Advances in Smart Technologies Applications and Case Studies

Cellular communications have evolved to Fifth Generation (5G) to accommodate versatile use cases (UCs) requirements. These UCs are ensured to provide higher data rates, massive connection density and ultra-low latencies into a new era of Internet of Things (IoT) and smart cities. ITU-R has classified UCs into three categories: Enhanced mobile broadband (eMBB), Massive machine-type communication (mMTC) or mIoT (massive IoT), and Ultra-reliable and low-latency communications (URLLC). It is very important to plan the network based on different UC requirements using 5G new radio (NR). In this paper, we design the numerology and corresponding bandwidth parts (BWPs) to support the desired requirements. Exercise of coverage and capacity dimensioning of the network is performed to determine the required cell sites. We then propose mixed-integer linear programming (MILP) based cost optimization model and heuristic. Finally, we evaluate and compare MILP and heuristic topology solutions for network planning in the context of cost minimization.

Radio-Over-Fiber (RoF), the unique blend of both optical and wireless systems, is the last mile solution for increasing the capacity and mobility as well as decreasing the expenses. The idea behind this work is to adapt the RoF system parameters in terms of modulation order, code rate and optical transmitting power according to the channel conditions. This system seeks of optimum parameters in order to deliver high data rates as well as high reliability. Simulation results show better BER performance and optical power efficiency compared to conventional fixed RoF systems.

This paper proposes a solution to distribute 4 × 4 MIMO (Multiple Input Multiple Output) over optical fiber at 60 GHz. This solution is based on the subcarrier multiplexing and OCS (Optical Carrier Suppression) modulation. The 60 GHz signal is generated by using two parallel modulators MZM. The First modulator is set to realize the OCS modulation and the subcarrier multiplexing, while the second MZM is set to generate two frequencies of 24 GHz and 48 GHz by heterodyne at reception. This system is realized in OptiSystem software. The results show a 40 Gb/s of data rate transported through 25 km SMF (Single Mode Fiber).

In this paper, we present an evaluation of railway communications system based on 5G-RoF technology and millimeter wave band. Filter Bank MultiCarrier (FBMC) modulation is used in this work to generate the 5G waveform, and Radio over Fiber (RoF) technology is used to transport the Radio Frequency (RF) signal from Central Station (CS) to Base Stations (BSs). RoF system is based on Single Side Band (SSB) and Optical Carrier Suppression (OCS) modulations. For railway environment, we have considered the mobile radio channel taking into consideration a high velocity of receiver (train), which is characterized by Doppler effect. We have evaluated, by simulation, the performance of this system in terms of Bit Error Rate (BER) and Error Vector Magnitude (EVM). The simulations show that the length of optical link decreases by increasing the velocity of receiver, and the maximum length of railway track, that can be covered by each CS, is approximately 76 km for a velocity up to 300 km/h.

Modeling and simulation of communications systems is an efficient and fast way to highlight the performance and the main design difficulties of the latter. The experience and the real measure are still indispensable tools for the validation of the simulation results and its improvement.Our work consists of real experience of satellite transmission/reception. The experience was done in the national society of radio and television, where we measured and compared the transmitted and the received signal by changing the parameters at the input such as power, frequency, type of coding and modulation.In these experiences, we used a GSERTEL tool as measurement system in order to show the performance according to gain and emission power. As result, we have different types of parameters (C/N, MER BCHBER, LDPCBER, EIRP, Constellation and Link margin).In this paper, we present the main characteristics of satellite transmission/reception and the description of modulation, demodulation, and the encoding types. These characteristics have been experimented and measured by doing a transmission with different frequencies within the National Broadcasting and Television Company, where we have transmitter and receive some data with different parameters. Finally, obtained results are evaluated, compared and discussed.

The purpose of this study is to design a patch antenna for ultra wideband (UWB) applications. The antenna consists of a circular patch printed on the FR4 epoxy dielectric substrate. The global dimension of the proposed antenna is 24 × 14 × 0.8 mm3. We inserted a split ring in the radiating patch in order to reject the band 5–6 GHz to avoid interferences with other applications working in the same band.To study the performance of this antenna and reinforce the results, the antenna parameters such as reflection coefficient, radiation pattern, voltage standing wave ratio (VSWR), current distribution, and gain have been simulated and analyzed using Ansoft HFSS (High frequency structural simulator) and CST (Microwave Studio MWS). From these results, we can conclude that the designed antenna operates in the frequency band 3.85–12.38 GHz with good radiation characteristic which makes the proposed antenna a better device for UWB technology.

A new approach to implement a Radio over Fiber (RoF) system for millimeter wave (mm-wave) is proposed and investigated. At the Central Station (CS) the mm-wave signal is produced using Impulse Radio Multiband On-Off Keying (IR-MBOOK) architecture. Then, we use an external modulator to modulate the optical signal that propagates to the Base Station (BS) through the optical fiber. This system is proposed as a solution to deal with the demands of multi-Gbps data transmission in the 60 GHz band and beyond, for nomadic applications. Low complexity, cost reduction, and performance enhancement are achieved by simplifying the mm-wave generation method. The IR-MBOOK design and external modulation are jointly used in this work. The optical link is based on Single Mode Fiber (SMF) to reach a long distance. At the receiver, a non-coherent receiver has been used in order to down-convert the signal to the baseband. The system efficiency is evaluated and analyzed by quality factor (Q factor) performance. Simulation results show the efficiency of monocycle pulse compared with Gaussian pulse shape, and RoF system with transmission rate of 4 Gbps is successfully achieved up to 45 km.

Calibration of RF exposimeter is a serious problem when conventional measurement methods are used, especially when the morphology of the user, workers or public, changes a lot. In this paper we present a study of the behavior of new measurement methods based on linear regression with respect to morphological change. For this we use three models: Child, Gustav and Emma of different sex, age and dimensions. The study is conducted in the near field for a base station antenna at the DCS band. The study showed a low dependence of the proposed methods on morphology.

In this paper, we present a study of the behavior of new RF-exposimeter measurement methods based on linear regression as a function of frequency and distance change. For this, we use a realistic human model (Gustav) exposed to the radiation of a base station antenna. The frequency bands used are GSM, DCS and LTE-Wimax. The distance varies from the near field zone to the far field zone. The errors induced by the human body have been evaluated. The study showed that errors increase with frequency and decrease in Far Field zone. But in general, the proposed methods have a good response to variations in frequency and distance.

An assessment of the similarity of situations and developments in decision-making processes is under consideration. Specifically, mathematical and IT tools that make it possible to assess plausibility and similarity based on the limited amount of information available. Under observation is the assessing the plausibility of occurrence of armed conflict. The data this study is based on is limited periodically and regionally: starting from the end of World War II, until the year 2008. Developments and situations will be treated as algebraic systems.

Local Binary Pattern is a descriptor whose purpose is to summarize the local structure of the images. The goal is to be able to discriminate different images. This method has gone through a large number of changes and adjustments in different types of applications. This paper reviews various LBP methods for facial expression analysis and proposes a new set of variants. Firstly, the principle of LBP is presented and the main variants for face recognition and facial expression analysis are described. Then, new variants are proposed and finally, a comparison with the referred existing ones is made and analyzed through experiments conducted on facial recognition databases YaleB and ORL, and on facial expressions recognition database JAFFE.

The LBP (Local Binary Pattern) texture descriptor has demonstrated its superiority in several image applications: texture characterization, facial identification and macro-expression recognition. Featuring by LBP characterizes image by its local structures, observing its micro patterns and building a histogram. Each observed pixel is featured and then encoded into one byte. All these codes constitute bins for the histogram. For an efficient classification, encoded bytes can be divided into uniform and non-uniform codes. In standard applications, only uniform codes are used leading to 59 codes. The present work proposes an additional process for the reduction of these codes. The proposal is developed and comparatively evaluated with success to classical LBP. Experimental evaluations are performed on 2 different databases for facial identification and macro-expression recognition respectively, and this for different reduction of code length. Though for macro-expression recognition the proposed features can give lower but comparable performance with the traditional LBP, for facial identification they perform very well and keep excellent efficiency. This approach can be extended to most part of LBP variants while keeping the simplicity of LBP.

We address the problem of image-to-video face retrieval. Given a query image of a person, the aim is to retrieve videos of that same person. The methods proposed so far are based on hand-crafted features. In this work we investigate the use of an off-the-shelf object detection network as a feature extractor by building an image-to-video face retrieval pipeline composed of an offline feature extractor and online filtering and re-ranking steps that use the object proposals learned by a Region Proposal Network (RPN) and their associated representations taken from a CNN. Moreover we study the relevance of features from a fine-tuned network. In addition to that we explore the use of face detection before extracting the features and we test the impact of different similarity metrics. The results obtained are promising.

Image segmentation is one of the main tasks for many computer vision problems. In this paper, a GPU acceleration for a Fractal features extraction method is proposed, followed by our neuro-morphological approach that will allow to segment the images based on the Fractal texture features. In the first step, we use the CUDA environment on an NVIDIA GPU to compute the Fractal features in parallel for each pixel of our image, this makes it possible to optimize the extraction phase before starting the image segmentation by using our approach which is divided into two main steps. Firstly, we train a Kohonen self-organized Map (KSOM) using the extracted features. In the final step, we use our watershed method to extract the modals regions from the KSOM, these regions define the final regions found in the segmented image. To highlight the effectiveness of our parallel implementation, the performance results of the GPU extraction method are compared to his sequential counterpart based on CPU. In addition, the segmentation rate of the proposed approach is compared to the K-means results.

The security of digital images is an important issue that has been receiving considerable attention in the recent past. Different image encryption techniques have been proposed in the literature. Among the fundamental theories in the number theory, we can find the elliptic curve (EC) which is widely used to construct cryptographic primitives. This paper investigates the security of image encryption schemes based on elliptic curves cryptography (ECC) and Chaos theory. More precisely, in this paper, we propose a new image encryption scheme that utilize a new mapping method based on Piecewise Linear Chaotic Map (PWLCM) that converts each pixel of plain image into a point on an elliptic curve. Encryption and decryption process are given in details. After applying encryption, security analysis is performed to show that our scheme cannot only achieve good encryption, but also resist the statistical attacks.

In this article we present a suite of our research work on object shape recognition based on uncoded structured light that is used to acquire 3D information in the form of lines distorted by the object’s relief, from these lines we extract the 1D signals corresponding to the object. These signals are used to extract the features of the 3D object shape. In this article we propose a new approach to determine 3D shape descriptors using 1D signals. And to improve the performance of the recognition system based on 1D signal processing, we thought about implementing more information on the object shape by adding to the characteristic vector other descriptors calculated in frequency domain called frequency-based descriptors. Once the shape descriptors are calculated, we proceed to the classification of descriptor vectors in order to recognize the different shapes of 3D objects. The results of the proposed approach allow 3D object recognition with an accuracy of 99.6% using the ANN classifier on a database formed by 10 objects. We present a comparison between the results obtained by applying our approach to different databases made up of 6, 7 and 10 objects and treat these results according to two classifiers KNN (K Nearest Neighbors) and ANN (Artificial Neural Network).

Fermentation degree is one of the main important indicators of cocoa bean quality. Therefore, accurate estimation of fermentation degree is very important for ensuring the quality of final products. This paper presents a quantitative method for assessing the cocoa beans fermentation degree by image analysis. In this approach, the image of cocoa beans are acquired using a camera and processed to obtain the bean’s target. Then, the target’s pixels are clustered into red, green and blue regions where each region’s pixel presents respectively a maximal value of R, G and B in RGB color space. After that, the first three color moments of each region are calculated from RGB space and used to describe the fermentation degree of the bean. Finally, multi-class support vector machine (SVM) algorithm is used as classifier to discriminate cocoa beans sample into unfermented (UF), partly fermented (PF) and well fermented (WF) categories. Experimental results show that 99.17% of UF beans, 97.50% of PF beans and 100% of WF beans were detected successfully. This results revealed that the proposed method could be used as a fast, accurate and a reliable tool to discriminate cocoa beans according to their fermentation degree for quality assurance.

Plants are the primary food source of humans. They are the raw material of most medicines. Therefore, it was necessary to use artificial intelligence to help those interested in plants to classify and identify various plants types quickly and accurately. In this article we present Neural Shifted Legendre-Fourier Moments, we used shifted Legendre-Fourier moments to extract features from leaves images and build descriptor vectors. These vectors are the inputs of the artificial neural network. We tested this model on MalayaKew (MK) Leaf dataset and we got important results. The validity of this proposed method has been provided under different transformations.

Internet of Things (IoT) is a growing field and its use in home automation is one of the dominating application areas. However, the end-users lack security awareness, whereas the system designers lack the incentives for building secure IoT systems. To address this challenge, we propose the notion of security classes to assess and present the security of complex IoT systems both for the users and for developers. Furthermore, regulatory bodies can use our security classification method as a reference to derive requirements for adequate security. This paper presents a security classification methodology and extends it towards Smart Home Energy Management Systems (SHEMS). We demonstrate its applicability by performing a systematic security classification assessment of an industrial SHEMS. Our results show that the use of security classes is a good indication of the level of security, as well as a guide to improve the security of IoT system.

The problem of secure linear regression calculation has been widely considered in the literature. It involves multiple parties, with a private dataset each, wanting to collectively carry out linear regression on the union of their datasets but are unable to combine the data due to privacy restrictions. The solutions suggested in the literature use different methods from cryptography to securely calculate the regression parameters while keeping the parties’ data private. In this paper, we compare the different algorithms in terms of security, efficiency and accuracy.

lately, Mobile ad hoc network continues to increase their existence, hence becoming a essential in several fields. However, these networks operate without any dedicated infrastructure; therefore they are vulnerable to a numerous of threats. The traditional securing networks methods are unsufficient and currently requires to couple a reactive security solution, such as an Intrusion Detection System (IDS). However, most existing IDS suffer from a number of drawbacks, e.g., high rates of false positives, low efficiency, etc. Using a semantic resource such as ontology could be an effective way to enrich the data on intrusions.This paper presents hybrid architecture based on multi-agent system, which uses complete ontology by considering the intrusion at a higher level of abstraction, also to improve the detection process.

An Intrusion Detection System (IDS) is a mechanism which is intended to spot the abnormal activities in a network traffic. The recent systems are based on intelligent methods such as the Artificial Neural Network, Naïve Bayes, Random Tree...In fact, due to the learning capacity of the intelligent methods especially the Artificial Neural Network, the IDS becomes able to detect the known attacks and also the unknown or the recent ones. For this reason, it is crucial to use an extensive database in learning phase and also to test the system performance. KDD dataset is the most commonly known set and contains a large variety content.The aim of our research consists on exploiting the relevant data of the KDD data set in order to obtain the optimum system based on neural network.

In a comparative analysis, this paper investigates the performance of two open source intrusion detection systems (IDSs) namely SNORT and SURICATA for accurately detecting the malicious traffic on computer networks, an evaluation approach, based on a series of tests. These experiments consisted of a test bed which compared SNORT and SURICATA’s reaction; consist in injecting various traffic loads, characterized by different transmission times, packet numbers, packet sizes and bandwidths, and then analyzing, for each situation, the processing performed on the packets. The study demonstrates that SURICATA would process a higher speed of network traffic than SNORT with lower packet drop rate but it consumed higher computational resources.

It is important to state that, from the intrusion detection point of view, the number of steps that appears in a certain session of an attack process is arbitrary. Indeed, in order to prevent detection, attackers can proceed slowly, following different steps, in many days, or even in many weeks. The main objective of this study is to learn how to shape the progression of an attack process in time. We propose a stochastic model based on Markov chains. The basic assumption of our model consists of evaluating malware valid attack scenarios. The algorithm we have proposed shows that we will be able to identify an attack scenario while the attack process is not yet complete which will help IDS to improve the detection rate of malware attacks.

The Internet of Things (IoT) is now destroying the barriers between the real and digital worlds. However, one of the huge problems that can slow down the development of this global wave, or even stop it, is security. So, it’s considered as a crucial problematic for IoT from the fact that the minimal capacity “things” being used, the physical accessibility to sensors, actuators, and objects, and the openness of the systems, including the fact that most devices will communicate wirelessly. However, existing security solutions and techniques are not adapted to these developments, which impact security management efficiency. Therefore, there is a need to automate certain security management tasks mainly the detection of security attacks, the deploying reaction and assisting the security administrators for taking the right decisions. Based on several works in that paradigm, we provide an adequate and appropriate solution to IoT security management problems by proposing a dynamic security management model which will aim to simplify the security management process.

This paper presents a comparative study between a classical and an improved Direct Torque Control (DTC) strategy for induction motor powered by two level three phase voltage source inverter. The main objective is to make a comparative analysis of two control techniques with a variable and a constant switching frequency to which this latter shows a great improvement of the system performance by reducing stator and rotor flux ripples and improving the current sinusoidal form by optimizing the total harmonic distortion (THD). The hysteresis regulators and voltage vectors selection table of the classical DTC, which directly control the inverter states by reducing the torque and flux errors within prefixed band limits, are replaced by proportional–integral (PI) controllers connected to a sine pulse width modulation (SPWM). This latter is used to generate the quadrature and direct voltages. A fair comparison between the two control strategies has been made and simulated under MATLAB/SIMULINK software.

This paper investigates the effectiveness of the Maximum Power Point Tracking (MPPT) algorithms of solar photovoltaic (PV) systems. Indeed, two efficient new control MPPT algorithms presented are based on the Perturb and Observe (P&O) method with a fixed step. The two suggested controllers are Artificial Neural Network (ANN). The first one uses Multilayer perceptron (MLP) learned by Levenberg Marquard (LM) learning algorithm as a neural regulator named (POPI-LMNN). The second one uses an Adaptative linear Neuron type (Adaline) learned by least mean square algorithm (LMS). The controllers are applied to a DC-DC boost converter inside a standalone solar photovoltaic conversion system used to feed an isolated area. As a result, the responses achieved through the learning approaches are more convincing. That is to say, they are faster and more efficient in terms of the power conversion. This is used to override the limitations of the traditional P&O technique, which are the fluctuations around the maximum power point with their low response tracking performance in the most severe cases of changing weather conditions. A comparative study between the three algorithms is done using Matlab/Simulink Simpower system environment. The results of simulation demonstrate that the Adaline presents a very high performance in terms of rapidity and elimination of oscillations with a conserving energy.

This paper presents the simulation results of a Dish-Stirling power plant with a nominal power of 100 MW considered to be installed in northeastern Morocco: Oujda (latitude: 34.68 °N, Longitude: −1.9 °E). The Dish-Stirling System of the Stirling Energy system (SES) Company with 4–95 kinematic type motor, contains 4 cylinders and the hydrogen uses as a working fluid. The number of collectors required is 4000 with an area of 900000 m2. For this study, the System Advisor Model (SAM) software has been used in order to investigate a potential technical and economical installation of a 100 MW concentrating solar thermal power plant. The simulation results for an annual Direct Normal Irradiation of 1990 kWh/m2/yr predicted that the system would produces 159.3 GWh annually, achieving a maximum power in May. The Levelized Cost of Electricity (LCOE) of the plant and the capacity factor would be 0.16 $/kWh and 18.3%, respectively. All these results should encourage the Moroccan government to exploit this technology for electricity production which would lead to reduction of CO2 and a sustainable development of this region of Morocco.

For many years, the evolution of power electronics became very important in a world where energy aspects have become an essential issue. The appearance of multilevel inverter is one of the results of this evolution. This type of inverter provides high power quality with fewer harmonics. So as to improve the performance of multilevel inverters, many modulation techniques have been proposed. Generally, pulse width modulation (PWM) control is the most used. In this paper, we used a sinusoidal Pulse Width Modulation (SPWM) strategy to control our new 5-level inverter. The performance of our proposed five-level inverter with respect to harmonic content and number of switches is simulated using MATLAB/Simulink. A hardware prototype is developed to verify the performance of the developed system using microcontroller ATmega2560.

In this paper a state space averaged modeling and control is proposed for a dc-dc boost convert. The constitutional time variation nature and the non-linearity made the control of the power electronics an arduous task. Therefore, linear control techniques cannot achieve effective control effect. Furthermore, the paper advances a SMC controller to regulate the boost converter. Moreover, to a comparison to a linear controller in variable conditions.

This paper deals with the nonlinear control of a direct-drive PMSG wind turbines using the super-twisting algorithm. The studied system is assumed supplying a water pumping system for the use in isolated sites and areas. The aim of the proposed control strategy is tracking the wind turbine maximum power point. The designed controllers are based on one of the high-order sliding mode controller (HOSM) versions, which is the super-twisting algorithm. This latter possess many attractive features as the chattering-free behavior, finite time convergence, less information demand, simplicity, stability and robustness against external disturbances. The performance of the whole system in closed-loop mode is assessed through computer simulations.

Covering the necessary power for the stand-alone work of water pumping systems for irrigation using photovoltaic energy is the primary objective of this paper. To achieve this goal, we have adopted the Power Pinch Analysis as a guideline, and the method is based on the Electric System Cascade Analysis for the sizing of the water supply system, which necessarily contains photovoltaic panels and a storage tank.The sizing procedure for obtaining an optimal design of the system technically and economically is preceded by a modeling of system components, solar radiation model and optimal tilt angles for each month to obtain the largest amount of solar radiation. Then, the procedure start from a developed algorithm with multiple inputs, hourly solar radiation, hourly water demand for irrigation, total dynamic head, as well as cost data. The results prove the capability and accuracy of the method in optimally sizing stand-alone photovoltaic water pumping systems based on load profil and climate resources of the worst month of the year.

In the present study a method for sizing of a PV/Wind autonomous hybrid water pumping system is modeled, the procedure followed in the sizing of system components is based on the Electric System Cascade Analysis (ESCA) method. The developed algorithm gives all the possible configurations of the PV panels number, wind turbines and the storage tank size. The choice of the optimal solution is based on economic analysis with the Life Cycle Cost. The algorithm has been demonstrated on a case study with the daily data of the worst month.

Unbalance is one of the most common defects in rotating machinery that causes important vibrations and subsequently an increase in energy consumption. Unbalance is defined as un-equal distribution of weight around the center of rotation. Although vibration, heat and noise are usually the results of mechanical defects, vibrations present the most relevant indicator for the identification of unbalanced shaft. By monitoring time indicators such as the Root Mean Square (RMS), unbalance is detected. This paper aims to experimentally analyze the energy losses due to shaft unbalance. The proposed methodology is meant to emphasise on the relationships between vibration level and energy consumption for different degree unbalance defect. A laboratory test rig was designed to create unbalance by adding weights at various eccentricities. For each case, RMS indicator and the electrical characteristics were measured. The results obtained from this study show that vibration analysis is an effective technique to identify unbalance severity and energy losses.

The brushless DC motors (BLDC) are used in several applications such as electrical cars, medical and industrial equipment, where speed control with high efficiency is required to satisfy specifications regarding load and tracking references variations. In this paper we propose a comparative study between the classical PI controller and a Sliding Mode Controller (SMC) for closed loop speed control of a trapezoidal back-EMF BLDC using a DC/DC buck converter under various load conditions. Simulation results of speed response, torque and current behaviors of the studied BLDC are also presented to support the noted improvements.

The Wind System Power varies according to the wind speed, and the wind turbine (WT) operates at an optimal point which depends on its rotation speed. For that, this paper presents a Fuzzy Logic Approach (FLA) to tracking the maximum available power of a WT system based on Permanent Magnet Synchronous Generator (PMSG) to supply the load. Its main advantage is not requiring a mechanical sensor or a prior knowledge of the WT characteristic. The PMSG–PWM rectifier combination used is compared with other topologies types in term of cost and efficiency. By modifying the PWM rectifier modulation index, the reflected voltage at the PMSG is varied and consequently its rotational speed, which allows tracking the maximum power point (MPP) of WT. So as to check the proposed strategy feasibility, a PSIM and Matlab co-simulation is made. The simulation results demonstrated the competitiveness of FLA, from the point of view of oscillations around the maximum point and response time compared to P&O technique.

This paper investigates the efficiency of an original approach for Maximum Power Point Tracking (MPPT) algorithm applied to a Wind Electric Water Pumping System (WEWPS). The studied model is developed under Matlab/Simulink software and consists of an asynchronous wind turbine, a Static Var Compensator (SVC) and a centrifugal water pump driven by a three phase Induction Motor (IM). The proposed control technique seeks to improve water flow rate by exploring the maximum amount of electrical power produced by the asynchronous wind turbine in a wide range of wind speed. Theoretical analysis as well as simulation results have shown that the highest electrical power rate depends on the value of the produced voltage which can be controlled by the SVC using single input fuzzy logic regulator. Modified Enhanced Perturb and Observe (MEPO) algorithm is then used in this application to calculate the optimal value of the voltage reference that ensure maximum electric power extraction and hence maximal water flow rate. Moreover, a comparison have been made with the conventional P&O algorithm to prove the superior performance of the proposed approach which does neither require to measure wind speed nor to know the WEWPS parameters.

Our study focuses on the problem of multi-source load management in a hybrid energy production system, photovoltaic/wind, associated with a storage system.The connection of these elements to the photovoltaic panel and the wind generator is performed at a DC voltage bus. This continuous bus has the advantage of interconnecting more easily the different elements of the hybrid system. This solution being the one adopted in this work. From the DC bus, the connection to the load is made using DC/DC power converters.The goal is to find a strategy for managing power exchanges between the different elements of the hybrid system.This strategy should optimize the overall performance of the system, properly utilize each source and load, and adapt to the configuration change. The system command should be distributed so that items can be removed without having to change the command policy.It is necessary to study the operation of hybrid power systems in extreme northern climates in order to optimize the efficiency of these systems once in use in these environments.

In this paper, we address the problem of identifying the quality of citation as important or unimportant to the developments presented in the research papers. We gather features represented by four state-of-the-art machine learning techniques and combined them with newly engineered, natural language-based features. Using a known dataset of 465 citations, manually labeled by experts, our approach out-performed state-of-the-art by using fine-tuned Random Forest Classifier with 90.7% F1 score and 97.7% precision. We also employ Convolutional Neural Networks with AdamW optimizer with focal loss function - that converges quickly on small data to achieve considerably significant results.

Allowing Robots to navigate automatically in unknown environments where human being cannot access is a needed characteristic in many fields, such as military, industry, civil engineering. This paper presents an efficient and fast path-planning algorithm for mobile robots, which navigate in a priori unknown indoor environments based on their local capacities of sensing and acting. For navigate in a priori unknown environments, re-planning the path it’s necessary each time the current one cut off by an obstacle, thus the path-planning algorithm must had the ability to re-plan the path once the tracked path become unavailable (online planning). The proposed path-planning algorithm deploy a grid map to characterize the environment, where the environment described by an occupancy and trajectory grids, which smooth the planning task. In order to make the planning faster we had split up the algorithm into two modules: obstacle avoiding and path re-planning. The experiments results had shown the applicability and the fusibility of the proposed approach.

Multi-Agents System (MAS) and Ontology are two technologies capable of creating intelligent reasoning and inferring new knowledge useful for decision making. In this paper we propose a platform model called Agent-SSSN whose agents reason as a human actor and collaborate with them to create a Collective Intelligence in Economic Intelligence (EI) coordinated network monitoring. To organize the knowledge in Agent-SSSN and facilitate the reasoning ability of the agent, an ontology-based approach called Onto-Agent-SSSN is presented in this paper.

Providing the knowledge of solar irradiance under such location is nowadays one of the crucial issue and over the world. This paper presents our performed work regarding the prediction of direct normal irradiance (DNI) using artificial neural networks (ANN) method under Oujda climate (Eastern of Morocco). To explore ANN, we have developed and validated a Matlab code with different statistical performance indicators. DNI is the most sensitive component of solar irradiance because of its high variability. For this purpose, various parameters are chosen as inputs: Global Horizontal Irradiance (GHI), Diffuse Horizontal Irradiance (DHI), Temperature (T), Wind Speed (WS), Wind Direction (WD), and Barometric Pressure (BP), which are recorded from the meteorological station installed at a roof of Mohammed first University in Oujda for two years of measurement period (2015–2016). Making combinations of mentioned variables leads to detect the best ANN architecture with accurate correlations and low errors. To assess the performance of the ANN method, we calculated different statistical indicators in addition to the correlation coefficient of training, testing and validation of the ANN models. Results show that the most appropriate model for our prediction case includes seven inputs (GHI, DHI, T, WS, WD, BP and RH) with 28 hidden neurons in which performs with low errors; 0.16%, 6.6%, 4.2% for the bias, root mean square and mean absolute errors respectively.

Corpora construction is an indispensable task in many natural language processing (NLP) domains. Thus, this paper aims to construct a linguistic resource that covers two NLP domains simultaneously, namely language identification, that distinguishes Moroccan Dialectal Arabic from Modern Standard Arabic, and sentiment analysis. It presents the construction concept, defines data sources, explains the annotation process and describes corpus characteristics. All along the paper, the description is enhanced by statistics and vivid examples. Moreover, the paper studies either language identification or sentiment analysis specificities separately.

Over the past years, there is a growing concern in analyzing social networks and modeling their dynamics at different scales. Most social networks are dynamic and evolve gradually. Also, the communities in these dynamic networks usually have changing members and could grow and shrink over time. Therefore, one of the central challenges is to predict the future orientation of community evolution using the community features mined at different time intervals. Though, both the massive size of data and the dynamic nature of the network make it difficult to efficiently calculate these features. In this paper, we suggest a new approach that studies the structural and temporal features of the network and identify the most important subset of community features in order to predict the future orientation of communities in dynamic social networks. Our framework is to select the significant features associated to a community – its structure and history that guides to precise community event evolution. Contrasting to common methods that result huge number of features at each time interval, our suggested approach demands identifying essential number of community features to adequately define if a community will continue stable or experience certain events like shrinking, splitting or merging. Our experiments on real world datasets confirms the efficiency of the suggested framework.

CropSAT is an interactive decision support system (DSS) that provides vegetation index (VI) maps from Sentinel-2 data all across the globe and lets users delineate fields, design variable-rate application of user specified inputs (mainly nitrogen, but also e.g. fungicides or growth regulators) based on the VI maps. The CropSAT DSS was initially developed in a research project at the Swedish University of Agricultural Sciences (SLU), and has since its launch in 2015 been continuously developed in a private-public-partnership between SLU, private companies and the Swedish Board of Agriculture. Now it has global coverage, is continuously updated with new satellite images, and is provided free-of-charge in multiple languages (including Arabic and French). The present study aims at describing the CropSAT systems, summarizing research results from the ongoing developmental process and pointing to opportunities for applications in precision agriculture, e.g. in Morocco and other countries in North Africa.

In this paper a rice crop prediction performance analysis of five machine learning and two multilinear regression algorithms is presented. A five hectares rice plot was selected. For the database, in the plot, 72 sampling points spatially distributed were defined. For each sampling point, physicochemical, biomass and leaf chlorophyll content measurement were taken at vegetative stage. Additionally, the plot was flown with a quadcopter to take multispectral images in order to calculate vegetation indices maps. As output variable, the crop yield was defined. The machine learning (ML) algorithms used in this analysis were: Random Forest, eXtreme Gradient Boosting, Support Vector Regression Machines, Multilayer Perceptron Regression Neural Networks, and K-Nearest Neighbors; the multilinear algorithms were Partial Least Squares and Multiple Linear regression (MLR). The results show the best performance for K-Nearest Neighbors with an average absolute error for the testing point of 10.74%. The worst case was the MLR with a root mean square error (RMSE) of 2712.26 kg-ha $$^{-1}$$ in the testing dataset, while KNN regression was the best with 1029.69 kg-ha $$^{-1}$$ .

A precise estimate of the evapotranspiration (ET) partitioning is fundamental for determining the crop water needs and optimizing irrigation management. The plant transpiration (T) is generally considered to be the most desirable component, increasing the flow of T within the ET could be one of the most important actions to save water in semi-arid agricultural regions. Given the lack of reference method to estimate the E/T partitioning of wheat crop, this study inter-compares four different methods based on eddy covariance, sap flow and lysimetry measurements and FAO modeling. The objectives are: i) quantify T and ET flows using different approaches and ii) evaluate the response of the FAO dual approach model to different periods of stress. Results indicate that despite the small surface sensed by mini-lysimeters, the partitioning ratio is evaluated more precisely (19% relative error) with lysimetry than with the other systems (any combination of eddy covariance, lysimetry and sap flow measurements). Moreover, stem-scale T measurements from sap flow sensors are subject to representativeness issues at the field scale, and to systematic errors during water-stress and senescence periods.

The present study aims at modeling and analyzing the climate change effects on the runoff of the Rheraya stream (Morocco), taking into account the snow component. For this purpose, we combined the GR4J conceptual rainfall-runoff model with the CemaNeige snow module calibrated over 1989–2009, for daily time step simulations. Med-Cordex climate projections for the period 2020–2040 were used to simulate the future evolution of precipitations, snow cover area and runoff according to the Representative Concentration Pathway scenarios (RCP); RCP 4.5 (stabilization of emissions) and RCP 8.5 (trend scenario).The results showed a decrease in rainfall, snow cover area and runoff for both RCP 4.5 and RCP 8.5 scenarios at the 2040’s horizon when compared to the 1989–2009 baseline. The annual rainfall decreases by −18.4% for the RCP 4.5 scenario and −19.6% for the RCP 8.5 scenario. The snow cover area decreases by −38% and −48% for RCP 4.5 and RCP 8.5, respectively. Finally, the average annual runoff decreases by −9% for RCP 4.5 and −28% for RCP 8.5 scenarios. The forthcoming consequences would seriously affect irrigation and groundwater availability in the area.

The management of water resources in agricultural areas requires a precise knowledge of the evapotranspiration (ET). For this purpose, many measurement techniques are developed to quantify this variable. The eddy covariance system (EC) is the only method that can measure directly ET with high precision. However, over heterogeneous areas where the complexity is associated to the type and the cover of the vegetation canopy, to the soil moisture and to the changes in topography, a network of EC systems is needed which is costly and requires a continuous availability of well-trained staff to operate and maintain the devices. To overcome this issue, the scientific community develops other approaches based on modeling algorithms which can provide estimated values of the ET. In the present study, Shuttleworth-Wallace (SW) model was used to estimate ET over a mixed vegetation of olive trees and wheat in semi-arid climate conditions. The estimated ET shows good consistency during two years, 2017 and 2018 of study, with an RMSE of about 0.49 and 0.51 mm/day, respectively.

The agriculture development based essentially on water resource. In several places around the globe, fresh water is scarce and its cost become high, thus an optimal management of this resource is indispensable for a sustainable agriculture. Therefore, the implementation of an intelligent irrigation platform to insure the watering process by applying a precise amount of water to crops at precise locations at precise times in order to optimize crop yield and water productivity is necessary. The proposed platform uses sensory data and communication technologies, giving the farmer a way to consult data from different nodes sensor from any place in the field in a simple way. The proposed architecture based on gateway node and different wireless nodes equipped with LoRa connectivity. Each wireless node connected to different sensors and actuators, in addition to a solar panel, giving the node unlimited autonomy. Different industrial sensors used, such as soil moisture, temperature and humidity sensors for measuring different parameters of the soil, plant and atmosphere. The data is send and processed on a gateway node and a remote server, in order to store the sensory data, then allowing further consultation and analysis of data in a versatile way. Due to the energy autonomy and low cost, the platform can be useful for geographically isolated and water limited areas.

The study of human brain activity using electroencephalogram (EEG) is a growing multidisciplinary field that links electronics, psychology and cognitive science to learn the effect of human brainwave activities in various fields e.g. meditation. In this research, an experiment is performed to compare the EEG-Alpha rhythm between Zikr meditation and listening relaxing music. The hardware use is the EMOTIV EPOCH Neuroheadset of 16 electrode channels. The investigation are focused on the Electroencephalography (EEG) signal for exploring the relativity of attentional control by observing Topographic Map through Event Related Potential (ERP) analysis to see whether the brain experience any changes. Then, Power Spectrum Density (PSD) through Auditory Evoked Potential (AEP) analysis is recorded to see the different in each of brain lobes. The effect of the EEG spectral analysis on neurofeedback relaxation technique is also been observed. The results were observed from the frontal lobe and the average alpha power in the region is obtained. Comparison of the experiments works were carried out to observe the relaxation states of the subject while listening to Zikr and the relaxing music. Results from 5 subjects shows that 90% of the subjects’ alpha wave was dominant to Zikr. From the result, 4 out of 5 students tend to have better relaxation when listening to Zikr thus proving that the participants are much more relax after listening the Zikr.

Human computer interaction (HCI) is becoming an essential area of study these days. However, most contemporary HCI systems are unable to identify human emotional states and use this information in deciding upon proper actions to execute. To solve this problem, eye tracking has been introduced to record the eye gaze behaviours which can signify the insight of emotions. This manuscript will look into this study by empirically attempt to apply the eye tracking device and method to investigate the relationship of human emotions and eye-gaze behaviour. We explored the pupil size and fixation duration stimulated by film clips of different arousal content using the eye tracking technology. Fifteen students from Malaysian-Japan International Institute of Technology (MJIIT) are chosen. Emotions are analysed by studying the eye gaze behaviours using five emotional video stimuli e.g. Amusement, Joy, Neutral, Sad and Fear. These stimuli are displayed to the subjects to obtain and record their gaze behaviour using the Tobii TX300 eye tracker. The results which are obtained are analysed using statistical ANOVA analysis. The ANOVA analysis shows that the significance, p value less than 0.05 for both fixation duration and pupil dilation which indicates there is significant relationship between eye-gaze behaviour and human emotions.

In recent decades, researchers have proposed a number of string matching algorithms, searching for instances of one model string in another string or body of text. Channel matching is usually deployed in plagiarism, text mining, network intrusion detection, form recognition, information security, application in bioinformatics and other areas of expertise. In this study, a new algorithm will be programmed, adapted and applied in the field of DNA sequence analysis based on the approach of the transition from discrete to continuous unlike most algorithms, which are based on the matching channels concept. Experimental results, revealed that our approach ensures an efficient pairing of DNA sequences.

In this paper, we use student agency analytics to examine how university students who assessed to have low agency resources describe their study experiences. Students ( $$n=292$$ ) completed the Agency of University Students (AUS) questionnaire. Furthermore, they reported what kinds of restrictions they experienced during the university course they attended. Four different agency profiles were identified using robust clustering. We then conducted a thematic analysis of the open-ended answers of students who assessed to have low agency resources. Issues relating to competence beliefs, self-efficacy, student-teacher relations, time as a resource, student well-being, and course contents seemed to be restrictive factors among the students in the low agency profile. The results could provide guidelines for designing systems for smart education.

Making in education is an emergent practice focusing on learners as creators of things in a collaborative fashion while promoting knowledge construction through technology, design, and creative self-expression. Teachers’ ( $$n=33$$ ) opinions about making were studied using an online questionnaire after they had attended an online course for professional development about making in education. The results suggest that there exists a group of educators who consider making as a promising approach in education and want to promote its use in schools.

Teachers are expected to lead the innovative use of Information Communication and Technology (ICT) at the classroom level of context. However, research literature shows that a number of factors influence their ICT pedagogical practices. Therefore, the present study investigates the influence of school level support on teacher educators’ technology integration. A mixed method is used to collect data through three focused interviews (N = 19) and self-completion survey (N = 136). The data collected is analysed both qualitatively and quantitatively. The result shows support for the model hypothesized and suggests that the ICT pedagogical practices of the teacher educators do not predict their technology integration. Also, there was evidence that the school level context influenced teachers’ ICT competence, which is necessary for successful technology integration.

Entrepreneurship which has become a major influence for the development of modern nation and individual in respect of innovation, competitiveness and economic strength, is one of the important skills students nowadays should have as a preparation before they graduated from university. However, entrepreneurial venture is a challenging task especially for student. This study therefore is carried out to encourage student to be an entrepreneur by investigating the difference of mind behavior between students with and without entrepreneurial interest through functional Near-infrared Spectroscopy (fNIRS) technology. The study was conducted on thirty engineering students that got high score and low score during entrepreneurial opportunity recognition (EOR) survey. The result indicates that student with entrepreneurial interest has a high opportunity recognition during the moment of finding the entrepreneurial opportunities from stimulus given due to their high attention and working memory function. This investigation hopes to give motivation among students on their inclination and enthusiasm in welcoming the entrepreneurial call.

Artificial Intelligence (AI) is evolving quickly and is likely to be the most disruptive force in technology, which could reshape our future and transform our economies and societies. AI uses algorithms that change the way we interact with the world; it challenges us to examine our assumptions about the world and how we make decisions. Does this technological revolution raise new moral questions, as the industrial revolution did? We must use AI responsibly. But how? Is there a technological determinism? Do we need to reconsider the good and the bad related to innovations and their challenges? This article will propose a semantic investigation of the terms “AI” associated with “ethics” before electing the most appropriate approach to making an ethical AI. Would an ethical AI make utilitarian choices or choices of conviction?The proposed protocol is based on the idea of asking how to name “Ethics + AI” in light of technological imperatives and moral challenges. First, we will identify the constitutive concepts of ethics, “ethics of AI,” and “ethical AI.” (Then we will identify the foundations of the conceptual matrix needed to bring closer the two concepts. Or is it a single concept with two faces?).

This paper treats some ethical issues of open access publishing in the Moroccan academic context through a survey conducted among researchers and decision makers at Abdelmalek Essaâdi University and also among some notorious Moroccan scientific journal publishers. We questioned faculty researchers about their points of view on publishers’ commercial policy that appears to be against ethical and scientific values. We present some arguments against publishers’ market policy such as knowledge is a public good not a commodity and author must not pay to receive information that he provided. Ethic is present during all open access publishing process.

Most of the real world problems involving decisions are part of a complex environment, where multiple objectives and points of view have to be taken into account and the diversity of value systems are present. This observation motivated our study, which focuses on the development and/or validation of mathematical models and computer tools for complex-problems evaluation via expert knowledge. As an example of a complex situation, the evaluation of catch-up sessions is treated. Usually, in a catch-up evaluation the final score is either the average or the maximum value and the choice is in most cases unjustified. Thus, the evaluation of the objectives achievement could only be in turn deficient. Using our new intelligent evaluation system based on Knowledge Based Expert Systems (KBES) and fuzzy logic lead us to discover new models for evaluating such complex problems. The principle of this system is to reproduce the cognitive mechanisms of an expert in evaluation.

Databases store a significant amount of data from the world, however, to access this data, users must understand a query language such as SQL. In order to facilitate this task and to make the interaction with databases possible for all the world, researches has recently appeared to approach systems that understand the natural language questions and automatically convert them into SQL queries. The purpose of this article is to provide the state of the art text-to-SQL task in which we present the main models and existing solutions based on Artificial Neural Networks (ANN), precisely on Deep Learning (DL) and Natural Language Processing (NLP). We also specify the experimental settings of each approach, their limits as well as a comparison of the best existing ones.

Support Vector Machines algorithms, have been and are still widely used for several Machine Learning tasks, thanks to many advantages including efficiency, precision, and great performance. Lately, the need for processing large scale data is dramatically increasing, in which High-Performance Computing Clouds have proved to be the ultimate solution. However, in the specific case of SVMs, with their different types, these applications focus on the search for performance in terms of precision and reduction of the error rate, generally on one or several data sets representing a concrete use case. This work seeks to extract a clearer idea about the suitability of each SVM and Kernel type, according to the example studied, and the impact of the distributed computing on the results.

Planning a global IT project stands for facing two issues, complexity and visibility. So agile methods can be key of success, due to their ability to fast reacting when impacting events come up.In this work, we seek to provide assistance for planning Global IT projects, based on an agile planning approach, by improving the platform we developed for planning usual IT project. We propose a planning assistance platform based on solving the planning problem modelled as a constraint satisfaction problem (CSP). Which should help project managers to analyse the project feasibility and to generate useful schedules and charts. We also allow managers to supervise the project progressing and regenerate the schedule if an adjustment is made or a change has occurred at any phase of the project life cycle.

To reduce the cost of maintenance, software quality is more and more studied. In this area, to present software quality, many models have appeared, among ISO 25000. In other side, to evaluate software quality, several metrics have appeared. The two axes do not have a clear link to assert a complete evaluation. Our work consists to establish the link between this two axes by adding an intermediate layer containing the quality concepts. This work was applied to ISO 25000 model, and to validate it we were establish a survey on this application. In this paper, we present a new results on our survey that aims to validate our work.

In this work, we present a new method of semantic analysis and data allowing the automatic generation of source code from specifications and descriptions written in natural languages (NL2Code). Our long-term goal is to allow any user to create an application from a specification describing the need for a complete system. It involves the realization of a study, the design and the implementation of an intelligent system allowing the automatic generation of a computer project by answering the needs of the user (skeleton, configuration, scripts of initialization, ...) expressed in natural language. We are taking a first step in this area by providing a new dataset specific to our Novelis company and putting in place an approach that allows the machine to understand the user’s need, expressed in natural language in a specific area.

Software Testing is a process of investigating a software product to identify possible mismatches between expected and implemented system requirements. One of the main motivations of software testing is to ensure the correctness of a software system. Thus, due to the continuous quest for better quality of software system, the efforts spent on testing phase are enormous and costly toward both the user and the developer. Many tools and approaches were proposed to solve this problem. Model-Based Testing (MBT) is one of the new technologies to meet the challenges imposed on software testing. Compared to traditional testing methods, MBT is able to manage and accomplish testing tasks in a cheaper and more efficient way. MBT relies on models of the system under test and/or its environment to derive test cases for the system. Models can be used in different manners and for different purposes such as improving specifications quality, code generation, and test generation. In this paper, we use models as the core element of our approach that aims to generate test cases for the different abstraction levels of MDA development life cycle. Our approach uses UML diagrams and Semantic Business Vocabulary and Business Rules (SBVR) to model MDA levels then we propose transformations to generate test cases in order to evaluate if the generated code fulfills system requirements. We aim to implement transformations in an Eclipse plugin.

This paper presents the impact of weather conditions on the optical degradation of CSP solar reflectors under Eastern Moroccan climate while highlighting their contribution rates on the natural soiling effect. Within this context, two experiments were conducted at Mohammed First University to assess the optical degradation of soiled-CSP reflectors via a Real-time Cleanliness Monitoring System. The experimental results show that the natural accumulation of dust leads to a weekly decrease in a range of 5.33–16.6% of the cleanliness/reflectivity of the CSP reflectors. The precipitation intensity, wind speed and relative humidity of the site are the main meteorological factors contributing to the soiling of the CSP concentrators. Moreover, the cleanliness/reflectivity of CSP mirrors has been dramatically decreased to 45% in merely 11 weeks of exposure, reflecting the importance of regular cleaning of CSP solar fields. Further investigations and experimentations are needed to study the correlation between soiling and the above-mentioned contributing factors to develop effective site-specific cleanup methods for each CSP plant.

Historically, the study of fatigue failure was neglecting by numerous designers due to misunderstanding of this phenomenon. Then, it contribute a series of serious accidents what makes its analysis today very important and required in conception phase. Consequently, the fatigue occurs in three stages; the initiation of cracks and, the propagation of macroscopic cracks until a final step which represented the sudden rupture. In this paper, a fatigue and static study of welded structure used frequently for electric vehicles utilization was built using the numerical modeling, in order to evaluate the equivalent stress which is one of the factors that can influence to the fatigue behavior of structure. Moreover, the results indicated the impact of spring utilization on fatigue life in the critical areas. For that reason, a comparison of fatigue simulation results were analyzed between two models based on welded rectangular profiles with same dimensions but one is connected to a longitudinal spring and the other without. Finally, we have demonstrated that the use of spring increases more the durability of our welded structure than that of structure with connected spring.

This paper evaluates the direct normal irradiation under Oujda climate. To achieve this investigation, firstly, we compared DNI data obtained from the satellite-based irradiance data in Copernicus Atmosphere Monitoring Service (CAMS) to those recorded by the meteorological station located at a roof of the Oujda University (Morocco) from January 2015 to April 2017. CAMS data are accessible via SoDa Service website. The evaluation relies on a comparison of hourly, daily and monthly irradiation and the performance of CAMS is given by calculating classical statistical performance indicators. As a result, this satellite dataset exhibits satisfactory performances with almost great correlations. Secondly, we studied the impact of different atmospheric components, namely cloud parameters (clear sky index and cloud coverage), aerosols and water vapor on the DNI error issued by CAMS under the same climate and during the same measurement period. Therefore, correlations between these atmospheric parameters and CAMS error are carried out and results show that cloud is the first important factor affecting the solar irradiance derivation, followed by aerosols then the water vapor.

In this paper, the Earth-to-Air Heat Exchanger is modeled and validated in TRNSYS. Then, the model is used to evaluate the performance of the EAHE for cooling and heating purposes during the warm June-September and the cold December-March periods of the semi-arid climate of Oujda city. The studied EAHE is a 52 m PVC pipe buried at 2.5 m, with a diameter of 15 cm and an air flow rate of 280 m3/h, which is sufficient for the air renewal of a 100 m2 house. The results are very convincing, so, for the hottest day of the summer, when the outside temperature is 39.1 °C, the EAHE provides an outlet air temperature of 22.67 °C, making it very conceivable for direct use in the building with a total cooling potential of 1102 kWh. While for heating, the results show that the direct use of this system is not suitable for winter in spite of its 861 kWh heating potential, though, it may be used for the preheating of the air.

The NaI(Tl) detectors are widely used in Gamma-Ray Spectrometry researches and applications due to many advantages of their detection efficiency, power consumption, cost, and applicability. These detectors are connected with a multichannel analyzer device that must be correctly calibrated for optimal results. In this paper, we present calculation methods and results of the energy calibration, energy resolution and detector efficiency of a NaI(Tl) 3” $$\times $$ 3” scintillator. A validation of MCNPX simulation was established to evaluate the response of the NaI(Tl) with five different sources.

GATE (Geant4 Application for Tomographic Emission) is a user-friendly environment and an open source software. It presents a new release of specific codes despite it was initially focused on imaging simulations like CTscanner, PET and SPECT. The new version v8.2 (February, 2019) offers new tools dedicated to brachytherapy and especially to radiation therapy simulations. The aim of this work is to simulate the interaction and distribution of particles in the dependent-patient part of a 6 MV X-Ray Beam, generated by Elekta linac, based on Monte Carlo Simulation, using GATE software. In order to accelerate our simulation, the phase space (PS) data freely available in the International Atomic Energy Agency database and cluster computing (Slurm HPC-MARWAN, CNRST, Morocco) are used. However, the results show that there are many interactions at all positions and internal structures of the dependent patient part. Thus, contribution of electrons and positrons appears in the created field previously designed to be an X-ray beam.

In this study, the potential of deploying a large-scale solar tower plant in eastern Morocco is investigated. A detailed analysis of the influence of solar field (SM) and energy storage sizes (TES) on the annual performance of central receiver concentrated solar thermal plants with thermal molten salt direct energy storage is realized. Also, a parametric study on the effect of SM and TES hours on the Levelized Cost of Energy (LCOE) has been carried out to determine which plant has lowest LCOE using the System Advisor Model software. From results obtained, for 100 MWe solar tower plants with direct molten salt storage, it has been found that the best configuration of the heliostats field to minimize the LCOE is determined. Also, the optimal solar multiple under Oujda climate conditions is 3 for 14 h storage sizes and LCOE equal to 10.81 c$/kWh.

The Ad hoc network is a part of the IoT environment, including several types of networks like MANET. It uses different categories of routing protocol, but the OLSR routing protocol is the best one for MANET. The mobility concept has an impact on the evolution of network performances. In the OLSR routing protocol, this mobility influences on the choice of the MultiPoint Relay (MPR). In this paper, the main objective is to develop an algorithm enhancing MPR selection process in such networks. This algorithm based on Average Spatial Dependency metric “ASD” linked to the average Relative Speed “RS” and the average Relative Acceleration “RA”. Each node shares these values with its neighbors using the messages of OLSR routing protocol (Hello and TC messages). Furthermore, if the nodes have equal reachability, the highest ASD value will be selected as a criterion of the choice of their MPR set. In the simulation, we have applicated the Manhattan mobility model in the MANET network, and we have used the NS3 simulator. The results of this simulation have shown that the mobility concept could change network performances in terms of packet loss, end-to-end delay, and Throughput.

The need to manage the electrical grid in real time has led researchers to propose reliable solutions based on new technologies. Our contribution in this subject is to develop a smart meter connected to an advanced metering infrastructure. The chosen communication technology and the developed network are presented in this paper. The presented network solution is based on LoRa modulation technology and LoRaWAN network. We adapted the network to handle a hug number of smart meters using frequency and time multiplexing. Based on real measurement of LoRaWAN ranges for different spreading factors, we simulated the network on MATLAB by modeling the LoRaWAN based smart metering network and the urban environment. As results, with the proposed network structure 98.45% of sent packets was received.

Early detection and preventive measures is the primary way of minimizing the damage and casualties caused to natural and human resources. However, the disasters caused by fires require an immediate response. The work presented in this paper describes a proposal to make a short-term forest fire risk assessment, this is to improve the response time of time of emergency corps and existing forest fire prevention, detection, and monitoring systems. In order to do it, the introduction and implementation of Wireless Sensor Networks will enable real - time environmental monitoring of dynamic forest fire risk factors.

In wireless sensor networks (WSNs) the lifespan of nodes is limited due to the scares energy in their batteries. According to tasks allowed by nodes, many variables affect the amount of their energy and influence the lifetime. Among them the number of nodes, the diameter of the network, the location of the base station, the initial system energy, etc. In this work, we have attempted an approach based on global (GDoE) and sequential design of the experiment (SDoE). Specifically, we have modeled the dissipated energy in function of significant factors. In both cases, the investigation has led to a deep exploration of the overall domain defined by the variables. Besides, we have obtained good polynomial models (Taylor approach) with high statistical characteristics. However, although optimal values in terms of energy are comparable for both approaches, the study domain is greatly explored and different irregularities are involved only with the SDoE approach. In addition, through this methodology, we have conducted the hybridization of the present approaches with hierarchical routing protocol in WSNs. Thence, we have studied and evaluated a routing mechanism such as LEACH protocol and then displayed its potentials. Finally, the resulted polynomial models can be good alternatives to hierarchical routing protocols.

Currently, in electrical energy sector, due to the population growth and the increasing of energy consumption demand, the electrical grid is becoming more and more complex. This creates new challenges in term of electrical energy management. Our contribution in the area, presented in this paper, is the design and the implementation of a Smart Electrical Energy Meter (SEEM) integrated in an advanced metering infrastructure. This SEEM is two way communicating with the electrical operator server using LoRa communication protocol and through LoRaWAN network. We describe the hardware and software architecture of the SEEM as well as the developed communication strategy to enhance the network capacity while reducing collision problems. The performance of the developed system was evaluated in term of measurement and communication by means of experimental tests, which prove the reliability of the system.

Reliability and timeliness are two extremely important factors in Wireless Body Area Networks (WBANs) that make the design of MAC protocols a challenging task. Indeed, with the popularity of these networks, it has become crucial to design an appropriate MAC protocol that provides a high level of reliability within a relatively low delay. Since numerous researchers have considered the IEEE standards 802.15.4 and 802.15.6 for WBANs, and most of the MAC protocols proposed for such networks are primarily derived from one of these two norms, it will be useful to consider the following question: which norm best supports the communication requirements in WBANs? In this context, the current paper presents a fair performance comparison of the IEEE standards 802.15.4 and 802.15.6 under WBANs based on the MAC sub-layer, and provides useful insights about the best solution to be chosen for the regular traffic applications of these networks.