Biblio

The emerging Internet of Things needs edge-computing - this is an established fact. In turn, edge computing needs infrastructure decentralisation. What is not necessarily established yet is that infrastructure decentralisation needs a distributed model of Internet governance and decentralised trust schemes. We discuss the features of a decentralised IoT and edge-computing ecosystem and list the components that need to be designed, as well the challenges that need to be addressed.

Logistic regression (LR) is used in many areas due to its simplicity and interpretability. While at the same time, those two properties limit its classification accuracy. Deep neural networks (DNNs), instead, achieve state-of-the-art performance in many domains. However, the nonlinearity and complexity of DNNs make it less interpretable. To balance interpretability and classification performance, we propose a novel nonlinear model, Deep Embedding Logistic Regression (DELR), which augments LR with a nonlinear dimension-wise feature embedding. In DELR, each feature embedding is learned through a deep and narrow neural network and LR is attached to decide feature importance. A compact and yet powerful model, DELR offers great interpretability: it can tell the importance of each input feature, yield meaningful embedding of categorical features, and extract actionable changes, making it attractive for tasks such as market analysis and clinical prediction.

The aim of this investigation is to classify real-life facial images into one of six types of emotions. For solving this problem, we propose to use deep machine learning algorithms and convolutional neural network (CNN). CNN is a modern type of neural network, which allows for rapid detection of various objects, as well as to make an effective object classification. For acceleration of CNN learning stage, we use supercomputer NVIDIA DGX-1. This process was implemented in parallel on a large number of independent streams on GPU. Numerical experiments for algorithms were performed on the images of Multi-Pie image database with various lighting of scene and angle rotation of head. For developed models, several metrics of quality were calculated. The designing algorithm was used in real-time video processing in human-computer interaction systems. Moreover, expression recognition can apply in such fields as retail analysis, security, video games, animations, psychiatry, automobile safety, educational software, etc.

With the widespread diffusion of powerful media editing tools, falsifying images and videos has become easier and easier in the last few years. Fake multimedia, often used to support fake news, represents a growing menace in many fields of life, notably in politics, journalism, and the judiciary. In response to this threat, the signal processing community has produced a major research effort. A large number of methods have been proposed for source identification, forgery detection and localization, relying on the typical signal processing tools. The advent of deep learning, however, is changing the rules of the game. On one hand, new sophisticated methods based on deep learning have been proposed to accomplish manipulations that were previously unthinkable. On the other hand, deep learning provides also the analyst with new powerful forensic tools. Given a suitably large training set, deep learning architectures ensure usually a significant performance gain with respect to conventional methods, and a much higher robustness to post-processing and evasions. In this talk after reviewing the main approaches proposed in the literature to ensure media authenticity, the most promising solutions relying on Convolutional Neural Networks will be explored with special attention to realistic scenarios, such as when manipulated images and videos are spread out over social networks. In addition, an analysis of the efficacy of adversarial attacks on such methods will be presented.

It is challenging to detect weak hidden information in a JPEG compressed image. In this paper, we propose a 32-layer convolutional neural networks (CNNs) with feature reuse by concatenating all features from previous layers. The proposed method can improve the flow of gradient and information, and the shared features and bottleneck layers in the proposed CNN model further reduce the number of parameters dramatically. The experimental results shown that the proposed method significantly reduce the detection error rate compared with the existing JPEG steganalysis methods, e.g. state-of-the-art XuNet method and the conventional SCA-GFR method. Compared with XuNet method and conventional method SCA-GFR in detecting J-UNIWARD at 0.1 bpnzAC (bit per non-zero AC DCT coefficient), the proposed method can reduce detection error rate by 4.33% and 6.55% respectively.

Vehicular ad hoc networks (VANETs) have become a promising technology in intelligent transportation systems (ITS) with rising interest of expedient, safe, and high-efficient transportation. VANETs are vulnerable to malicious nodes and result in performance degradation because of dynamicity and infrastructure-less. In this paper, we propose a trust based dueling deep reinforcement learning approach (T-DDRL) for communication of connected vehicles, we deploy a dueling network architecture into a logically centralized controller of software-defined networking (SDN). Specifically, the SDN controller is used as an agent to learn the most trusted routing path by deep neural network (DNN) in VANETs, where the trust model is designed to evaluate neighbors' behaviour of forwarding routing information. Simulation results are presented to show the effectiveness of the proposed T-DDRL framework.

With the amount of data has been rapidly growing over recent decades, binary hashing has become an attractive approach for fast search over large databases, in which the high-dimensional data such as image, video or text is mapped into a low-dimensional binary code. Searching in this hamming space is extremely efficient which is independent of the data size. A lot of methods have been proposed to learn this binary mapping. However, to make the binary codes conserves the input information, previous works mostly resort to mean squared error, which is prone to lose a lot of input information [11]. On the other hand, most of the previous works adopt the norm constraint or approximation on the hidden representation to make it as close as possible to binary, but the norm constraint is too strict that harms the expressiveness and flexibility of the code. In this paper, to generate desirable binary codes, we introduce two adversarial training procedures to the hashing process. We replace the L2 reconstruction error with an adversarial training process to make the codes reserve its input information, and we apply another adversarial learning discriminator on the hidden codes to make it proximate to binary. With the adversarial training process, the generated codes are getting close to binary while also conserves the input information. We conduct comprehensive experiments on both supervised and unsupervised hashing applications and achieves a new state of the arts result on many image hashing benchmarks.

This paper presents DeepSecure, the an scalable and provably secure Deep Learning (DL) framework that is built upon automated design, efficient logic synthesis, and optimization methodologies. DeepSecure targets scenarios in which neither of the involved parties including the cloud servers that hold the DL model parameters or the delegating clients who own the data is willing to reveal their information. Our framework is the first to empower accurate and scalable DL analysis of data generated by distributed clients without sacrificing the security to maintain efficiency. The secure DL computation in DeepSecure is performed using Yao's Garbled Circuit (GC) protocol. We devise GC-optimized realization of various components used in DL. Our optimized implementation achieves up to 58-fold higher throughput per sample compared with the best prior solution. In addition to the optimized GC realization, we introduce a set of novel low-overhead pre-processing techniques which further reduce the GC overall runtime in the context of DL. Our extensive evaluations demonstrate up to two orders-of-magnitude additional runtime improvement achieved as a result of our pre-processing methodology.

Software testing is very important for software quality assurance. However, the test activity is not a simple task and requires good planning to be successful. It is in this context that the automation of tests gains importance. This paper presents the experience of defining and implementing a test automation strategy for functional tests based on the Brazilian Test Process Improvement Model (MPT.Br) in an IT company. The results of this work include the improvement of the testing process used by the company, the increase in the test coverage and the reduction of time used to perform regression tests.

The rise of Big Data is leading to an increasing demand for large-scale data-intensive applications (DIAs), which have to analyse massive amounts of personal data (e.g. customers' location, cars' speed, people heartbeat, etc.), some of which can be sensitive, meaning that its confidentiality has to be protected. In this context, DIA providers are responsible for enforcing privacy policies that account for the privacy preferences of data subjects as well as for general privacy regulations. This is the case, for instance, of data brokers, i.e. companies that continuously collect and analyse data in order to provide useful analytics to their clients. Unfortunately, the enforcement of privacy policies in modern DIAs tends to become cumbersome because (i) the number of policies can easily explode, depending on the number of data subjects, (ii) policy enforcement has to autonomously adapt to the application context, thus, requiring some non-trivial runtime reasoning, and (iii) designing and developing modern DIAs is complex per se. For the above reasons, we need specific design and runtime methods enabling so called privacy-by-design in a Big Data context. In this article we propose an approach for specifying, enforcing and checking privacy policies on DIAs designed according to the Google Dataflow model and we show that the enforcement approach behaves correctly in the considered cases and introduces a performance overhead that is acceptable given the requirements of a typical DIA.

Many recent advances of network caching focus on i) more effectively modeling the preferences of a regional user group to different web contents, and ii) reducing the cost of content delivery by storing the most popular contents in regional caches. However, the context under which the users interact with the network system usually causes tremendous variations in a user group's preferences on the contents. To effectively leverage such contextual information for more efficient network caching, we propose a novel mechanism to incorporate context-aware collaborative filtering into demand-driven caching. By differentiating the characterization of user interests based on a priori contexts, our approach seeks to enhance the cache performance with a more dynamic and fine-grained cache allocation process. In particular, our approach is general and adapts to various types of context information. Our evaluation shows that this new approach significantly outperforms previous non-demand-driven caching strategies by offering much higher cached content rate, especially when utilizing the contextual information.

Synchrophasor technology is used for real-time control and monitoring in smart grid. Previous works in literature identified critical vulnerabilities in IEEE C37.118.2 synchrophasor communication standard. To protect synchrophasor-based systems, stealthy cyber-attacks and effective defense mechanisms still need to be investigated.This paper investigates how an attacker can develop a custom tool to execute stealthy man-in-the-middle attacks against synchrophasor devices. In particular, four different types of attack capabilities have been demonstrated in a real synchrophasor-based synchronous islanding testbed in laboratory: (i) command injection attack, (ii) packet drop attack, (iii) replay attack and (iv) stealthy data manipulation attack. With deep technical understanding of the attack capabilities and potential physical impacts, this paper also develops and tests a distributed Intrusion Detection System (IDS) following NIST recommendations. The functionalities of the proposed IDS have been validated in the testbed for detecting aforementioned cyber-attacks. The paper identified that a distributed IDS with decentralized decision making capability and the ability to learn system behavior could effectively detect stealthy malicious activities and improve synchrophasor network security.

In recent years, service providers, such as mobile operators providing wireless services, collected location data in enormous extent with the increase of the usages of mobile phones. Vertical businesses, such as banks, may want to use this location information for their own scenarios. However, service providers cannot directly provide these private data to the vertical businesses because of the privacy and legal issues. In this demo, we show how privacy preserving solutions can be utilized using such location-based queries without revealing each organization's sensitive data. In our demonstration, we used partially homomorphic cryptosystem in our protocols and showed practicality and feasibility of our proposed solution.

Data lineage is a fundamental type of information that describes the relationships between input and output data items in a workflow. As such, an immense amount of data-intensive applications with logic over the input-output relationships can be expressed declaratively in lineage terms. Unfortunately, many applications resort to hand-tuned implementations because either lineage systems are not fast enough to meet their requirements or due to no knowledge of the lineage capabilities. Recently, we introduced a set of implementation design principles and associated techniques to optimize lineage-enabled database engines and realized them in our prototype database engine, namely, Smoke. In this demonstration, we showcase lineage as the building block across a variety of data-intensive applications, including tooltips and details on demand; crossfilter; and data profiling. In addition, we show how Smoke outperforms alternative lineage systems to meet or improve on existing hand-tuned implementations of these applications.

We present a new type of attack termed denial of engineering operations in which an attacker can interfere with the normal cycle of an engineering operation leading to a loss of situational awareness. Specifically, the attacker can deceive the engineering software during attempts to retrieve the ladder logic program from a programmable logic controller (PLC) by manipulating the ladder logic on the PLC, such that the software is unable to process it while the PLC continues to execute it successfully. This attack vector can provide sufficient cover for the attacker»s actual scenario to play out while the owner tries to understand the problem and reestablish positive operational control. To enable the forensic analysis and, eventually, eliminate the threat, we have developed the first decompiler for ladder logic programs. Ladder logic is a graphical programming language for PLCs that control physical processes such as power grid, pipelines, and chemical plants; PLCs are a common target of malicious modifications leading to the compromise of the control behavior (and potentially serious consequences). Our decompiler, Laddis, transforms a low-level representation to its corresponding high-level original representation comprising of graphical symbols and connections. The evaluation of the accuracy of the decompiler on the program of varying complexity demonstrates perfect reconstruction of the original program. We present three new attack scenarios on PLC-deployed ladder logic and demonstrate the effectiveness of the decompiler on these scenarios.

Inspired by the simple, yet effective, method of tweeting gibberish to attract automated social agents (bots), we attempt to create localised honeypots in the South African political context. We produce a series of defined techniques and combine them to generate interactions from users on Twitter. The paper offers two key contributions. Conceptually, an argument is made that honeypots should not be confused for bot detection methods, but are rather methods to capture low-quality users. Secondly, we successfully generate a list of 288 local low quality users active in the political context.

This paper deals with the developing model of a honeynet that depends on the Internet of things (IoT). Due to significant of industrial services, such model helps enhancement of information security detection in industrial domain, the model is designed to detect adversaries whom attempt to attack industrial control systems (ICS) and supervisory control and data acquisition (SCADA) systems. The model consists of hardware and software aspects, designed to focus on ICS services that managed remotely via SCADA systems. In order to prove the work of the model, a few of security tools are used such as Shodan, Nmap and others. These tools have been applied locally inside LAN and globally via internet to get proving results. Ultimately, results contain a list of protocols and ports that represent industry control services. To clarify outputs, it contains tcp/udp ports 623, 102, 1025 and 161 which represent respectively IPMI, S7comm, KAMSTRAP and SNMP services.

The internet-of-things (IoT) consists of embedded devices and their networks of communication as they form decentralized frameworks of ubiquitous computing services. Within such decentralized systems the potential for malicious actors to impact the system is significant, with far-reaching consequences. Hence this work addresses the challenge of providing IoT systems engineers with a framework to elicit privacy and security design considerations, specifically for indoor adaptive smart environments. It introduces a new ambient intelligence indoor adaptive environment framework (CORE) which leverages multiple forms of data, and aims to elicit the privacy and security needs of this representative system. This contributes both a new adaptive IoT framework, but also an approach to systematically derive privacy and security design requirements via a combined and modified OCTAVE-Allegro and Privacy-by-Design methodology. This process also informs the future developments and evaluations of the CORE system, toward engineering more secure and private IoT systems.

Research has observed context factors like occasion and time as influential factors for predicting whether or not to share a location with online friends. In other domains like social networks, personality was also found to play an important role. Furthermore, users are seeking a fine-grained disclosement policy that also allows them to display an obfuscated location, like the center of the current city, to some of their friends. In this paper, we observe which context factors and personality measures can be used to predict the correct privacy level out of seven privacy levels, which include obfuscation levels like center of the street or current city. Our results show that a prediction is possible with a precision 20% better than a constant value. We will give design indications to determine which context factors should be recorded, and how much the precision can be increased if personality and privacy measures are recorded using either a questionnaire or automated text analysis.

Research has observed context factors like occasion and time as influential factors for predicting whether or not to share a location with online friends. In other domains like social networks, personality was also found to play an important role. Furthermore, users are seeking a fine-grained disclosement policy that also allows them to display an obfuscated location, like the center of the current city, to some of their friends. In this paper, we observe which context factors and personality measures can be used to predict the correct privacy level out of seven privacy levels, which include obfuscation levels like center of the street or current city. Our results show that a prediction is possible with a precision 20% better than a constant value. We will give design indications to determine which context factors should be recorded, and how much the precision can be increased if personality and privacy measures are recorded using either a questionnaire or automated text analysis.

Wireless transmission of power has been in research for over a century. Our project aims at transmitting electric power over a distance of room. Various methods using microwaves, lasers, inductive coupling, capacitive coupling and acoustic medium have been used. In our project, we are majorly focusing on acoustic method of transferring power. Previous attempts of transferring power using acoustic methods have employed the usage of ultrasonic sound. In our project, we are using infrasonic sound as a medium to transfer electrical power. For this purpose, we are using suitable transducers and converters to transmit electric power from the 220V AC power supply to a load over a considerable distance. This technology can be used to wirelessly charge various devices more effectively.

The general idea to achieve error detection and correction is to add some extra bit to an original message, in which the receiver can use to check the flexibility of the message which has been delivered, and to recover the noisy data. Turbo code is one of the forward error correction method, which is able to achieve the channel capacity, with nearer Shannon limit, encoding and decoding of text and images are performed. Methods and the working have been explained in this paper. The error has also introduced and detection and correction of errors have been achieved. Transmission will be secure it can secure the information by the theft.

Examining implementations of the 6LoWPAN Internet Standard in major embedded operating systems, we observe that they do not fully interoperate. We find this is due to some inherent design flaws in 6LoWPAN. We propose and demonstrate four principles that can be used to structure protocols for low power devices that encourage interoperability between diverse implementations.

Users often need to authenticate at situated displays in order to, for example, make purchases, access sensitive information, or confirm an identity. However, the exposure of interactions in public spaces introduces a large attack surface (e.g., observation, smudge or thermal attacks). A plethora of authentication models and input modalities that aim at disguising users' input has been presented in the past. However, a comprehensive analysis on the requirements for secure and usable authentication on public displays is still missing. This work presents 13 design considerations suitable to inform practitioners and researchers during the development process of authentication systems for situated displays in public spaces. It draws on a comprehensive analysis of prior literature and subsequent discussion with five experts in the fields of pervasive displays, human-computer-interaction and usable security.

Robots remain the focus of researchers and developers, and now they are moving towards IoT based devices and mobile robots to take advantage of the different sensor enables facilities. A robot is a machine capable of carrying out a complex series of actions automatically, especially one programmable by a computer. A robot can be controlled by a human and can be modified by its functionality at runtime by the operator. From past few decades, researchers are contributing towards Robotics. There is no end of technology, creativity, and innovation. The project is designed to develop a robot using android application for remote operation attached to the wireless camera for monitoring purpose. Surveillance using the camera can help the soldier team to make strategies at run-time. This kind of robot can be helpful for spying purpose in war fields. The android application loaded on mobile devices can connect to the security system and easy to use GUI and visualization of the Warfield. The security system then acts on these commands and responds to the user. The camera and the motion detector are attached to the system for remote surveillance using wireless protocol 802.11, ZigBee and Bluetooth protocols. This robot is having the functionality of mines detection, object detection, GPS used for location and navigation and a gun to fire the enemy at the runtime.