Search

Graphs are ubiquitous for modeling complex systems involving structured data and relationships. Consequently, graph representation learning, which aims to automatically learn low-dimensional representations of graphs, has drawn a lot of attention in recent years. The overwhelming majority of existing methods handle unsigned graphs. However, signed graphs appear in an increasing number of application domains to model systems involving two types of opposed relationships. Several authors took an interest in signed graphs and proposed methods for providing vertex-level representations, but only one exists for whole-graph representations, and it can handle only fully connected graphs. In this article, we tackle this issue by proposing two approaches to learning whole-graph representations of general signed graphs. The first is a SG2V, a signed generalization of the whole-graph embedding method Graph2vec that relies on a modification of the Weisfeiler-Lehman relabelling procedure. The second one is WSGCN, a whole-graph generalization of the signed vertex embedding method SGCN that relies on the introduction of master nodes into the GCN. We propose several variants of both these approaches. A bottleneck in the development of whole-graph-oriented methods is the lack of data. We constitute a benchmark composed of three collections of signed graphs with corresponding ground truths. We assess our methods on this benchmark, and our results show that the signed whole-graph methods learn better representations for this task. Overall, the baseline obtains an F-measure score of 58.57, when SG2V and WSGCN reach 73.01 and 81.20, respectively. Our source code and benchmark are publicly available online.

Refactoring is one of the techniques mostly employed by software developers to improve the quality attributes of their systems. However, little has been done to investigate how refactoring operations specifically aimed at improving the internal structure of software can impact its security. Refactoring usually entails different code change operations including the decomposition of classes, methods, and the reallocation of code elements. While this refinement aims to improve the internal design of a system, it might inadvertently disperse security-critical code elements throughout the codebase. Consequently, such dispersion could affect the software attack surface. To this end, this paper presents an empirical study conducted on 30 open-source software systems that were developed in Python, C, and Javascript. The study scrutinized two subsequent versions of each subject application to uncover the refactoring operations applied and the trend of the software attack surface. Specifically, the study focused on the injection or removal of bugs, code smells and other vulnerabilities aiming to discern the impact of refactorings on the software attack surface. Data was collected using well-known tools, namely SonarQube, RefDiff, and PyReff. The findings suggest that refactorings can have multiple impacts (i.e., positive, negative, or neutral) on bugs, code smells, and vulnerabilities. The findings further confirm that developers must be aware of the combination or sequence of refactoring operations that can improve software quality without compromising its security.

Slow-read Distributed Denial of Service (DDoS) attacks are complex to detect and mitigate. Although existing tools allow one to identify these attacks, these tools mainly generate alerts. However, in real scenarios, a large number of attack detection alerts will put the security workforce in a bottleneck, as they will not be able to implement mitigation actions in a complete and timely manner. Furthermore, since most existing security solutions for DDoS attack mitigation are tested using datasets and simulated scenarios, their applicability to production networks could be unfeasible or ineffective due to possibly incomplete assumptions in their design. Therefore, automated security solutions against DDoS attacks are needed not only to be designed but also to be implemented and evaluated in real scenarios. This study presents a Software-Defined Networking (SDN)-based security framework, which automates the monitoring, detection, and mitigation of slow-rate DDoS attacks. The framework is implemented in a physical network that uses equipment from the European Experimental Facility Smart Networks for Industry (SN4I). The results demonstrate that the framework effectively mitigates malicious connections, with a mitigation efficiency between 91.66%– 100% for different conditions of the number of attackers and victims. In addition, the SDN-SlowRate-DDoS dataset is presented, which contains multiple experiments of slow-rate DDoS attacks performed on the real testbed. The resources provided in this security dataset are useful to the scientific and industry communities in designing and testing realistic solutions for intrusion detection systems.

Electronic Government (e-Government) systems constantly provide greater services to people, businesses, organisations, and societies by offering more information, opportunities, and platforms with the support of advances in information and communications technologies. This usually results in increased system complexity and sensitivity, necessitating stricter security and privacy-protection measures. The majority of the existing e-Government systems are centralised, making them vulnerable to privacy and security threats, in addition to suffering from a single point of failure. This study proposes a decentralised e-Government framework with integrated threat detection features to address the aforementioned challenges. In particular, the privacy and security of the proposed e-Government system are realised by the encryption, validation, and immutable mechanisms provided by Blockchain. The insider and external threats associated with blockchain transactions are minimised by the employment of an artificial immune system, which effectively protects the integrity of the Blockchain. The proposed e-Government system was validated and evaluated by using the framework of Ethereum Visualisations of Interactive, Blockchain, Extended Simulations (i.e. eVIBES simulator) with two publicly available datasets. The experimental results show the efficacy of the proposed framework in that it can mitigate insider and external threats in e-Government systems whilst simultaneously preserving the privacy of information.

Wireless communication systems working in millimeter-wave (mmWave) frequency bands offer higher bandwidths than traditional radio frequency schemes. This technology allows multibeam steering and data multiplexing with the help of massive multiple-input multiple-output (MIMO) systems. However, supporting large bandwidths at mmWave frequencies is challenging due to the use of large antenna arrays with beamforming, sampling signals with large bandwidths, and baseband signal processing operations at gigabit data rates. Due to the wider bandwidth and higher signal processing requirements of mmWave systems, low-complexity receiver algorithms become important. Previously reported investigations assumed the use of hybrid beamforming structures that reduce power consumption and signal processing tasks. Therefore, the use of artificial neural networks (ANNs) becomes relevant for the processing of mmWave signals as reported in earlier works. In this article, to carry out MIMO combining processing for mmWave communications, we propose a fully complex multilayer extreme learning machine (M-ELM) neural network. We investigate the tuning of the number of neurons in each hidden layer for the proposed method to maximize the system performance and decrease the complexity of the receiver. We compare the results of the introduced M-ELM algorithm with a fully complex extreme learning machine (ELM), fully real ELM, and M-ELM defined in the real plane in terms of spectral efficiency, bit error rate, computational complexity, and processing time. Furthermore, we compare the novel M-ELM strategy with traditional linear MIMO receivers, such as Maximum Ratio and Minimum Mean Square Error, as well as to a multilayer perceptron (MLP) neural network trained offline. The numerical results show that with a good balance between the overall performance and computational cost of the ANN, the fully complex M-ELM MIMO receiver outperforms the other evaluated schemes.

Linear topologies arise naturally in the context of Internet-of-Things (IoT) applications for smart cities, where the infrastructure itself commonly has a linear or semi-linear structure. This is the case of buildings, public transportation systems, road infrastructure, and utility distribution networks. Given the prevalence of this type of topologies, several Medium Access Control (MAC) protocols have been designed to take advantage of their particular properties. Unfortunately, most of them do not scale well as the node density and the distance in hops to the sink increases. The result is that packets generated many hops away from the sink tend to experience unacceptable high end-to-end delay and low delivery probabilities. This paper introduces HP-MAC, a synchronized duty-cycled MAC protocol for Linear Sensor Networks (LSNs) that assigns transmission priorities to nodes to avoid collisions, through the implementation of distributed elections based on hash functions. HP-MAC also implements a packet queuing scheme that acts as a mechanism to control the amount of network resources allocated to data flows generated at different distances to the sink. This way, packets can reach their destination with loss probability and end-to-end delay that do not depend on their distance to the sink. We use a Discrete-Time Markov Chain (DTMC) to model the performance of the proposed protocol. Numerical solutions of this model show that HP-MAC outperforms state-of-the-art representatives in terms of throughput, end-to-end delay, power consumption, and packet loss probability. These results are validated through extensive discrete-event simulations.

In this paper, we compare two artificial neural networks (ANNs) approaches designed to perform channel equalization for millimeter-wave (mmWave) signals operating in the 28 GHz frequency band. We used an in-house deterministic Three-Dimensional Ray-Launching (3D-RL) code to simulate the spatial structure of mmWave channels considering the material properties of the obstacles within the scenario at the frequency under analysis. We performed offline training of a multilayer perceptron (MLP) neural network with the simulated mmWave channels to equalize the received signal. We also performed online training of an extreme learning machine (ELM) neural network to directly get the equalized symbols at the receiver, given as input the received mmWave signal. The ANN solutions were tested in terms of the achievable spectral efficiency, bit error rate, and time to process. We compared the ANN techniques to the minimum mean square error and the zero-forcing equalizers, considering an orthogonal frequency-division multiplexing communication based on the 5G New Radio standard. We present numerical results on the performance of the proposed ANNs and show that the ELM strategy outperforms the MLP method, requiring significantly less processing time than the reviewed equalization methods.

Autonomous robot visual navigation is a fundamental locomotion task based on extracting relevant features from images taken from the surrounded environment to control an independent displacement. In the navigation, the use of a known visual map helps obtain an accurate localization, but in the absence of this map, a guided or free exploration pathway must be executed to obtain the images sequence representing the visual map. This paper presents an appearance-based localization method based on a visual map and an end-to-end Convolutional Neural Network (CNN). The CNN is initialized via transfer learning (trained using the ImageNet dataset), evaluating four state-of-the-art CNN architectures: VGG16, ResNet50, InceptionV3, and Xception. A typical pipeline for transfer learning includes changing the last layer to adapt the number of neurons according to the number of custom classes. In this work, the dense layers after the convolutional and pooling layers were substituted by a Global Average Pooling (GAP) layer, which is parameter-free. Additionally, an L2 -norm constraint was added to the GAP layer feature descriptors, restricting the features from lying on a fixed radius hypersphere. These different pre-trained configurations were analyzed and compared using two visual maps found in the CIMAT-NAO datasets consisting of 187 and 94 images, respectively. For evaluating the localization tasks, a set of 278 and 94 images were available for each visual map, respectively. The numerical results proved that by integrating the L2-norm constraint in the training pipeline, the appearance-based localization performance is boosted. Specifically, the pre-trained VGG16 and Xception networks achieved the best localization results, reaching a top-3 accuracy of 90.70% and 93.62% for each dataset, respectively, overcoming the referenced approaches based on hand-crafted feature extractors.

In this paper, a constant speed control of slider-crank mechanisms for machine tools is proposed. A joint-task space hybrid controller based on a second-order sliding mode control and time-base generator was used to guarantee a constant speed trajectory tracking and a complete turn of the mechanism crank. A switching criterion was implemented in order to avoid the singularities located at the two extreme positions of the slider stroke. A trapezoidal speed profile with parabolic blends was designed directly over task space slider trajectory considering a constant cutting speed, the workpiece dimensions and the slider stroke length. Stability of the second-order sliding mode control was validated with the Lyapunov stability theory. Simulations were carried out to verify this approach.

The fire influence on regional to global environments and air quality (FIREX-AQ) field campaign was conducted during August 2019 to investigate the impact of wildfire and biomass smoke on air quality and weather in the continental United States. One of the campaign's scientific objectives was to estimate the composition of emissions from wildfires. Ultraspectrally resolved infrared radiance measurements from aircraft and/or satellite observations contain information on tropospheric carbon monoxide (CO) as well as other trace species present in fire emissions. A methodology for retrieving tropospheric CO from such remotely sensed spectral data has been developed for the National Airborne Sounder Testbed-Interferometer (NAST-I) and is applied herein. Retrievals based on NAST-I measurements are used to demonstrate CO retrieval capability and characterize fire emissions. NAST-I remotely sensed CO from ER-2 flights are evaluated with concurrent in situ measurements from the differential absorption carbon monoxide measurements flown on the NASA DC-8 aircraft. Enhanced CO emissions along with plume evolution and transport from the fire ground site locations were captured by moderate vertical and high horizontal resolution observations obtained from the NAST-I IR spectrometer; these were intercompared and verified by the cloud physics lidar and the enhanced MODIS airborne simulator also hosted on the NASA ER-2 aircraft. This study will be beneficial to the science community for studying wildfire-related topics and understanding similar remotely sensed observations from satellites, along with helping to address the broader FIREX-AQ experiment objectives of investigating the impact of fires on air quality and climate.

Distributed Denial of Service (DDoS) attacks represent the most common and critical attacks targeting conventional and new generation networks, such as the Internet of Things (IoT), cloud computing, and fifth-generation (5G) communication networks. In recent years, DDoS attacks have become not only massive but also sophisticated. Software-Defined Networking (SDN) technology has demonstrated effectiveness in counter-measuring complex attacks since it provides flexibility on global network monitoring and inline network configuration. Although several works have proposed to detect DDoS attacks, most of them did not use up-to-date datasets that contain the newest threats. Furthermore, only a few previous works assessed their solutions using simulated scenarios, easing the migration to production networks. This document presents the implementation of a modular and flexible SDN-based architecture to detect transport and application layer DDoS attacks using multiple Machine Learning (ML) and Deep Learning (DL) models. Exploring diverse ML/DL methods allowed us to resolve which methods perform better under different attack types and conditions. We tested the ML/DL models using two up-to-date security datasets, namely CICDoS2017 and CICDDoS2019 datasets, and they showed accuracy above 99% on classifying unseen traffic (testing set). We also deployed a simulated environment using the network emulator Mininet and the Open Network Operating System (ONOS) SDN controller. In this experimental setup, we demonstrated high detection rates, above 98% for transport DDoS attacks and up to 95% for application-layer DDoS attacks.

In this work, a position control in task space for slider-crank mechanisms is presented. In order to apply linear controllers it is required to linearize the mechanism dynamics at an equilibrium point. However, complete dynamic knowledge is needed and the linearization technique gives an oversimplified model that affects the control performance. In this work, it is proposed a novel method to design task space controllers without using the complete knowledge of the mechanism dynamics and linearization methods. From the extended dynamic model of parallel robots, it can be seen that the end-effector (slider) dynamics is expressed as a linear system that can be used directly for the control design instead of the complete mechanism linear dynamics. The approach requires a minimal knowledge of the mechanism dynamics and avoids linearization methods. To verify our approach, it is used pole placement and sliding mode controllers whose gains are tuned according to the slider dynamics. A linear sensor is mounted at the slider to measure its position and avoids considering noise and disturbances at links before the slider. Simulations and experiments are presented to validate our approach using two kinds of slider-crank mechanisms.