability of vulnerabilities over the due course of time.

Fuzz testing is a popular program testing technique. However, it is difficult to find hard-to-reach vulnerabilities that are nested with complex branches. In this paper, we propose semi-symbolic fuzz testing to discover hard-to-reach vulnerabilities. Our method groups inputs into high frequency and low frequency ones. Then symbolic execution is utilized to solve only uncovered branches to mitigate the path explosion problem. Especially, in order to play the advantages of fuzz testing, our method locates critical branch for each low frequency input and corrects the generated test cases to comfort the branch condition. We also implemented a prototype\textbackslashtextbarS2F, and the experimental results show that S2F can gain 17.70% coverage performance and discover more hard-to-reach vulnerabilities than other vulnerability detection tools for our benchmark.

Vulnerability detection is an import issue in information system security. In this work, we propose the deep learning method for vulnerability detection. We present three deep learning models, namely, convolution neural network (CNN), long short term memory (LSTM) and convolution neural network — long short term memory (CNN-LSTM). In order to test the performance of our approach, we collected 9872 sequences of function calls as features to represent the patterns of binary programs during their execution. We apply our deep learning models to predict the vulnerabilities of these binary programs based on the collected data. The experimental results show that the prediction accuracy of our proposed method reaches 83.6%, which is superior to that of traditional method like multi-layer perceptron (MLP).

A high definition(HD) wide dynamic video surveillance system is designed and implemented based on Field Programmable Gate Array(FPGA). This system is composed of three subsystems, which are video capture, video wide dynamic processing and video display subsystem. The images in the video are captured directly through the camera that is configured in a pattern have long exposure in odd frames and short exposure in even frames. The video data stream is buffered in DDR2 SDRAM to obtain two adjacent frames. Later, the image data fusion is completed by fusing the long exposure image with the short exposure image (pixel by pixel). The video image display subsystem can display the image through a HDMI interface. The system is designed on the platform of Lattice ECP3-70EA FPGA, and camera is the Panasonic MN34229 sensor. The experimental result shows that this system can expand dynamic range of the HD video with 30 frames per second and a resolution equal to 1920*1080 pixels by real-time wide dynamic range (WDR) video processing, and has a high practical value.

Block recursive least square (BRLS) algorithm for dictionary learning in compressed sensing system is developed for surveillance video processing. The new method uses image blocks directly and iteratively to train dictionaries via BRLS algorithm, which is different from classical methods that require to transform blocks to columns first and then giving all training blocks at one time. Since the background in surveillance video is almost fixed, the residual of foreground can be represented sparsely and reconstructed with background subtraction directly. The new method and framework are applied in real image and surveillance video processing. Simulation results show that the new method achieves better representation performance than classical ones in both image and surveillance video.

The world is witnessing a remarkable increase in the usage of video surveillance systems. Besides fulfilling an imperative security and safety purpose, it also contributes towards operations monitoring, hazard detection and facility management in industry/smart factory settings. Most existing surveillance techniques use hand-crafted features analyzed using standard machine learning pipelines for action recognition and event detection. A key shortcoming of such techniques is the inability to learn from unlabeled video streams. The entire video stream is unlabeled when the requirement is to detect irregular, unforeseen and abnormal behaviors, anomalies. Recent developments in intelligent high-level video analysis have been successful in identifying individual elements in a video frame. However, the detection of anomalies in an entire video feed requires incremental and unsupervised machine learning. This paper presents a novel approach that incorporates high-level video analysis outcomes with incremental knowledge acquisition and self-learning for autonomous video surveillance. The proposed approach is capable of detecting changes that occur over time and separating irregularities from re-occurrences, without the prerequisite of a labeled dataset. We demonstrate the proposed approach using a benchmark video dataset and the results confirm its validity and usability for autonomous video surveillance.

In video surveillance, face recognition (FR) systems seek to detect individuals of interest appearing over a distributed network of cameras. Still-to-video FR systems match faces captured in videos under challenging conditions against facial models, often designed using one reference still per individual. Although CNNs can achieve among the highest levels of accuracy in many real-world FR applications, state-of-the-art CNNs that are suitable for still-to-video FR, like trunk-branch ensemble (TBE) CNNs, represent complex solutions for real-time applications. In this paper, an efficient CNN architecture is proposed for accurate still-to-video FR from a single reference still. The CCM-CNN is based on new cross-correlation matching (CCM) and triplet-loss optimization methods that provide discriminant face representations. The matching pipeline exploits a matrix Hadamard product followed by a fully connected layer inspired by adaptive weighted cross-correlation. A triplet-based training approach is proposed to optimize the CCM-CNN parameters such that the inter-class variations are increased, while enhancing robustness to intra-class variations. To further improve robustness, the network is fine-tuned using synthetically-generated faces based on still and videos of non-target individuals. Experiments on videos from the COX Face and Chokepoint datasets indicate that the CCM-CNN can achieve a high level of accuracy that is comparable to TBE-CNN and HaarNet, but with a significantly lower time and memory complexity. It may therefore represent the better trade-off between accuracy and complexity for real-time video surveillance applications.

With crimes on the rise all around the world, video surveillance is becoming more important day by day. Due to the lack of human resources to monitor this increasing number of cameras manually, new computer vision algorithms to perform lower and higher level tasks are being developed. We have developed a new method incorporating the most acclaimed Histograms of Oriented Gradients, the theory of Visual Saliency and the saliency prediction model Deep Multi-Level Network to detect human beings in video sequences. Furthermore, we implemented the k - Means algorithm to cluster the HOG feature vectors of the positively detected windows and determined the path followed by a person in the video. We achieved a detection precision of 83.11% and a recall of 41.27%. We obtained these results 76.866 times faster than classification on normal images.

We present an object tracking framework which fuses multiple unstable video-based methods and supports automatic tracker initialization and termination. To evaluate our system, we collected a large dataset of hand-annotated 5-minute traffic surveillance videos, which we are releasing to the community. To the best of our knowledge, this is the first publicly available dataset of such long videos, providing a diverse range of real-world object variation, scale change, interaction, different resolutions and illumination conditions. In our comprehensive evaluation using this dataset, we show that our automatic object tracking system often outperforms state-of-the-art trackers, even when these are provided with proper manual initialization. We also demonstrate tracking throughput improvements of 5× or more vs. the competition.

This work introduces concepts and algorithms along with a case study validating them, to enhance the event detection, pattern recognition and anomaly identification results in real life video surveillance. The motivation for the work underlies in the observation that human behavioral patterns in general continuously evolve and adapt with time, rather than being static. First, limitations in existing work with respect to this phenomena are identified. Accordingly, the notion and algorithms of Dynamic Clustering are introduced in order to overcome these drawbacks. Correspondingly, we propose the concept of maintaining two separate sets of data in parallel, namely the Normal Plane and the Anomaly Plane, to successfully achieve the task of learning continuously. The practicability of the proposed algorithms in a real life scenario is demonstrated through a case study. From the analysis presented in this work, it is evident that a more comprehensive analysis, closely following human perception can be accomplished by incorporating the proposed notions and algorithms in a video surveillance event.

Due to the transition from analog to digital format, it possible to use IP-protocol for video surveillance systems. In addition, wireless access, color systems with higher resolution, biometrics, intelligent sensors, software for performing video analytics are becoming increasingly widespread. The paper considers only the calculation of the error probability (BER — Bit Error Rate) depending on the realized value of S/N.

Surveillance video systems are gaining increasing attention in the field of computer vision due to its demands of users for the seek of security. It is promising to observe the human movement and predict such kind of sense of movements. The need arises to develop a surveillance system that capable to overcome the shortcoming of depending on the human resource to stay monitoring, observing the normal and suspect event all the time without any absent mind and to facilitate the control of huge surveillance system network. In this paper, an intelligent human activity system recognition is developed. Series of digital image processing techniques were used in each stage of the proposed system, such as background subtraction, binarization, and morphological operation. A robust neural network was built based on the human activities features database, which was extracted from the frame sequences. Multi-layer feed forward perceptron network used to classify the activities model in the dataset. The classification results show a high performance in all of the stages of training, testing and validation. Finally, these results lead to achieving a promising performance in the activity recognition rate.

Detecting faces and heads appearing in video feeds are challenging tasks in real-world video surveillance applications due to variations in appearance, occlusions and complex backgrounds. Recently, several CNN architectures have been proposed to increase the accuracy of detectors, although their computational complexity can be an issue, especially for realtime applications, where faces and heads must be detected live using high-resolution cameras. This paper compares the accuracy and complexity of state-of-the-art CNN architectures that are suitable for face and head detection. Single pass and region-based architectures are reviewed and compared empirically to baseline techniques according to accuracy and to time and memory complexity on images from several challenging datasets. The viability of these architectures is analyzed with real-time video surveillance applications in mind. Results suggest that, although CNN architectures can achieve a very high level of accuracy compared to traditional detectors, their computational cost can represent a limitation for many practical real-time applications.

Sensors incorporated into mobile devices provide unique opportunities to capture detailed environmental information that cannot be readily collected in other ways. We show here how data from networked navigational sensors on leisure vessels can be used to construct unique new datasets, using the example of underwater topography (bathymetry) to demonstrate the approach. Specifically, we describe an end-to-end workflow that involves the collection of large numbers of timestamped (position, depth) measurements from "internet of floating things" devices on leisure vessels; the communication of data to cloud resources, via a specialized protocol capable of dealing with delayed, intermittent, or even disconnected networks; the integration of measurement data into cloud storage; the efficient correction and interpolation of measurements on a cloud computing platform; and the creation of a continuously updated bathymetric database. Our prototype implementation of this workflow leverages the FACE-IT Galaxy workflow engine to integrate network communication and database components with a CUDA-enabled algorithm running in a virtualized cloud environment.

Underwater acoustic communications experiments often involve custom implementations of schemes and protocols for the physical and data link layers. However, most commercial modems focus on providing reliable or optimized communication links, rather than on allowing low-level reconfiguration or reprogramming of modulation and coding schemes. As a result, the physical layer is typically provided as a closed, non-reprogrammable black box, accessible by the user only through a specific interface. While software-defined modems would be the ultimate solution to overcome this issue, having access to the symbols transmitted by the modems using a proprietary modulation format already opens up a number of research opportunities, e.g., aimed at the cross-layer design and optimization of channel coding schemes and communication protocols. In this paper, we take the latter approach. We consider the commercial EvoLogics modem, driven by a custom firmware version that bypasses the channel coding methods applied by the modem, and allows the user to set the transmit bit rate to any desired value within a given set. This makes it possible to evaluate different coding schemes in the presence of different bit rates. Our results show that the custom firmware offers sufficient flexibility to test different configurations of the coding schemes and bit rates, by providing direct access both to correctly decoded and to corrupted symbols, which can be separated at the receiver for further processing. In addition, we show that the DESERT Underwater framework can also leverage the same flexibility by employing low-level physical layer access in more complex networking experiments.

Multi-modal communication methods have been proposed for underwater wireless networks (UWNs) to tackle the challenging physical characteristics of underwater wireless channels. These include the use of acoustic and optic technology for range-dependent transmissions. Software-defined networking (SDN) is an appealing choice for managing these networks with multi-modal communication capabilities, allowing for increased adaptability in the UWN design. In this work, we develop a simulation platform for software-defined underwater wireless networks (SDUWNs). Similarto OpenNet, this platform integrates Mininet with ns-3 via TapBridge modules. The multi-modal communication is implemented by equipping each ns-3 node with multiple net devices. Multiple channel modules connecting corresponding net devices are configured to reflect the channel characteristics. The proposed simulation platform is validated in a case study for oceanographic data collection.

This paper considers the use of Combined Free/Demand Assignment Multiple Access (CFDAMA) for Underwater Acoustic Networks (UANs). The long propagation delay places severe constraints on the trade-off between end-to-end delay and the achievable channel utilisation. Free assignment is shown to offer close to the theoretical minimum end-to-end delay at low channel loads. Demand assignment is shown to have a much greater tolerance to increasing channel load over virtually the entire channel utilisation range, but with longer delay. CFDAMA is shown to exhibit significantly enhanced performance with respect to minimising end-to-end delay and maximising channel utilisation.

In this paper, the mathematical framework of behavioral system will be applied to detect the cyber-attack on the networked control system which is used to control the remotely operated underwater vehicle ROV. The Intelligent Generalized Predictive Controller IGPC is used to control the ROV. The IGPC is designed with fault-tolerant ability. In consequence of the used fault accommodation technique, the proposed cyber-attacks detector is able to clearly detect the presence of attacker control signal and to distinguish between the effects of the attacker signal and fault on the plant side. The test result of the suggested method demonstrates that it can be considerably used for detection of the cyber-attack.

Real-time localization of mobile target has been attracted much attention in recent years. With the limitation of unavailable GPS signals in the complex environments, wireless sensor networks can be applied to real-time locate and track the mobile targets in this paper. The multi wireless signals are used to weaken the effect of abnormal wireless signals in some areas. To verify the real-time localization performance for mobile targets, experiments and analyses are implemented. The results of the experiments reflect that the proposed location method can provide experimental basis for the applications, such as the garage, shopping center, underwater, etc.

This paper combines FMEA and n2 approaches in order to create a methodology to determine risks associated with the components of an underwater system. This methodology is based on defining the risk level related to each one of the components and interfaces that belong to a complex underwater system. As far as the authors know, this approach has not been reported before. The resulting information from the mentioned procedures is combined to find the system's critical elements and interfaces that are most affected by each failure mode. Finally, a calculation is performed to determine the severity level of each failure mode based on the system's critical elements.

On account of large and inconsistent propagation delays during transmission in Underwater Wireless Sensor Networks (UWSNs), wormholes bring more destructive than many attacks to localization applications. As a localization algorithm, DV-hop is classic but without secure scheme. A secure localization algorithm for UWSNs- RDV-HOP is brought out, which is based on reputation values and the constraints of propagation distance in UWSNs. In RDV-HOP, the anchor nodes evaluate the reputation of paths to other anchor nodes and broadcast these reputation values to the network. Unknown nodes select credible anchors nodes with high reputation to locate. We analyze the influence of the location accuracy with some parameters in the simulation experiments. The results show that the proposed algorithm can reduce the location error under the wormhole attack.

In this paper, we propose a theoretical framework to investigate the eavesdropping behavior in underwater acoustic sensor networks. In particular, we quantify the eavesdropping activities by the eavesdropping probability. Our derived results show that the eavesdropping probability heavily depends on acoustic signal frequency, underwater acoustic channel characteristics (such as spreading factor and wind speed) and different hydrophones (such as isotropic hydrophones and array hydrophones). Simulation results have further validate the effectiveness and the accuracy of our proposed model.

The widespread diffusion of the Internet of Things (IoT) is introducing a huge number of Internet-connected devices in our daily life. Mainly, wearable devices are going to have a large impact on our lifestyle, especially in a healthcare scenario. In this framework, it is fundamental to secure exchanged information between these devices. Among other factors, it is important to take into account the link between a wearable device and a smart unit (e.g., smartphone). This connection is generally obtained via specific wireless protocols such as Bluetooth Low Energy (BLE): the main topic of this work is to analyse the security of this communication link. In this paper we expose, via an experimental campaign, a methodology to perform a vulnerability assessment (VA) on wearable devices communicating with a smartphone. In this way, we identify several security issues in a set of commercial wearable devices.

Wearable medical devices are playing more and more important roles in healthcare. Unlike the wired connection, the wireless connection between wearable devices and the remote servers are exceptionally vulnerable to malicious attacks, and poses threats to the safety and privacy of the patient health data. Therefore, wearable medical devices require the implementation of reliable measures to secure the wireless network communication. However, those devices usually have limited computational power that is not comparable with the desktop computer and thus, it is difficult to adopt the full-fledged security algorithm in software. In this study, we have developed an efficient authentication and encryption protocol for internetconnected wearable devices using the recognized standards of AES and SHA that can provide two-way authentication between wearable device and remote server and protection of patient privacy against various network threats. We have tested the feasibility of this protocol on the TI CC3200 Launchpad, an evaluation board of the CC3200, which is a Wi-Fi capable microcontroller designed for wearable devices and includes a hardware accelerated cryptography module for the implementation of the encryption algorithm. The microcontroller serves as the wearable device client and a Linux computer serves as the server. The embedded client software was written in ANSI C and the server software was written in Python.