Search

This study introduces a novel dataset for wildfire prediction in Morocco, integrating multisource observations to address the country’s unique geographical and climatic challenges. We compile essential environmental indicators and employ state-of-the-art machine learning (ML) and deep learning (DL) algorithms to predict next-day wildfire occurrences. Our best-performing models achieve an accuracy of up to 90%, significantly improving upon traditional approaches. The key contributions include: 1) A localized dataset tailored to Morocco’s conditions; 2) benchmarking of advanced ML and DL algorithms; and 3) open sharing of the dataset and codebase. We discuss the potential applicability of our methodology to other regions and highlight future research directions. This work advances dataset creation techniques and emphasizes the importance of localized research for effective wildfire management strategies in underrepresented areas.

Federated Learning (FL) is a decentralized machine learning (ML) technique that allows a number of participants to train an ML model collaboratively without having to share their private local datasets with others. When participants are unmanned aerial vehicles (UAVs), UAV-enabled FL would experience heterogeneity due to the majorly skewed (non-independent and identically distributed -IID) collected data. In addition, UAVs may demonstrate unintentional misbehavior in which the latter may fail to send updates to the FL server due, for instance, to UAVs' disconnectivity from the FL system caused by high mobility, unavailability, or battery depletion. Such challenges may significantly affect the convergence of the FL model. A recent way to tackle these challenges is client selection, based on customized criteria that consider UAV computing power and energy consumption. However, most existing client selection schemes neglected the participants' reliability. Indeed, FL can be targeted by poisoning attacks, in which malicious UAVs upload poisonous local models to the FL server, by either providing targeted false predictions for specifically chosen inputs or by compromising the global model's accuracy through tampering with the local model. Hence, we propose in this article a novel client selection scheme that enhances convergence by prioritizing fast UAVs with high-reliability scores, while eliminating malicious UAVs from training. Through experiments, we assess the effectiveness of our scheme in resisting different attack scenarios, in terms of convergence and achieved model accuracy. Finally, we demonstrate the performance superiority of the proposed approach compared to baseline methods.

With the proliferation of service-oriented architectures, the ubiquity of mobile networks, and an expanding spectrum of services available to users, the service selection process has become paramount in designing and implementing contemporary applications. Service selection strategies in mobile networks involve choosing the optimal services from many available options based on QoS (quality of service), bandwidth, and mobility, ultimately augmenting network efficiency and user experience. Nonetheless, particular challenges can impact the efficacy of these strategies. This paper presents a comprehensive survey of service selection strategies in mobile networks, transitioning from traditional approaches to emerging techniques, while also introducing a taxonomy classifying service selection methods based on criteria including mobile devices as service providers, architectural perspectives, computing devices allocation, evaluation metrics, and service selection methods. Additionally, We offer insights into the latest QoS-aware and energy-centric service selection methodologies, while addressing challenges related to battery life and energy efficiency, Quality of Service (QoS), security and privacy, as well as scalability and resource management. Additionally, we discuss prospective trajectories and future direction.

This paper presents a very compact, wideband, and enhanced-gain antenna for 5G applications. A simple single-layer millimeter wave (mm-wave) metamaterial lens (meta-lens) is used to improve the gain, aperture efficiency, and gain bandwidth of a slotted-patch antenna over a wide range of frequencies from 25 GHz to 31 GHz. The lens exhibits a metamaterial negative refractive index behavior, which is attributed to a substantial gain enhancement of around 4–5 dBi over the whole band compared to the gain values of the slotted patch antenna alone. The lens’s unit cell comprises a simple single-layer split ring resonator (SRR) whose dimensions are carefully chosen to improve transmitted power and suppress absorbed and reflected power. The meta-lens consists of $8\times 8$ subwavelength SRR unit cells. Each cell has an area of $1.6\times 1.6$ mm2, it is located in the near-field region closely above a slotted patch antenna to produce a total antenna size of $12.8\times 12.8\times 7.27$ mm3 ( $1.2~\lambda \times 1.2~\lambda \times 0.68~\lambda $ , where $\lambda $ is the free space wavelength at 28 GHz). The maximum gain of the proposed antenna is 12.7 dBi, the 1 dB gain bandwidth is 18%, the maximum aperture efficiency is 92%, and the −10 dB impedance bandwidth (10 dB B.W.) is 17%. This excellent combination of essential metrics is hard to realize at mm-wave using narrowband antenna structures (microstrip patch antennas), and the aperture efficiency is the highest thus far for such a class of antennas.

The brain computer interface is defined as the way of acquiring the brain signals that analyse and translates them into commands that are relayed to intelligent devices for carrying out various actions. Through number of BCI mechanism and approaches have been proposed by various scientists to empower the individuals for directly controlling their objects via their thoughts. However, the actual implementation and realization of this method faces number of challenging with low accuracy and less interoperability. In addition, the pre-processing signals and feature extraction process is further time consuming and less accurate. In order to overcome the mentioned issues, this paper proposes an accurate and highly inter-operable system using genetic fuzzy system along. The predictive model and analysis can be further improved using canonical correlation analysis. The proposed framework is validated and demonstrated using brain typing system analysis. The results are computed against accuracy, latency and interoperability of the signals received from brain with less SNR along with traditional method. The proposed mechanism shows approximately 87% improvement as compare to existing approaches during the simulation over various performance metrics.

The issue of cybersecurity has gained significant prominence in the context of safeguarding the privacy and integrity of information, especially with the increasing prevalence of interconnected devices such as Mobile Ad-hoc networks (MANETs). Due to their nature, particularly in light of the unavoidable security flaws that arise from the complexity of current and future systems and services, it is crucial to explore how to prevent cyberattacks efficiently. On the other hand, game theory is one of the most significant methods used in this field. Yet, survey articles that consider game theory for security in MANETs are less common. Therefore, in this paper, we survey the articles related to game theory for security in MANETs, presenting an overview of the basic MANETs concepts and their main vulnerabilities. Also, we will propose a taxonomy for recent works, highlighting the limits of the existing works, and outlining some potential future directions for cybersecurity in Ad-hoc networks.

Egocentric vision data captures the first person perspective of a visual stimulus and helps study the gaze behavior in more natural contexts. In this work, we propose a new dataset collected in a free viewing style with an end-to-end data processing pipeline. A group of 25 participants provided their gaze information wearing Tobii Pro Glasses 2 set up at a museum. The gaze stream is post-processed for handling missing or incoherent information. The corresponding video stream is clipped into 20 videos corresponding to 20 museum exhibits and compensated for user’s unwanted head movements. Based on the velocity of directional shifts of the eye, the I-VT algorithm classifies the eye movements into either fixations or saccades. Representative scanpaths are built by generalizing multiple viewers’ gazing styles for all exhibits. Therefore, it is a dataset with both the individual gazing styles of many viewers and the generic trend followed by all of them towards a museum exhibit. The application of our dataset is demonstrated for characterizing the inherent gaze dynamics using state trajectory estimator based on ancestor sampling (STEAS) model in solving gaze data classification and retrieval problems. This dataset can also be used for addressing problems like segmentation, summarization using both conventional machine and deep learning approaches.

Vehicular networks aim to support cooperative warning applications that involve the dissemination of warning messages to reach vehicles in a target area. Due to the high mobility of vehicles, imperative technologies such as software-defined network (SDN) and edge computing (EC) have been proposed for the next-generation vehicular networks. The SDN separates the control plane from data plane entities and executes the control plane software on general purpose hardware. On the other hand, EC aims to reduce the network latency and packet loss rate by pushing the computations to the edge of the network. Nevertheless, the current solutions that integrate SDN and EC could not satisfy the latency requirements for data dissemination of vehicle-to-everything (V2X) services. To bridge the gap between the two technologies, the conventional EC is enhanced to multi-access edge computing (MEC) by collocating the edge computing servers with the radio access networks. In order to improve the latency for V2X services, we propose in this paper, an SDN-based multi-access edge computing framework for the vehicular networks (SDMEV). In the proposed solution, two main algorithms are implemented. First, a fuzzy logic-based algorithm is used to select the head vehicle for each evolved node B (eNB) collocated with road-side unit (RSU) for the purpose of grouping vehicles based on their communication interfaces. Afterward, an OpenFlow algorithm is deployed to update flow tables of forwarding devices at forwarding layers. In addition, a case study is presented and evaluated using the object-oriented modular discrete event network (OMNeT++) simulation framework which includes the INET framework-based SDN. Simulation results depict that the data dissemination based-SDN supported by multi-access edge computing over SDMEV can improve the latency requirements for V2X services.

Just noticeable difference (JND) for stereoscopic 3D content reflects the maximum tolerable distortion; it corresponds to the visibility threshold of the asymmetric distortions in the left and right contents. The 3D-JND models can be used to improve the efficiency of the 3D compression or the 3D quality assessment. Compared to 2D-JND models, the 3D-JND models appeared recently and the related literature is rather limited. In this paper, we give a deep and comprehensive study of the pixel-based 3D-JND models. To our best knowledge, this is the first review on 3D-JND models. Each model is briefly described by giving its rationale and main components in addition to providing exhaustive information about the targeted application, the pros, and cons. Moreover, we present the characteristics of the human visual system presented in these models. In addition, we analyze and compare the 3D-JND models thoroughly using qualitative and quantitative performance evaluation based on Middlebury stereo datasets. Besides, we measure the JND thresholds of the asymmetric distortion based on psychophysical experiments and compare these experimental results to the estimates from the 3D-JND models in order to evaluate the accuracy of each model.

The so-called mode-selective transmission line or simply “MSTL” is studied theoretically and experimentally. This low-loss and low-dispersion transmission line operates with a frequency-dependent mode-switching behavior. This self-adaptive mode-selective guided-wave structure begins with the propagation of electromagnetic waves over the lower frequency range in the form of a quasi-TEM fundamental mode similar to the microstrip line case, then followed by a fundamental quasi-TE10 mode with reference to rectangular waveguide over the higher frequency region. To gain insight into the physical mechanism and fundamental features of this mode-selective transmission line, a detailed semi-analytical hybrid-mode analysis is developed through the application of a method of lines. This method allows accurate and effective modeling of MSTL guided-wave properties. Propagation characteristics of this proposed mode-agile structure in terms of dispersion, modal, and loss properties are examined, which leads to the establishment of some basic MSTL design guidelines. Numerical results confirm the expected mode conversion and low-loss behavior through the observation of field evolutions along the structure. For experimental verification, a set of MSTL prototypes are fabricated on two different substrates through dissimilar fabrication processes. Measurements are carried out from dc-to-500 GHz using a vector network analyzer. Excellent agreement between theoretical and experimental results is observed. It is confirmed that the low-dispersion and low-loss behavior of MSTL makes it an outstanding integrated waveguide in support of high-performance super-broadband signal transmission and/or ultra-fast pulse propagation in a fully-integrated platform.

Recent video coding standards typically use the Rate-distortion optimization (RDO) method, which is essential to appropriately perform mode decisions during encoding process. The newest standard high efficiency video coding (HEVC) introduces complex encoding structures and strong dependency between coding units. Particularly, the Lagrangian multiplier is a primary factor in RDO procedure, which directly affects the rate-distortion (R-D) performance and is defined for an entire video frame. This paper proposes a novel approach for perceptually guiding the RDO process in HEVC. The reference encoder does not consider effectively the perceptual characteristics of the input video and further, the visual sensitivity of each coding tree unit (CTU) in a frame. Inspired by the mechanisms of the human visual system, the proposed solution is a CTU-level adjustment of Lagrangian value based on a set of complementary perceptual features. The proposed scheme concerns important visual information of a CTU and its temporal dependency with adjacent blocks. Feature extraction is implemented in the frequency domain using efficient spatio-temporal analysis. In our experiments, we opted a perceptual mean squared error (MSE) metric and structural similarity (SSIM) index. According to perceptual MSE metric, the BD-rate savings using the Bjontegaard delta measurements, were fairly convincing over the state-of-the-art HEVC software HM16.12; 4.41% and 6.14% for random access (RA) and low delay (LD) encoding settings, respectively. Using SSIM, the BD-Rate achieved 6.95% and 9.86% for RA and LD settings, respectively. The proposed method further demonstrates a superior R-D performance over a compared approach adopting a similar scheme.

Cooperative Intelligent Transportation Systems, mainly represented by vehicular ad hoc networks (VANETs), are among the key components contributing to the Smart City and Smart World paradigms. Based on the continuous exchange of both periodic and event triggered messages, smart vehicles can enhance road safety, while also providing support for comfort applications. In addition to the different communication protocols, securing such communications and establishing a certain trustiness among vehicles are among the main challenges to address, since the presence of dishonest peers can lead to unwanted situations. To this end, existing security solutions are typically divided into two main categories, cryptography and trust, where trust appeared as a complement to cryptography on some specific adversary models and environments where the latter was not enough to mitigate all possible attacks. In this paper, we provide an adversary-oriented survey of the existing trust models for VANETs. We also show when trust is preferable to cryptography, and the opposite. In addition, we show how trust models are usually evaluated in VANET contexts, and finally, we point out some critical scenarios that existing trust models cannot handle, together with some possible solutions.