main-content

Über dieses Buch

This book presents peer-reviewed articles from the 6th International Conference on Wireless Technologies, Embedded and Intelligent Systems (WITS 2020), held at Fez, Morocco. It presents original research results, new ideas and practical lessons learnt that touch on all aspects of wireless technologies, embedded and intelligent systems. WITS is an international conference that serves researchers, scholars, professionals, students and academicians looking to foster both working relationships and gain access to the latest research results. Topics covered include Telecoms & Wireless Networking Electronics & Multimedia Embedded & Intelligent Systems Renewable Energies.

Inhaltsverzeichnis

A Hybrid Indoor Localization Framework in an IoT Ecosystem

The Global Position System (GPS) does not work in the indoor environment because of the satellite signal attenuation. To overcome this lack, we propose a Hybrid Indoor Positioning and Navigation System (HIPNS), based on Li-Fi (Light-Fidelity) localization and optical camera positioning analyses deployed in an indoor environment. The localization approach is based on the fuse of two positioning strategies where the camera-based part is responsible for localizing individuals and recovering their trajectories in zones with low coverage of Li-Fi LEDs. A third-party element is planned to operate in the event of loss of contact. So, the step detection technique and heading estimation are applied in a smartphone-based indoor localization context between two referenced points. The main contribution of this paper focuses on the use of techniques, algorithms, and methods from different spheres of application that generate heterogeneous data. We apply a data integration approach based on REST Web service architecture to allow localization operations in this hybrid indoor positioning system (HIPS). In this work-in-progress paper, we also present a state-of-the-art survey of techniques and algorithms for indoor positioning with the help of smartphones, as well as the main concepts and challenges related to this emergent area.

Marc Junior Pierre Nkengue, Ivan Madjarov, Jean Luc Damoiseaux, Rabah Iguernaissi

Current Works on IDS Development Strategies for IoT

Intrusions into the networks of the connected objects are rapidly evolving and affect its entire architecture (physical, network, application layers), as devices, networks and applications are increasingly connected and integrated. Securing these systems, which are generally constrained in resources, is becoming a necessity. Intrusion Detection Systems have proven to be an important security tool to detect attacks on the IoT network and resources. To create the IDS, Security researchers have recently used machine learning techniques because of the excellent results given by these methods (image and voice recognition, product recommendation, detection of spam and financial fraud …). Deep learning methods known for his or her successful ability to extract high-level functionality from big data are often a resilient mechanism for detecting small variants of attacks. The target of this work is to provide a general study on IDSs implementation techniques for IoT, precisely the classical methods and also the machine learning techniques. Finally, we give some recommendations of selected works that have practiced each of the methods presented.

New Metrics to Measure the Quality of the Ranking Results Obtained by the Multi-criteria Decision-Making Methods

Nowadays, there is a panoply of multi-criteria decision-making methods which are proposed in the literature to solve the ranking problematic, where each method has its resolution process and has its drawbacks and advantages. These methods aim to rank from best to worst a finite set of alternatives while taking into account a set of conflictual criteria. The purpose of this article is to propose specific metrics that will be useful to measure the quality of the rankings obtained by different methods. Thus, these quality measures can help the decision-maker to choose the best ranking objectively when adopting several methods. To show and prove the importance and relevance of the metrics proposed, a set of twenty-five examples of rankings will be examined. The results of the experiment conclusively show that all the proposed metrics lead to significant and equivalent quality measures.

Mohammed Chaouki Abounaima, Loubna Lamrini, Fatima Zahra EL Mazouri, Noureddine EL Makhfi, Mohammed Talibi Alaoui, Mohamed Ouzarf

LiteNet: A Novel Approach for Traffic Sign Classification Using a Light Architecture

Naim, Soufiane Moumkine, Noureddine This paper presents a deep convolutional neural network architecture to classify the traffic signs of the GTSRB dataset. Our method uses a very light architecture with a few number of parameters that achieve good results without the need of hard computation. To get at our goal, we use a filter bank. The aim of which being to extract more features, which will be used as input to a fully connected classifier. The recognition rate of our model gets an accuracy of 99.15%, overpassing the human performance being 98.81%. This way, LiteNet competes the best state of art architectures since our approach uses less memory and less computation.

Soufiane Naim, Noureddine Moumkine

The Attitude of Moroccan University Students Towards an Online Assistive Application of Stress Management

Many studies show the efficacy of mental health applications to reduce several types of mental problems encountered by university students. Indeed, this type of mental treatment is not adopted at the national level. In this paper, we present the results of an online survey of acceptance of an assistive stress’ management application by Moroccan university students. A total of 421 medical students were invited to complete an online survey published in the official web site of the Faculty of Medicine and Pharmacy of Fez. The mean age was 21.52 (SD = 2.05) and females represented the majority of our population (63.4%). The results of our investigation show a general acceptance of an online antistress application by a good proportion of our population (36.6%). A mobile app seems to be more accepted by our students (22.6%) than a web app (14.0%). The findings of this paper will be explored to design an evidence-based antistress app that will be designed to help our university students to access to professional online help to better manage their psychological problem.

Hakima EL Madani, Ikrame Yazghich, Maryem Baya, Mohamed Berraho

Detection and Prediction of Driver Drowsiness for the Prevention of Road Accidents Using Deep Neural Networks Techniques

Driver drowsiness is one of the reasons for a large number of road accidents in the world. In this paper, we have proposed an approach for the detection and prediction of the driver’s drowsiness based on his facial features. This approach is based on deep learning techniques using convolutional neural networks CNN, with Transfer learning and Training from Scratch, to train a CNN model. A comparison between the two methods based on model size, accuracy and training time has also been made. The proposed algorithm uses the cascade object detector (Viola-Jones algorithm) for detecting and extracting the driver’s face from images, the images extracted from the videos of the Real-Life Drowsiness Dataset RLDD will act as the dataset for training and testing the CNN model. The extracted model can achieve an accuracy of more than 96% and can be saved as a file and used to classify images as driver Drowsy or Non-Drowsy with the predicted label and probabilities for each class.

Ismail Nasri, Mohammed Karrouchi, Hajar Snoussi, Kamal Kassmi, Abdelhafid Messaoudi

A New Framework to Secure Cloud Based e-Learning Systems

Aissaoui, Karima Amane, Meryem Berrada, Mohammed Madani, Mohammed AmineCombining cloud computing with e-learning has led to a new form of systems called: cloud-based e-learning systems. Those systems take advantages and benefits of cloud computing, and combine them with e-learning systems. This combination offers some solutions to make e-learning systems more efficient and easier for use, and contribute to deal the best conditions of using distance learning systems. However, cloud-based e-learning systems present some challenges in two principal axis: security and storage. In this paper, we propose a new architecture that aims to resolve the problems of these systems, related to security and storage. It is based on a new security layer, responsible of controlling and storing all transactions, in order to generate a security key, and to give us the ability to use generated data to offer recommended systems in the future. Also, this architecture is proposed after a study that we conducted to cover many works done related with this field.

A Term Weighting Scheme Using Fuzzy Logic for Enhancing Candidate Screening Task

Habous, Amine Nfaoui, El HabibThe candidate screening is an essential task in the recruitment process. It is about choosing a suitable candidate that satisfies the recruiter requirements for a given job position. The evolution of information technologies leads to an increase in the use of the recruitment web portals by the candidates that apply for the job positions published in the job boards. Thus the candidate screening process automation becomes necessary to handle the enormous volume of CVs applying for the job positions. In Information Technology (IT) domain, the technology skills are the key competencies to identify the job profile; Consequently, they have priority to the candidate screening task. In this paper, we enhance the candidate screening task in the IT field. For this purpose, we propose a fuzzy-based weighting scheme using domain ontology for Information Retrieval (IR). Experimental results on a recruiter company data show the effective results of our proposed solution.

Amine Habous, El Habib Nfaoui

E-learning Recommendation System Based on Cloud Computing

E-learning in higher education has been known as great technology to improve efficiency, augment focus and thereby, give better academic outcomes, and given its several advantages and benefits, e-learning is considered among the best strategies for instruction. Furthermore, the e-learning system can help students save time and showing further information improving student learning. However, the traditional system for conducting research work and choosing courses is a time-consuming and uninteresting activity, which not only seriously affects students’ academic performance, but also affects students' learning experience, and due to information overload, it becomes more difficult to choose relevant learning resources. To resolve this problem, this paper presents a model of a recommender system for the e-learning platform that will recommend and motivate the student in selecting the courses according to their requirements; this system based on cloud computing infrastructure and particularly with the use of Google cloud services.

Mounia Rahhali, Lahcen Oughdir, Youssef Jedidi, Youssef Lahmadi, Mohammed Zakariae El Khattabi

An Intelligent System Based on Heart Rate Variability Measures and Machine Learning Techniques for Classification of Normal and Growth Restricted Children

Growth restricted children have higher predisposition of developing metabolic syndrome, type-2 diabetes, hypertension and cardiovascular problems in later life. Numerous intelligence systems that have proved their effectiveness for detection of cardiac abnormalities to support medical diagnosis. Previous studies used heart rate variability (HRV) analysis techniques for distinguishing normal and growth restricted children, however those studies did not use intelligent systems for this purpose. The aim of present study is to develop an intelligent system using HRV analysis measures and machine learning (ML) techniques for early detection of cardiac abnormalities in growth restricted children. We performed two sets of experiments using interbeat interval time series data of the normal and growth restricted children and different combinations of individual characteristics of the subjects. Several ML algorithms such as linear discriminant analysis (LDA), support vector machine with linear and sigmoid kernels (SVML and SVMS), random forest (RF), and RPart are used for developing intelligent system to classify normal and growth restricted children. We evaluated the performance of the classifiers using sensitivity, specificity, area under receiver operator characteristic curve and total accuracy. The results reveal that the LDA is robust for classifying normal and LBW-IUGR children with 100% accuracy at all cross validation formulations. The SVMS and LDA revealed highest accuracy, whereas, RF and Rpart were robust for classifying LBW-IUGR and ABW_IUGR. Our findings show that the intelligent system developed using HRV analysis markers and ML techniques could be a reliable tool for identifying future risk of cardiac abnormalities in IUGR children.

Abdulrhman Wassil Al-Jedaani, Wajid Aziz, Abdulrahman A. Alshdadi, Mohammed Alqarni, Malik Sajjad Ahmed Nadeem, Mike P. Wailoo, Fernando S. Schlindwein

Predicting Student’s Performance Based on Cloud Computing

COVID-19 Coronavirus epidemic has created a calamitous worldwide situation. The Moroccan University Sidi Mohammed Ben Abdellah of Fez mobilized to develop platforms for distance learning. New tools for e-learning (online learning) were developed, and advanced learning management systems (LMSs) were deployed. Predicting student's performance is more difficult because of the large amount of data including the huge number of learners and the educational content variety. Currently, in the University of Fez, the lack of current mechanisms to assess and control the development and performance of the students is not discussed. In this context, Cloud computing is becoming a hot research subject when faced with large-scale data and is commonly used to solve this issue. Big Data is a leading concept because of its permanent optimization and opportunities offering data collection, analysis, storage, optimization, processing, and data representation to e-learning professionals. The objective of this paper is to develop a model of predicting student’s performance based on cloud computing as part of the normal enhancement of online learning by incorporating new information and communication technologies.

Youssef Jedidi, Abdelali Ibriz, Mohamed Benslimane, Mehdi Tmimi, Mounia Rahhali

Contribution to the Optimization of Industrial Energy Efficiency by Intelligent Predictive Maintenance Tools Case of an Industrial System Unbalance

Today's industry presents many challenges whose the competitiveness weighs heavily on productivity. The future industry or industry 4.0 requires a new way for organizing industrial processes and must integrate smarter maintenance tools capable of greater adaptability in production. This new organization must respond to competitiveness challenges to achieve customer expectations but with a short deadline to market and an optimized cost production in terms of energy consumed reduced breakdowns, etc. One of the failures encountered in the industry, object of our study, is the unbalance corresponds to a rotor imbalance, shaft … due to the non-coincidence of the principal axe of inertia and the inertia center with the rotation axis. Our contribution is to develop the main components surveillance of an industrial installation continuously and follow the evolution through quantifiable and qualifiable data which allows preventing a dysfunction before stopping the production. This surveillance uses very precise predictive maintenance technologies and can tracks parameters in real time: vibration, consumed energy and the various components temperature.

Ali Elkihel, Yosra Elkihel, Amar Bakdid, Hassan Gziri, Imane Derouiche

Automobile Insurance Claims Auditing: A Comprehensive Survey on Handling Awry Datasets

Soufiane, Ezzaim EL Baghdadi, Salah-Eddine Berrahou, Aissam Mesbah, Abderrahim Berbia, Hassan Fraud is a very costly criminal activity. Insurance companies face the very challenging task of identifying and preventing fraudulent claims. Just like any big problem in recent years, Machine Learning has been heavily applied to fraud detection in both a supervised and non-supervised manner. But, usually supervised models do not perform well in the presence of awry, asymmetrical Datasets. This paper presents a novel approach for auditing claims in automobile insurance. Our data pipeline consists of preprocessing, feature selection, data balancing, and classification. This robust fraud detection model, built upon existing fraud detection research, gives very promising results compared to state of the art in the industry.

Ezzaim Soufiane, Salah-Eddine EL Baghdadi, Aissam Berrahou, Abderrahim Mesbah, Hassan Berbia

Artificial Intelligence Based on the Neurons Networks at the Service Predictive Bearing

In the industrial environment, production systems are increasingly complex and cannot be free from disturbances and failures. Indeed, the following study is considered as a point of change in the service domain to effectively track disturbances and failures, by allowing the transition from old maintenance to smart maintenance. However, the following document represents a sort of passage between the old and the new maintenance by treating the operation of the bearings in the rotating mechanical systems, the study consists in studying the modes of failures of the bearings. A prediction model is developed based on neural networks.

Ali Elkihel, Imane Derouiche, Yosra Elkihel, Amar Bakdid, Hassan Gziri

Intersection Management Approach based on Multi-agent System

For several decades, urban congestion causes various problems such us pollution, road wares, and congestion in intersections which deteriorates the quality of life of citizens who live in big cities. Different methods proposed to reduce urban congestion, notably traffic regulation that attend tremendous attention recently. In past years, the usage of tools from artificial intelligence, particularly distributed methods and multi-agent systems, which allow to design new methods for traffic regulation. In this context, a Multi-Agent approach for intersection management system based on the principle of trajectory reservation has been proposed to reduce the travel time average and air pollution.

Meryem Mesbah, Ali Yahyaouy, My Abdelouahed Sabri

A Model of an Integrated Educational Management Information System to Support Educational Planning and Decision Making: A Moroccan Case

The planning of education in Morocco represents an essential element in the projects implementation of the educational system, on which rests the various operations of diagnostics, realization and evaluation of the educational strategic choices. The planning profession has benefited very well from technological advance, and the country has been in the process of automating information systems for a long time. But according to our analysis, the education information system will be able to be more effective if it can adopt the techniques proposed, especially with regard to the establishment of an Integrated Information System (IIS) which groups operational systems, then use a Decision Support System (DSS) to help decision makers. As well as an Early Warning System (EWS) to predict problems, and a Recommendation System (RS) to propose realistic and effective measures. The unification of such systems will improve both the quality of the educational data management and the educational administration processes.

Mustapha Skittou, Mohamed Merrouchi, Taoufiq Gadi

Variational Autoencoders Versus Denoising Autoencoders for Recommendations

Recommender systems help users explore new content such as music and news by showing them what they will find potentially interesting. There are many methods and algorithms that can help recommender systems create personalized recommendations. All recommendation approaches can be divided into three categories: Content-based recommendation, Collaborative filtering and Hybrid methods. In this paper, we explore and compare Variational Autoencoders and Denoising Autoencoders for Collaborative Filtering with implicit feedback. A Variational Autoencoders(VAE) is a non-linear model, so it can capture patterns that are more complex in the data and since the forward pass is sufficient to obtain the recommendation of a given user then the query time is fast. A Denoising AutoEncoder is a specific type of AutoEncoder, which is generally classed as a type of deep neural network and is trained to use a hidden layer to reconstruct a particular model based on its inputs. Comparison results between Variational Autoencoders (VAE) and Denoising Autoencoders (DAE) show that VAE has the upper hand when it comes to large datasets while DAE is better when using small datasets. We explore and evaluate both methods on three public datasets and using different metrics.

Toward Moroccan Virtual University: Technical Proposal

Due to the Corona virus (COVID-19) and in order to ensure education and school continuity, universities around the world; including Moroccan ones; have moved abruptly and integrally to distance learning mode due to the confinement decision. Unfortunately, this transition occurred in an abrupt manner without the universities being sufficiently prepared for it. In this context, many countries around the world become motivated to virtual university projects, which will undoubtedly make it possible to overcome, without any difficulties, some constraints, related to the lack of qualified human resources, cost and time limitation. This concept presents the same advantages of the e-learning one but in an official form, that may target a large number of students around the world, which respects the famous slogan “Education for everyone”. In this paper, we will propose a Moroccan virtual university architecture, followed by an overview of technology proposal that could help in the implementation and data treatments steps.

Ayoub Korchi, Sarah Benjelloun, Mohamed El Mehdi El Aissi, Mohamed Karim Khachouch, Nisrine El Marzouki, Younes Lakhrissi

Data Lake Versus Data Warehouse Architecture: A Comparative Study

Each day huge quantities of data are generated from digital technologies and information systems. Therefore, processing these massive data requires a specific architecture and a good knowledge on how to handle data. Traditional databases management system can no longer be used for this type of data since they were originally designed for limited and structured data. Moreover, dedicated architecture known as Data Lake has been developed in order to extract valuable information hidden in data. The main objective of this paper is to explore the two architectures, namely, data warehouse and data lake. Furthermore, it describes the main differences and exposes key factors of each one.

Mohamed El Mehdi El Aissi, Sarah Benjelloun, Yassine Loukili, Younes Lakhrissi, Abdessamad El Boushaki, Hiba Chougrad, Safae Elhaj Ben Ali

Machine Learning for Credit Card Fraud Detection

Moumeni, Loubna Saber, Mohammed Slimani, Ilham Elfarissi, Ilhame Bougroun, ZinebE-commerce and many other online sites have increased the online payment modes, increasing the risk for online frauds. Many errors are lost due to fraudulent card transactions each year. The development of performance fraud detection methods is obligatory to minimize such losses. This article examines the usefulness of applying different learning approaches for detecting credit card fraud. Three algorithms will be applied on a database of an American bank and the data will be exploited based on supervised and unsupervised learning techniques, namely the MLP (multilayer perceptron), LR (logistic regression) and PCA (Principal Component Analysis). The main purpose of the study is comparing the classification performance of each algorithm using real dataset of fraudulent user accounts in a telecommunication network.

Loubna Moumeni, Mohammed Saber, Ilham Slimani, Ilhame Elfarissi, Zineb Bougroun

OctaNLP: A Benchmark for Evaluating Multitask Generalization of Transformer-Based Pre-trained Language Models

Kaddari, Zakaria Mellah, Youssef Berrich, Jamal Belkasmi, Mohammed G. Bouchentouf, Toumi In the last decade, deep learning based Natural Language Processing (NLP) models achieved remarkable performance on the majority of NLP tasks, especially, in machine translation, question answering and dialogue. NLP language models shifted from uncontextualized vector space models like word2vec and Glove in 2013, and 2014, to contextualized LSTM-based model like ELMO and ULMFit in 2018, to contextualized transformer-based models like BERT. Transformer-based language models are already trained to perform very well on individual NLP tasks. However, when applied to many tasks simultaneously, their performance drops considerably. In this paper, we overview NLP evaluation metrics, multitask benchmarks, and the recent transformer-based language models. We discuss the limitations of the current multitask benchmarks, and we propose our octaNLP benchmark for comparing the generalization capabilities of the transformer-based pre-trained language models on multiple downstream NLP tasks simultaneously.

Zakaria Kaddari, Youssef Mellah, Jamal Berrich, Mohammed G. Belkasmi, Toumi Bouchentouf

Comparative Study of Regression and Regularization Methods: Application to Weather and Climate Data

Raouhi, El Mehdi Lachgar, Mohamed Kartit, Ali Regression analysis is a powerful statistical method that support to inspect the relationship between two or more variables of interest. While there are many types of regression analysis, at their core they all examine the impact of one or more independent variables on a dependent variable. It is one of the most commonly used methods in many scientific fields. Satisfying the assumptions such as collinearity between variables ought to be a significant issue in data science. Advanced level tools such as Linear, Lasso, Ridge and ElasticNet regression are methods designed to overcoming a problem of overfitting a model. This study discusses comparing regression and regularization algorithms. It also deals with how the concept of model complexity unfolds for each of these models and provides an overview of how each algorithm builds a model. Moreover, it examines the strengths and weaknesses of each algorithm, as well as the type of data to which they can best be applied to irrigation water use efficiency under climate change. Finally, this work aims also to explain the meaning of the most important regularization criteria. It remains to say that the main contributions of this study are (1) Comparing linear and multilinear regression methods: case of climate change dataset using regression metrics (2) comparing regularization methods: Ridge, Lasso and ElasticNet.

El Mehdi Raouhi, Mohamed Lachgar, Ali Kartit

Big Data Architecture for Moroccan Water Stakeholders: Proposal and Perception

The water data are produced from synchronous and/or asynchronous observations of water bodies. The aim of these observations is to support water resource management. The water entities and concepts give meaning to all collected data. These entities are heterogeneous, with a wealth of attributes and very varied functions. In this paper we will present the steps followed for the characterization of the entities for water resources in Morocco. From trip works, interviews with experts and observations with Moroccan stakeholders. Then in a second step, we will propose a vision on a big data architecture for water resources in Morocco, which can be used by the public administrations.

Towards an Integrated Platform for the Presentation and Preservation of the Scientific Heritage of Drâa-Tafilalet

The region of Drâa-Tafilalet in the south-east of Morocco is known for its unique and varied tourist qualifications, combining the greenery of the oasis with the quietness of the desert and its deep-rooted heritage, which has made it one of the best known tourist poles at the national and international levels. The region has benefited from national programs aimed at promoting the tourism sector and restoring the rich and unique local heritage. This region has benefited from national programs aimed at promoting the tourism sector and restoring the rich and unique local heritage. The main objective of this article is to present the heritage of the Drâa Tafilalet region with emphasis on the diverse and rich scientific heritage in the region. This paper presents the latest results obtained in this field, based on applications, technologies, etc., aimed at preserving this heritage from extinction and presenting it to future generations.

Keratoconus Classification Using Machine Learning

The diagnosis of several ophthalmic diseases such as age-related macular degeneration, glaucoma, diabetic retinopathy and keratoconus involves the analysis of the eye topographic maps. The dependence between ophthalmology and images processing represents a point of attraction for researchers to benefit of capacity and performance of deep learning tools in image processing. These tools allow a better differentiation between a sick eye and a normal one based on the analysis of the eye topographic maps and can change potentially the practices of ophthalmologists in diagnosis and treatment of similar diseases. Among the diseases already mentioned, keratoconus, this non-inflammatory disease characterized by a progressive thinning of the cornea is often accompanied by aspens of vision. The increasing number of people diagnosed with keratoconus has made this disease the subject of several research studies.This paper represents an overview of artificial intelligence application in keratoconus classification and a proposal system of keratoconus classification based on neural networks.

Aatila Mustapha, Lachgar Mohamed, Kartit Ali

0.18 μm GaAs-pHEMT MMIC Frequency Doubler for Radar Area Scanning Application

This paper proposes a GaAs-pHEMT MMIC frequency doubler for 60 GHz radar area scanning application using 0.18 μm GaAs technology. The aim of this study is to enter a frequency f0 = 30 GHz and to recover at the output a frequency of 2*f0, i.e. 60 GHz. The proposed multiplier is designed and optimized thanks to OMMIC library. The latter gives to the proposed frequency doubler added values in term of circuit performances. Indeed, it has low power consumption Pdc = 0.126 mW, high spectral purity and small size; it occupies an area of (1.35 × 0.63mm2) and achieves a conversion gain of −2.189 dB and an output power of 11.376 dBm. The results show a total efficiency of around 13.87%.

H. El Ftouh, Moustapha El Bakkali, Naima Amar Touhami, A. Zakriti

Fail-Safe Remote Update Method for an FPGA-Based On-Board Computer System

As part of a university project to design a Low Earth Orbit (LEO) nano-satellite payload, we investigate a System-on-Chip (SoC) solution exploiting the features of Xilinx's Spartan 6 FPGA technology to design an On-Board Computer System (OBC). Thus, the increased flexibility of the FPGA implementation will enable on-orbit updates and modifications to the software and hardware OBC architecture, in lodge to support dynamic mission requirements. Within this context, this paper introduces a method to safely remote update an FPGA-based embedded system. The proposed architecture is based on the Xilinx soft processor, i.e. the Microblaze, which controls the remote update channel (Ethernet in our case) to upload hardware and/or software application images in the system by using the Trivial File Transfer Protocol (TFTP). An on-board flash memory is used to store FPGA Hardware and firmware images. The soft processor is implemented in the Spartan-6 XC6SLX45 FPGA device and uses the fallback features and the Internal Configuration Access Port (ICAP) primitive in order to manage fail-safe FPGA reconfiguration to maintain safe and stable state after updates.

Ahmed Hanafi, Mohammed Karim, Tajjeeddine Rachidi, Ibtissam Latachi

Autonomous Vehicle Lateral Control for the Lane-change Maneuver

During the fourth industrial revolution, the automotive industry is seeking to develop a completely autonomous ground vehicle capable of adapting to all situations encountered; including the lane change maneuver during normal driving on the highway. This paper describes, at first, the planning of the lane change maneuver using a quintic mathematic function, as a second step; a law of control based on higher-order sliding mode is applied. The adopted strategy achieves good results in terms of sideslip angle and tracking error, especially with the consideration of a high longitudinal speed.

Lhoussain El Hajjami, El Mehdi Mellouli, Mohammed Berrada

Integral Sliding Mode Control of Power Transfer in a Vehicle to Grid (V2G) Charging Station

Numerous overhauls are carried out to the electrical system in the last few years striving for more flexible and stable electric grid. Still, due to the continuous rise in demand in addition to the integration of new heavy loads such as EVs charging stations, this task is becoming more challenging. The conventional solutions include hydro storage stations and spinning reserves, be that as it may, the high cost of these solutions limits their use. As an alternative solution, vehicle to grid (V2G) offers the possibility to use the EV storage units to provide ancillary services to the grid. In this paper we propose a robust integral sliding mode (ISM) approach, to control the bidirectional power transfer in the V2G charging station. The proposed ISMC, controls the current flow in the bidirectional DC-DC converter, based on a reference signal derived from the grid power requirements. The robustness of the control scheme and its capability to respond to the desired performances were tested using different realistic scenarios. The obtained results demonstrated the effectiveness, stability, and tracking performances of the proposed controller compared to the conventional PID.

Hicham Ben Sassi, Chakib Alaoui, Fatima Errahimi, Najia Es-Sbai

Design and Analysis of an Integrated Class-D Power Output Stage in a 130 nm SOI-BCD Technology

In this work, a design and analysis of an integrated class-D power output stage in 130 nm SOI-BCD technology is will be described. An output stage with two NMOS transistors is used for economize cost and area. A bootstrap circuit is important to affording the gate overdrive voltage of the NMOS transistor, especially in tall-current gate drivers with great transistors, it is mush large for integration. The proposed class-D power output stage utilizes an on-chip bootstrap circuit with integrated bootstrap capacitor. The class-D power output stage achieving a total root-mean-square (RMS) output power of 0.2 W, a THD + N (total harmonic distortion + noise) at the 8-Ω load less than 0.06%, and a power efficiency of 93%. The final design occupies approximately 1.25 mm2.

Mustapha El Alaoui, Karim El khadiri, Ahmed Tahiri, Hassan Qjidaa

Digital Implementation of SPWM 7-Level Inverter Using Microcontroller

Chadli, Hajar Bikrat, Youssef Chadli, Sara Saber, Mohammed Fakir, Amine Tahani, Abdelwahed Currently, green energy is knowing a massive growth in the world with the growth of newer energy sources such as wind energy, hydro energy, tidal energy geothermal energy, biomass energy and of Corse the Solar energy which is considered the second biggest source of electricity worldwide including morocco. The production of electricity via these centrals requires optimization at the different conversion levels. To obtain electricity that meets the standards of the electrical grid (sine wave of frequency 50 Hz), the inverter remains the first element to design and build. The structures based on multi-level inverters have brought an undeniable advantage to alternative continuous conversion, especially in high power applications. In this article a new 7-level inverter architecture with only six switches is presented and compared along with the other seven level inverter topologies. To improve the performance of our proposed multilevel inverter, we used a digital sinusoidal Pulse Width Modulation (SPWM) strategy using the Arduino wich leads to further reduction of THD. In this paper, the inverter was tested using Proteus software and Matlab Simulink simulator for harmonic analysis. Then real-time implementation of inverter was tested for a resistive load.

Hajar Chadli, Youssef Bikrat, Sara Chadli, Mohammed Saber, Amine Fakir, Abdelwahed Tahani

Embedded and Parallel Implementation of the Stereo-Vision System for the Autonomous Vehicle

Stereo-vision is the most widely used technique in the development of environmental perception systems for intelligent transportation. The main requirement for the application of stereo vision on a vehicle is the processing time which must be very fast for autonomous driving in real time, whereas the computation of the correspondence of the images in the algorithm of stereo-vision requires a more computing power. This article presents an implementation of the stereo vision system to generate a scene disparity map using the sum of absolute differences (SAD), and the triangulation method for calculating the distance between the obstacle and the stereoscopic system. A parallel treatment is used to speed data processing and this algorithm is implemented in embedded platform. Several experiments are performed from a profiling analysis to have a statistical analysis for optimization.

Mohamed Sejai, Anass Mansouri, Saad Bennani Dosse, Yassine Ruichek

An Efficient Implementation of an Effective PFD-CP for Low Power Low-Jitter CP-PLL

Zouaq, Karim Bouyahyaoui, Abdelmalik Aitoumeri, Abdelhamid Alami, MustaphaA new efficient Phase-frequency Detector (PFD) paired with a new Charge-Pump (CP) is presented in this paper. This PFD-CP topology uses minimum sized devices of optimal minimum energy. The proposed PFD uses two pre-charged logic stages followed by static CMOS inverters without reset path cell or additional delay cells which leads to having a minimum blind zone, removing dead zone and occupying small chip area. Besides, the proposed design topology allows lowering power dissipation and improving speed, and can be used in high-speed CP-PLL applied in RF communication systems. Also, the paper presents an effective CP design for low-power and wide Voltage Controlled Oscillator (VCO) control voltage swing, that solely uses two CMOS inverters for charging and discharging processes, and therefore can form an effective PFD-CP topology suitable for applications where performance is needed, meanwhile minimizing energy consumption is key. Simulation results of the implemented circuit are discussed accordingly.

Karim Zouaq, Abdelmalik Bouyahyaoui, Abdelhamid Aitoumeri, Mustapha Alami

Adaptive Fast Terminal Sliding Mode Control for Uncertain Quadrotor Based on Butterfly Optimization Algorithm (BOA)

This paper proposes a robust Adaptive Fast Terminal Sliding Mode Control (AFTSMC) for the quadrotor UAV in the presence of external perturbations. First, the quadrotor position and attitude dynamics are obtained using the Newton–Euler principle, then the closed-loop stability of the suggested controller is proved by the Lyapunov theorem. The parameters of the proposed AFTSMC were automatically selected via the novel Butterfly Optimization Algorithm (BOA). Simulation experiments are conducted in Matlab/Simulink and comparison with other controllers is performed to validate the effectiveness of our new control strategy. The obtained results demonstrate that the proposed controller possesses multiple salient advantages such as fast response, high accuracy, and strong immunity to the variations across system model parameters.

Hamid Hassani, Anass Mansouri, Ali Ahaitouf

Localization and Navigation System for Blind Persons Using Stereo Vision and a GIS

Loss of vision caused by infectious diseases has decreased significantly; however aging will increase the risk that more people acquire vision impairment. Visual information is the basis of most navigation tasks; a person is considered visually impaired when he has no appropriate information on the surrounding environment. With the latest evolution of digital technologies, the assistance provided to visually disabled people during their mobility can be improved. In this context, we propose a system to help the visually impaired move quickly and to know their environment. Indoors, the system uses a stereoscopic camera, a portable computer, and a headset to direct and help visually impaired persons navigate comfortably and securely in familiar and unfamiliar environments. Outdoors, a GPS is used as a positioning method to keep the visually impaired person on the right path; with its dynamic routing and rerouting capabilities, it provides the user with an optimized path. The system can work on an outdoor and indoor environment. A stereoscopic camera is used to detect visual indicators that are used to trace and validate user navigation, provide accurate indoor location measurements, and recognize objects in front of the user. This article is mainly focused on this system and detailed outlines description.

Moncef Aharchi, M.’hamed Ait Kbir

New Delay Dependent Stability Condition for a Carbon Dioxide Takagi Sugeno Model

Elmajidi, Azeddine Elmazoudi, Elhoussine Elalami, Jamila Elalami, NoureddineThe growing interest in the preservation of environment leads several researchers to investigate the causes behind the high level of $${\text {CO}}_2$$ CO 2 and how to decrease it. This paper, deals with continuous time delay nonlinear systems (TDNS) stability conditions using the Takagi Sugeno Fuzzy Modeling. First a Nonlinear Carbon Dioxide Model is defined and transformed to a corresponding Fuzzy Takagi Sugeno (TS) multi-model. Then, by using the Lyapunov-Krasovskii Functionals (LK-F) and extending some linear time delay systems dependent delay stability technique to TS Fuzzy Modeling, a new relaxed stability conditions involving uncommon free matrices are addressed in Linear Matrix Inequalities (LMI). Finally a numerical simulation is also carried out to support the analytic results and to compare the conservativeness of the proven condition to other existing methods.

Azeddine Elmajidi, Elhoussine Elmazoudi, Jamila Elalami, Noureddine Elalami

Simulation-Based Optimization for Automated Design of Analog/RF Circuits

Automation tools for circuit optimization have proven their usefulness in solving design issues by considering the technological aspects of downscaling. Recent advances have proven that the optimization method based on simulation is a powerful and important solution for the optimal sizing of electronic circuits. In this paper, we propose a simulation-based methodology for automatic optimization of the multi-objective design of an analog/RF circuit. As applications, we use both analog and RF circuits, respectively the LC tank Voltage Controlled Oscillator (VCO) and the new Current-Feedback Operational Amplifier (CFOA). For the LC-VCO, we optimize the power consumption and the phase noise. For the CFOA, we optimize its important performances such as bandwidth and parasitic resistances, for low-voltage, low-power applications. All simulations are performed by HSPICE using 0.13 µm RF CMOS and 0.18 µm CMOS technologies for the LC-VCO and the CFOA, respectively.

Abdelaziz Lberni, Amin Sallem, Malika Alami Marktani, Abdelaziz Ahaitouf, Nouri Masmoudi, Ali Ahaitouf

Readout System of Piezoelectric Sensor Used for High Speed Weigh in Motion Application

The piezoelectric sensors are the main electronic devices widely used to measure the instantaneous force that is exerted on the sensor surface by vehicle wheels moving at normal speed. This application is typically implemented in High Speed Weigh in Motion system, which has been until today under research studies. The system aims to perform an accurate and reliable axle load measurement of heavy trucks in motion. The piezoelectric sensor produces a voltage when a strain is applied on its surface. In general, the voltage signal is in a pulse form as a deterministic signal, easy to operate, but often it is not so. This problem is due to more signal noise, for the most part, because of either the modest quality of piezoelectric sensor or the inefficiency of sensor signal conditioning stage design. The present paper introduces a newly readout system allowing the processing of the piezoelectric sensor output signal in response of exerted force. The studied system consists on the charge amplifier, which is coupled with a transistor used in decoupled common emitter. This circuitry scheme cancels noise and random effect on sensor signal. The purpose of the present study focuses on the achievement of a reliable and deterministic signal, using a sensor with modest quality. The experiment results show the efficiency of the proposed readout system. Even though more noises affect the piezoelectric sensor output signal, the proposed electric interface provides a deterministic and suitable analog signal offering more possibilities to process with the standard tools of signal processing.

Lhoussaine Oubrich, Mohammed Ouassaid, Mohammed Maaroufi

Towards the Implementation of Smartphone-Based Self-testing of COVID-19 Using AI

The new type of coronavirus COVID-19 has widely spread in most of the world and causing a pandemic according to the World Health Organization (WHO). Many mechanisms to detect the coronavirus disease COVID-19 are used like clinical analysis of chest CT scan images and blood test results. Several methods can be used to detect the presence of Covid-19 such as medical detection Kits. Though, such devices require huge costs and it takes time to install and use. In addition to that, the negatively diagnosed patients consume many of the needed resources and space in hospitals that could be used by other patients having higher chance of being infected. In this paper, a new framework is proposed to detect COVID-19 using built-in smartphone sensors. The proposal provides a low-cost solution, since most of radiologists have already held smartphones for different daily purposes. Not only that but also ordinary people can use the framework on their smartphones for the virus detection purposes. Nowadays, almost every household possesses at least one smartphone with powerful processors and advanced sensors. By combining the data collected by the various sensors, such as temperature data, coughing and breathing recording with questionnaires about the background of the phone user, artificial intelligence (AI) and advanced signal processing tools may analyze the recorded data in order to produce viable diagnosis of Covid-19 infection, and hence alleviate the ongoing pressure on the health system (hospitals and stuff).

Hajar Saikouk, Chakib Alaoui, Achraf Berrajaa

Design and Prototyping of an Embedded Controller Board for PV-EV Charging Station

This paper aims to present the design and realization of a fully embedded board, able to execute all the optimization, control and energy management algorithms developed in photovoltaic-electric vehicle charging stations (PVCS). It is intended to provide connectivity and interconnection between the different components of PVCS. This solution supports wired and wireless industrial and advanced computer communication protocols. It is based on the ESP32 Microcontroller, one of the most powerful microcontrollers in terms of performance, low energy consumption and cost, which makes the charging system smart, connected and simple to implement.The functions of the prototyped board are tested and a demonstration of the supervision capabilities in a small-scale PV system proved a good performance.

New Approach for Controlling PTW Vehicle Dynamics: Characterization of Critical Scenarios

At an exponential pace, the interest of ITS (intelligent transportation systems) solutions in the field of mobility has been increasing over the last two decades. However, progress in the field of driver assistance systems is unevenly distributed on the various modes of transport, with a modest development or at least not in terms of the vulnerability of users of Two Wheels. By approaching the mode of transport abbreviated in PTW (Powered Two-Wheeler), the study was devoted to bring out from the literature the different representative models as well as the different technologies known to date. In this paper, a dynamic model of a PTW vehicle based on the kinematic equations is developed and simulated using MATLAB Software. Also, a new approach based on Different scenarios of the PTW vehicle have been analyzed and discussed in both longitudinal and lateral mode. First, we have tackled the problem of collision by proposing a model of obstacle detection in longitudinal mode, then we used the NHTSA (National Highway Traffic Safety Administration) lateral dynamics model to determine the maximum cornering speed by studying the influence of road parameters. After analyzing the results, we find the developed dynamic model give a safe speed profiles suitable for two-wheeled vehicles.

Fakhreddine Jalti, Bekkay Hajji, Abderrahim Mbarki

Study of Parameters Influencing on the Performance of SiNW ISFET Sensor

In recent years, Silicon nanowires (SiNW)-based ISFET devices have been selected as promising better sensors because of their advantages such as real-time detection and the good sensitivity caused by high surface-to-volume ratio. In this work, a mathematical model is evaluated to study the performance of the SiNW ISFET sensor. Furthermore, the effects of the parameters such as the NW width (Wtop) and the gate insulator capacity COX on the performance of the SiNW ISFET sensor are investigated. The planar-ISFET structure and Si-nw-ISFET technology are also examined in our simulation and compared. Finally, the mathematical model of SiNW ISFET is verified with the experimental measurements and with other models indicated in the literature, gives a good accuracy at different pH values.

Nabil Ayadi, Bekkay Hajji, Abdelghafour Galadi, Ahmet Lale, Jerome Launay, Pierre Temple-Boyer

Modeling and Trajectory Tracking of an Unmanned Quadrotor Using Optimal PID Controller

Quadrotor Unmanned Aerial Vehicle (UAV) is a nonlinear, unstable, and coupled system which presents a great challenge in term of modeling and control design. In some cases, the nonlinear part of the quadrotor system is omitted to simplify the study and the control design, our goal is to derive the full dynamic model without neglecting the coupling influences between the translational and rotational dynamics. This paper conducted a detailed theoretical study of the dynamics and kinematics equations of the aerial vehicle based on the Newton–Euler formula. Then, a Proportional-Integral-Derivative (PID) controller is used to steer the quadrotor degrees of freedom. Ant Colony Optimization (ACO) algorithm is used to select the optimal configuration of the PID controller. Simulations for trajectory tracking are conducted to test the performance of the adopted PID controller.

Hamid Hassani, Anass Mansouri, Ali Ahaitouf

Dynamic Socket Design for Transtibial Prosthesis

Within the industry dedicated to build and adapt orthoses and prostheses, there has been a growing number of users or patients who require the use of prosthesis, for the replacement of one or several limbs (either lower or upper); for this reason, the innovation in devices and materials appears as a current need for the benefit of patients. Either for the manufacturing or adaptation of prosthetic components, along the experience, it has been evidenced that, in order to treat an amputee, the selection of the components to be used in the prosthesis is not the only aspect to take into account; consequently, one of the paramount factors is the proper management of adaptation The component with the highest relevance in the adaptation of a prosthesis is the socket. This device has been specially made in accordance with the anthropometric measures of the patient, and built using rigid, lasting compounds However, an issue arises regarding the moment when the patient, along adaptation or rehabilitation, changes his/her physical condition or stump. The previous fitting loosens and clearly depends on the suspension system to stay in position; on the other hand, if the patient’s body mass increases rather than decreases, the previously manufactured socket will not allow the use of said prosthesis Because of this, an unexplored research niche has been detected, corresponding to the design and manufacture of dynamic sockets, which can adapt to the user, depending on the change in morphology Another great field of action is the use of dynamic sockets in children.

Jhon Hernández Martin, Alejandra Santos Borraez, Catalina Ríos Bustos, Fran Pérez Ortiz, Phillip Meziath Castro

Towards an Enhanced Minimum Rank Hysteresis Objective Function for RPL IoT Routing Protocol

Hassani, Abdelhadi Eloudrhiri Sahel, Aa Badri, AbdelmajidRPL is designed as a routing protocol dedicated to Internet of Things. RPL is based on an objective function to build a Destination Oriented Directed Acyclic Graph that connects the leaf nodes in the wireless sensor network to the central node which collects all the informations. In its path selection, RPL does not offer the best network quality of service due to the unoptimized standard metrics. In this paper, we propose an improvement of Minimum Rank Hysteresis Objective function called E-MRHOF based on a new method of calculating the link metric. The simulations show that E-MRHOF can increase the number of packets delivered to the sink with low latency and power consumption, while offering a network convergence time and ICMPv6 packets inferior than the classical MRHOF based on ETX and Energy.

A Lightweight Hash Function for Cryptographic and Pseudo-Cryptographic Applications

In this paper, we design a lightweight hash function suitable for fast applications. The proposed hash function is intended to be used as a secondary component in cryptographic and pseudo-cryptographic applications (e.g., pseudorandom number generators, cryptosystems), where the required property is the fastness in addition to an acceptable security level. Based on simple logical operations used in designing primitive functions, and based on a Feistel-like network, the proposed hash function exhibits excellent performance (fastness) and good enough security properties.

Hybrid Intrusion Detection System for Wireless Networks

Local wireless networks (WLAN) are vulnerable to various types of security threats ranging from session hijacking to denial of service (DoS), and password attacks, to name a few. They are also subject to a wide range of 802.11-specific threats. The risks can become even higher and more serious when the WLAN network is made up of a number of IoT objects. As a remedy these failures, an intrusion prevention system (WIPS) has been on the network. However, the breadth of the network, the diversity of the elements to be secured and the approaches to be adopted make this integration sometimes complicated or ineffective in certain types of WLAN network. The main concern in this document is to develop, on the basis of free solutions, a flexible, easy-to-deploy and manage WIPS system that provides both intrusion detection and flow monitoring to reduce the rate of false positives, especially during home deployment or on small-scale networks.

Mohamed Amine Agalit, Ali Sadiqui, Youness Khamlichi, El Mostapha Chakir

Implementation and QoS Evaluation of Geographical Location-Based Routing Protocols in Vehicular Ad-Hoc Networks

Over the last few decades, Vehicular Ad hoc Networks (VANETs) in which vehicles communicate with each other in a high dynamic topology and a high speed, have attracted considerable attention. In this type of application, routing needs to be carefully designed to handle rapid network changes and therefore geographic routing protocols have been implemented to resolve this issue. The Greedy Perimeter Stateless Routing (GPSR) protocol is one of the most promising location-based routing protocols for wireless networks. In this paper, we implement three geographical routing protocols defined as Maxduration-Minangle GPSR (MM-GPSR), Modified GPSR (MGPSR) and traditional GPSR with different approaches for node selection in VANET. We carry out an overall performance evaluation with different levels of traffics in an urban environment. The tool-set of simulation integrates NS2, Intelligent Driver Model with Lane Change (IDM_LC) based on VANETMOBISIM. The paper provides insights the choice of a geographic protocol for Quality of Service (QoS) metrics and weight value that can improve robustness in VANET.

Safae Smiri, Abdelali Boushaba, Adil Ben Abbou, Azeddine Zahi, Rachid Ben Abbou

Congestion Control Management in High Speed Networks

High-speed networks are a key technology which has brought very innovative solutions to the world of computer networks while retaining mobility, scalability, reliability… It provides a very efficient service in terms of signal speed and optimality. However, such networks with such performance needs robust systems capable of adapting to the different changes that a network connection can experience, beyond this was born the need to design congestion control algorithms capable of preserving network performance while providing high speed with low latency and good signal quality. By the way, analyzing these algorithms is the main purpose of this paper.

Kaoutar Bazi, Bouchaib Nassereddine

Aerodynamic Analysis of Wind Turbine Blade of NACA 0006 Using a CFD Approach

In order to exploit wind energy and make it more and more promising and competitive with other forms of green energy, designers and researchers aim to optimize the aerodynamic performance of wind turbine blades. Our research project deals with the case of the aerodynamic profile of type NACA 0006 in order to evaluate the aerodynamic forces which act on a single blade. This article analyzes the two-dimensional model of the aerodynamic profile NACA 0006 for a specific Reynolds number as well as a range of angles of attack of [−10.5°; 10.5°]. The aerodynamic characteristics simulation is applied by the k − ω turbulence model (SST) namely; coefficient of lift, drag and pressure. That with these results it was possible to carry out the comparison of the numerical measurements with those experimental for the chosen profile of national advisory committee of aeronautics NACA in the wind tunnel. The purpose of our application is to verify the accuracy of the model and to predict the efficiency of the blade which will allow the achievement of the efficient design strategy.

Mohamed Hatim Ouahabi, Houda El Khachine, Farid Benabdelouahab

Educational Strategy Combining Technological Capacity and Ant Colony Algorithm to Improve the Ideal Dispatch Using Wind Energy

In this article, the reader will find an educational strategy, combining technological capacity and ant colonies to improve the ideal dispatch with plants of solid generation (hydraulic and thermal) and plants with variable generation (wind), as a case study for a Colombian electricity network. In order to achieve this process, we intend to use problem-based learning strategy as follows: posing the problem, listing the known data, dividing the main problem in analog form, looking for and implementing solutions, and consequently looking at the technological assets of all plants. This information is registered into the McKinsey matrix and finally, the solutions are analyzed.

Neider Duan Barbosa Castro, Jhon Alexander Hernández Martin, Fabiola Sáenz Blanco, Evy Fernanda Tapias Forero

A New Method for Photovoltaic Parameters Extraction Under Variable Weather Conditions

Hali, Aissa Khlifi, YaminaThis work suggests a new method for extracting parameters from the current-voltage characteristics of photovoltaic (PV) panel to establish translational relations that allows to predict these PV parameters in other conditions of temperature and irradiation levels. This method is based on the knowledge of few selected points which are the short-circuit point, the maximum power point, and the open circuit point. These points are extracted from the current-voltage characteristics provided by the manufacturer’s datasheet. Single diode model is chosen to represent the PV panel which is known as five parameters model. It includes two parasitic resistances, diode saturation current, diode ideality factor, and photo-current. The main results of simulations in Matlab environment show that the five PV parameters varies as function of irradiance and temperature. The proposed technique is compared to other published methods applied on Kyocera KC175GHT-2 PV panel. The obtained results show that the new suggested method of predicting PV parameters presents a lower statistical error whatever the weather conditions. Therefore, the new method is validated and more precise than extraction methods reported in the literature.

Aissa Hali, Yamina Khlifi

Applying CFD for the Optimization of the Drying Chamber of an Indirect Solar Dryer

The importance of renewable energy technologies is illustrated by their dependence on inexhaustible sources, and it has a friendly relationship with the environment, because of its several advantages. This type of systems are more attractive to investors as well for researchers in terms of development, In this paper, we are particularly interested in the modern system used for drying products, more precisely the one based on solar energy, like the indirect solar dryer, It generally consists of a solar air collector that generates a flow of air that passes through it and goes up to the drying chamber. In order to make the performance of the chamber more efficient, we try in this work to improve the efficiency of the solar dryer by improving the geometry, with the aim of reducing the drying time while guaranteeing the high quality of the dried product. The geometry of the chamber is divided into two sections: the first where the trays are located and the second one is a side-mounted plenum chamber. The width of this latter as well the distance between the first tray and the rooftop of the drying chamber where studied using ANSYS FLUENT software, under the climate of the oriental region of Morocco. And the results showed that the most suitable width for the side-mounted plenum chamber is 0.4 m with an optimal distance between the first tray (the upper tray) and the top of the chamber of 0.06 m.

Dounia Chaatouf, Mourad Salhi, Benyounes Raillani, Samir Amraqui, Ahmed Mezrhab

Performance Comparison of Regenerative Organic Rankine Cycle Configurations

In order to ameliorate the ORC performances, the Regenerative Organic Rankine Cycle (RORC) is presented. This study presents a comparison of an ORC with an open feedwater heater (ORC-OFWH) an ORC with a closed feedwater heater (ORC-CFWH) and double stage regenerative ORC (DSR-ORC) using both OFWH and CFWH. The thermodynamic performances of each system are investigated adopting R123 as working fluid. Summing up the results, it can be concluded that the double stage regenerative ORC shows greater performances in term of thermal efficiency and exergy efficiency with 14% and 17% respectively higher over the ORC-OFWH and ORC-CFWH.

Rania Zhar, Amine Allouhi, Abdelmajid Jamil, Khadija Lahrech

Performance Analysis of Combined Power and Refrigeration: ORC-VCC System

The paper aims to give a performance examination of a cogeneration system combining power and refrigeration, based on an Organic Rankine Cycle (ORC) as well as Vapour Compression Cycle (VCC). The following working fluids R123, R11 and R601a were assessed energetically and exergetically. The impact of different parameters including the boiler temperature and evaporator temperature are carried out on ORC thermal efficiency, also coefficient of performance (COP) as well as total system efficiency. The exergy efficiency and the amount of the exergy destroyed in each system component were detected. R123 was recommended.

Rania Zhar, Amine Allouhi, Abdelmajid Jamil, Khadija Lahrech

Fault Location Technique Using Distributed Multi Agent-Systems in Smart Grids

Efficient detection of fault and location can increase the safety and Efficiency of electrical power systems. The fault in power systems can create other faults and outages in the network. Fault analysis includes fault detection, location, isolation of the faulted section from the network, and energy restoration as soon as possible. In this study, an innovative multi-agent system approach proposed for fault location in the electrical network. The smart agents collect the power data (Voltage and current at each point) between different power components (sources, bus, relays, loads) for fault location and efficiently control of circuit breakers operations to restore service in the power systems. The simulation results show the performance of the proposed techniques for fault location in the power system.

Mohamed Azeroual, Younes Boujoudar, Tijani Lamhamdi, Hassan EL Moussaoui, Hassane EL Markhi

Hybrid Renewable Energy System Investigation Based on Power Converters Losses

Given the renewable sources complementarity, hybrid renewable energy systems (HRES) have been mostly used to address the limitations of single renewable source in terms of stability and reliability. In this regards many researches have been conducted to improve the hybrid efficiency. The aim of this work is to perform a comparative study of HRES architectures based on a qualitative investigation of the power converters efficiency used in each topology. Numerical application is performed based on technical features of converters in order to distinguish the best HRES configuration.

Ilham Tyass, Omar Bouamrane, Abdelhadi Raihani, Khalifa Mansouri, Tajeddine Khalili

Estimation of Daily Direct Normal Solar Irradiation Using Machine-Learning Methods

The sizing and simulation of all solar systems require the availability of reliable measurements of solar radiation at different time steps. Unfortunately, solar radiation measurements are not readily available for most worldwide locations. For this reason, it is desirable to develop accurate prediction models by developing relationships between available meteorological data and solar irradiation. Artificial Neural Networks (ANN) have been widely used for the estimation of different solar irradiation components. Recently, some machine learning methods have been reported and appear to be very promising. In this paper, we are interested in comparing the performance of ANN and three ensemble methods (Bagging, Boosting and Random Forests) in estimating the daily direct normal (DNI) solar irradiation from some commonly measured meteorological variables. Our study is performed using measurements data from five Moroccan cities: Oujda, Missour, Erfoud, Zagora, and Tan-Tan. The achieved results show that all developed models give good performances on training and validation datasets with a normalized Root Mean Squared Error (nRMSE) < 20%.

Zineb Bounoua, Abdellah Mechaqrane

Greenhouse Design Selection in Moroccan Climatic Conditions

Greenhouse cultivation is one of the most productive agricultural techniques, by its product yield, its quality and its all-around year production. This technique makes it possible to create a suitable microclimate for plant growth. For that, a quantity of energy is required, which raises the production cost. To reduce the costs and the energy needs, numerous greenhouse shapes and different covering materials are available. The selection of the right greenhouse design can significantly decrease the energy needs during heating and cooling periods, therewith sparing a great part of operating cost. This work aims to analyze the impact of conventional and innovative shapes (even span, uneven span, single slope, mansard, modified IARI, quonset, modified quonset, gothic-arch, modified gothic-arch) and covering materials (glass, Low-density polyethylene film (LDPE), Bubbled polyethylene plastic (BPE) and Ethylene vinyl acetate film (EVA)) on the energy needs of the greenhouse, in the climatic conditions of Fez, Morocco. For this purpose, dynamic models of different greenhouse designs are created in EnergyPlus environment.

Laila Ouazzani Chahidi, Abdellah Mechaqrane

Intelligent Architecture in Home Energy Management System for Smart Building, Moroccan Case Study

Managing energy in a building is an optimizing solution for smart homes. The user will be autonomous in managing his need in energy. It is an energy control and management that reacts to MPL (Maximum Power Limit), hybrid system Pv/Battery and temperature of heater. It manages the operation of appliances which are shiftable with respect to the tariffication of electricity, this minimizes the electricity bill. The proposed algorithm receives as input data, the tariffication profile, the power produced by the Pv, the Soc of batteries and the timing and sequencing of turning on the shiftable appliances. The turning on timing is fixed by the user’s priorities. The batteries are charged by the Pv panels only during periods where the cost of electricity is high. It guaranties the thermal comfort at 24 °C and the free usage of all appliances when the hybrid system Pv/Battery is activated. In the contrary if there is a default in the hybrid system which does not respond to the energy requested, it flips the building’s electricity supply to the public grid immediately. In this case the algorithm adjusts the temperature in the range 18–24 °C when the requested power is greater than the MPL and forbids the operation of shiftable appliances in periods where the cost of electricity is high. This will limit the power request to MPL and optimize the electricity bill. The study is realized according to the meteorological conditions of Oujda city and the tariffication of the Moroccan national office of electricity. The electric and thermal models of the building and the hybrid system Pv/Battery are implemented in Matlab/Simulink and the simulation results show that the cost is reduced to 16.41% if the building is uniquely supplied by the grid and to 55.22% for complete system.

Mohammed Dhriyyef, Abdelmalek El Mehdi, Mohammed Elhitmy

Evaluation of Adaptive Backstepping Control Applied to DFIG Wind System Used on the Real Wind Profile of the Dakhla-Morocco City

The conception of a control strategy of Doubly-Fed Induction Generator (DFIG) for providing a high quality of energy, without harmonic accumulations, to the electric network is a real challenge because of the no-linearity of the system and the variable wind speed. In this work, two commands based on two technics have been developed and then tested in order to control the system and the grid powers, respectively. Note that the main purpose of the control laws is to optimize the extraction of the wind energy in a real context and to push the DFIG working properly with the best performance in both static and dynamic modes. Therefore, the dynamic model turbine has been presented in this paper, also the FOC (Field Oriented Control) and the adaptive Backstepping commands are highlighted. Matlab/Simulink simulations analysis, with real parameters of the turbine and real wind profile of Dakhla-Morocco city, confirm the high accuracy of the adaptive Backstepping command, based on the Lyapunov stability technique, with a total harmonic distortion THD ~0.16%.

Comparative Study Between FOSMC and SMC Controllers for DFIG Under the Real Wind Profile of Asilah-Morocco City

This paper describes a comparative study between two advanced nonlinear controls strategies; the Sliding Mode Control (SMC) and the Fractional-Order Sliding Mode Control (FOSMC), in terms of both reactive and active powers to improve the quality of the energy injected into the distribution grid by the wind energy conversion system (WECS). This later is based on the doubly-fed induction generator (DFIG). The objective is to perform modeling and direct control of the (WECS). Firstly, the dynamic modeling of the different parts of the WECS is performed. Then, the second part of this work concentrates on the proposed nonlinear control laws that rely on FOSMC and SMC. Finally, the performance of those strategies has been simulated in the MATLAB/SIMULINK environment using two wind profiles. One of them is a real wind profile of Asilah-Morroco city to test the system robustness and dynamics as opposed to real conditions.

Mohamed Amine Beniss, Hassan El Moussaoui, Tijani Lamhamdi, Hassane El Markhi

Voltage and Power Control for a Grid Tied Single Phase Single Stage Transformer-Less Photovoltaic System Using Sliding Mode Control

This study presents a Sliding mode control (SMC) strategy applied to a single-phase grid connected photovoltaic plant. The proposed controller is designed in such a way to control the DC voltage, active and reactive power smoothly, and with insensitivity to system parameter variations and external perturbations. The global asymptotic stability of the developed control laws is guaranteed employing Lyapunov stability theory. So as to verify the effectiveness of the proposed controller, a simulation based comparative study has been carried out under a change in solar irradiance and ambient temperature. The results demonstrate that the SMC provides better dynamic performance, great decoupling of active and reactive power, good reference tracking and improved power quality even when facing operating point variation comparing with the conventional linear controller.

Khalid Chigane, Mohammed Ouassaid

Automatic Extraction of Photovoltaic Panels from UAV Imagery with Object-Based Image Analysis and Machine Learning

We develop an automatic pipeline for photovoltaic panels extraction based on Object-Based Image Analysis (OBIA) and machine learning (ML). Automatic optimization of segmentation parameters, statistical and morphological feature engineering, and ML segment-based classification are the main building blocks of the proposed pipeline. Experimentation was conducted on a dataset comprising RGB and thermal orthomosaics generated from UAV data. An F-factor of 98.7% was scored with a recall rate above 98%. Obtained results were also compared to a developed solution under the software eCognition.

Imane Souffer, Mohamed Sghiouar, Imane Sebari, Yahya Zefri, Hicham Hajji, Ghassane Aniba

Development of a Management Algorithm for a Compact Photovoltaic—Wind Turbine System

The exploitation of renewable energies despite their availability in domestic scale remains very insufficient. This is mainly due to three reasons. The first one is about the problem of intermittent power generation by solar and wind energy. The second problem is the high cost of traditional solar and wind installations. In addition, the third problem is the large space occupied by these systems. To help solve these problems and optimize the production of electricity and the space occupied, we proposed a compact system with flexible photovoltaic (PV) solar panels that automatically takes the shape of wind turbine blades named Savonius. This coupling ensures two modes of ecological production of green energy namely PV mode and wind turbine mode where these two technologies compensate to partially overcome the problem of intermittent and also save space. This is a new idea in the literature, unlike conventional hybrid systems which include wind turbines and PV solar panels separately. The proposed system is based on a management algorithm developed in MATLAB/Simulink and modeled with “Stateflow” which allows changing between the solar wind modes according of meteorological values.

Yahya Lahlou, Abdelghani Hajji, Mohammed Aggour

Impact of Solar Gain on Energy Consumption and Thermal Comfort

Energy consumption in the residential and tertiary sectors is particularly high in developed countries. There is great potential for energy savings in these sectors. Among the techniques that reduce the energy consumption of a building in the winter, we have the orientation. Optimal exposure ensures thermal and visual comfort with a minimum of energy consumption. We will begin this work with a simulation that shows the effect of solar gain in winter on a room with four supposed orientations: South, East, West and North. Equations which show solar contribution as well as factors influencing it have been developed. Then, we will conduct an experimental comparison between two equivalent rooms with different orientations. For each room, we will measure the temperature, relative humidity and energy consumption of its heat pump. The obtained results show that a south orientation saves up to 8.5% of heating energy, ensures good natural lighting and reduces considerably the relative humidity in the room. Concerning the room facing north, we propose a practical solution which will contribute to the heat pump consumption reduction by exploiting the solar gain and improve natural lighting.

Abdelghani Hajji, Yahya Lahlou, Ahmed Abbou

A Model-Based Predictive Control Approach for Home Energy Management Systems. First Results

The use of renewable energies in buildings are a needed solution to decrease the overall energy consumption. In countries such as Portugal and Morocco, this is translated in the use of Photovoltaic systems, and, hopefully, Energy storage systems. This paper presents a simplified Model-Based Predictive Control (MBPC) approach for a Home Energy Management System of a residence in the region of Algarve, Portugal. Simulation results show that MBPC achieves considerable savings in the use of electricity obtained from the grid, as well as economic savings.

Antonio Ruano, Hamid Qassemi, Inoussa Habou Laouali, Manal Marzouq, Hakim El Fadili, Saad Bennani Dosse

Comparative Study of Electricity Production by Photovoltaic Panels with Mirrors for Different Inclinations

This paper presents an analysis of a system consisting of a PV collector augmented with two reflectors to obtain more electrical energy. The model provides the possibility to predict the solar energy received by the panel in the day. To enhance the amount of solar radiation on the fixed photovoltaic Panel we daily changed the tilt angle of the two reflectors to get them in an optimal position which depends on the angle of elevation of the sun, the azimuth angle for typical days throughout the year. In this work, we calculated the electrical efficiency of the photovoltaic panel in two cases, without and with reflectors in the optimal position. The study is made on the site of Tetouan (longitude =  − 5°, latitude = 35.25°) for a daily variation (from sunset to sunrise) relatively for the days 21 March 21 Mai 21 June 21 September and 21 December, (the spring and autumn equinox and summer and winter solstices), days considered clear sky type.

Assia Benkaddour, Hanan Boulaich, Elhassan Aroudam

A Non Linear Autoregressive Neural Network Model for Forecasting Appliance Power Consumption

Laouali, Inoussa Habou Qassemi, Hamid Marzouq, Manal Ruano, Antonio Bennani, Saad Dosse El Fadili, HakimForecasting the electrical appliance power consumption is a necessary and important part of the management of electrical power system, in order to assess people’s penchant for using electricity. Even though several studies are focused on forecasting building consumption, less attention is given to forecasting the use of single appliances. Indeed, some of the energy needs of consumers may be relatively delayed or anticipated to obtain a better consumption profile while maintaining consumer comfort. This paper focuses on forecasting appliance power consumption using a non-linear autoregressive (NAR) neural network model. The results obtained on the UK-DALE public dataset demonstrate that NAR models are suitable for forecasting of energy consumption with a good accuracy. The proposed model obtained the best Mean Absolute Errors, compared with the LSTM, Autoencoder, Combinatory optimization, FHMM, and Seq2point techniques.

Inoussa Habou Laouali, Hamid Qassemi, Manal Marzouq, Antonio Ruano, Saad Bennani Dosse, Hakim El Fadili

Numerical Analysis of Bi-fluid PV/T Hybrid Collector Using the Finite Difference Method

The PV/T is called as a hybrid bi-fluid type solar collector when both fluids (water and air) are used as working fluids. This type of collector provide a wide range of thermal applications and several modes can be executed depending on energy needs and applications: air mode, water mode and simultaneous mode (air and water). This paper presents a numerical analysis of PV/T bi-fluid collector using the finite difference method. Energy balance equations were established for each layer and solved using the Gauss–Seidel iteration method. The performance of the PV/T hybrid is evaluated according to the operating mode of the fluid (independently and simultaneously). The simulation results are compared with literature (Modeling and performances assessments of PV/T bifluid hybrid collector: Three cooling modes operation case. https://doi.org/10.1109/iceit48248.2020.9113233 , [10]), and indicate that when both fluids are used simultaneously, the overall electrical and thermal performance of the hybrid collector is considered satisfactory compared to the situation where two fluids operate independently.

Oussama El Manssouri, Bekkay Hajji, Antonio Gagliano, Giuseppe Marco Tina

A Novel Cryptosystem for Color Images Based on Chaotic Maps Using a Random Controller

Hraoui, Said Gouiouez, Mounir Gmira, Faiq Berrada, Mohammed Jarjar, Abdellatif Jarrar, A. OulidiA new encryption technique for multicolor images is outlined in this study. After vectorization of the clear image, an initialization value will be calculated. This value, allows changing only the first pixel. This pixel will start the encryption process. In parallel, in a first step, we will make a confusion by chaotic vectors entirely controlled by another chaotic vector. In a second step, three chaotic substitution matrices of size (256, 256) will be generated. The passage of each pixel through these matrices will also be controlled by another chaotic vector. Finally, a strong link will be set up between the new pixel state and the previous ones to set up better diffusion/confusion. This step increases the avalanche effect. Testing performed through our algorithm on standard images shows the durability of our system.

Said Hraoui, Mounir Gouiouez, Faiq Gmira, Mohammed Berrada, Abdellatif Jarjar, A. Oulidi Jarrar

New Image Encryption Scheme Based on Dynamic Substitution and Hill Cipher

In this work, we propose a new color image encryption technique. After transiting of the original image into a vector and decomposing it into blocks of three pixels, along with modifying of a seed block by an initialization vector computed from the plain image, a preliminary confusion will be handled by a substitution matrix developed under the control of the two chaotic maps used in the system. The achieved block will be injected in affine transformation provided by an invertible matrix accompanied by a dynamic translation vector to surmount the problem of null or uniform blocks. The encrypted block will be linked to the original block to set up diffusion and avalanche effect to protect the system from differential attacks. Simulations applied to on a large number of color images prove the robustness of the proposed approach against known attacks.

Younes Qobbi, Abdeltif Jarjar, Mohamed Essaid, Abdelhamid Benazzi

Touchless Palmprint Identification Based on Patch Cross Pattern Representation

Over the last decade, palmprint recognition has been studied for many problems and applications. It has become one of the most well-known biometric recognition system. Its success is due to the rich features that can be extracted and exploited from the palmprint images captured by contact or contactless device. This paper presents a new representation based on textural structure of human palms for touchless palmprint identification. This representation method is called Patch Cross Pattern (PCP), which relies mainly on cross pattern encoder and the non-overlapping decomposition method. The feature vector is built using Cross Pattern (CP) encoder to capture the textural structure of palmprint image. Then, the non-overlapping decomposition on both directions is applied. Next, the feature vector representation of each palmprint image is constructed by concatenating all normalized histograms calculated at each patch. In addition, the reduced version of the PCP called R-PCP is obtained using whitened linear discriminant analysis. Finally, a K-nearest neighbor classifier is used for palmprint identification. The proposed system is successfully applied to IIT Delhi and CASIA touchless databases. Results show that, the proposed representation provides a significant performance improvement compared to the recent state-of-the-art in terms of accuracy.

Hakim Doghmane, Kamel Messaoudi, Mohamed Cherif Amara Korba, Zoheir Mentouri, Hocine Bourouba

Image Segmentation Approach Based on Hybridization Between K-Means and Mask R-CNN

In this article, we will introduce a hybrid method based on the combination of two image segmentation techniques. The first method adopted is the k-means algorithm which is an unsupervised machine learning technique used to group data points, the second is the Mask R-CNN which is a neural network architecture which combines two sub-problems: object detection and semantic segmentation. The main objective of this study is to approve the segmentation of the image using k-means. The first step is to apply Mask R-CNN on our original image to detect the objects that are present in the image, then, we will apply k-means clustering to have better segmentation. For our approach we used a set of metrics to evaluate our proposed approach such as the mean square error (MSE); the peak signal to noise ratio (PSNR) and many other measures such as the difference between two images. These measurements have shown satisfactory results and a performance of the proposed method.

Hanae Moussaoui, Mohamed Benslimane, Nabil El Akkad

Partial 3D Image Reconstruction by Cuboids Using Stable Computation of Hahn Polynomials

In this paper, we will present a new method of partial reconstruction of the 3D image. The latter is based on the following two concepts: one is the stable discrete orthogonal Hahn polynomials (DOHPs), which significantly reduces numerical defects. The other is the 3D image cuboid representation (ICR) to accelerate the computation time of discrete orthogonal Hahn moments (DOHMs) and improve the quality of 3D image reconstruction. The results of the simulation confirm the ability of the 3D image reconstruction using the proposed process.

Mohamed Amine Tahiri, Hicham Karmouni, Ahmed Tahiri, Mhamed Sayyouri, Hassan Qjidaa

Analysis of Online Spiral for the Early Detection of Parkinson Diseases

Parkinson’s disease (PD) is a neurodegenerative disorder that affects a person’s movement. As the early diagnosis of the disease is crucial, the main aim of this work is to implement an online analysis system of patients’ handwriting, through computer vision and signal processing techniques, using the database collected in the neurology department of the University Hospital Center Hassan II in Fez. For this, we studied the handwriting tests on a WACOM graphic tablet to retrieve the spatiotemporal data (position, pressure and angles of inclination), for each point (P(n)) of the trajectory. The features vector was obtained basing on five types of features: (a) Kinematic features related to the dynamics of spiral design, (b) Mechanical based on the pressure exerted on the writing surface, (c) Inclination angles, (d) Spatial interrelation feature and (e) Pen-Up. The used classification and clustering algorithms are respectively the Hoeffding tree and the FarthestFirst clusters. We observed coherence between the classification results and the clustering ones, thus the results being encouraging and promising with a recognition rate of 98.36%

Yassir Elghzizal, Ghizlane Khaissidi, Mostafa Mrabti, Aouraghe Ibtissame, Ammour Alae

Learning Hand-Crafted Palm-Features for a High-Performance Biometric Systems

Bouchemha, Amel Meraoumia, Abdallah Laimeche, Lakhdar Houam, LotfiThe extraction of distinctive image features is the most important step in pattern recognition systems due to their direct impact on learning the machine commonly used in these types of systems. In this paper, we propose a handcraft feature learning, which based on local distinctive image descriptors, for multispectral palmprint representation and recognition. In the training phase, a projection matrix (hash functions) and a codebook are obtained using the Pixel Difference Vectors (PDVs) of non-overlapping sub-blocks, in order to use it as prior knowledge in the feature extraction step. For the test phase, the extracted PDVs are encoded into binary codes using the projection matrix, then pooled as a histogram feature using the codebook. The experimental results carried out on the CASIA database show that the proposed framework achieves better performances compared to the state-of-the-art methods, in particular the handcrafted ones.

Amel Bouchemha, Abdallah Meraoumia, Lakhdar Laimeche, Lotfi Houam

CNN-Based Obstacle Avoidance Using RGB-Depth Image Fusion

In the last few years, deep learning has attracted wide interest and achieved great success in many computer vision related applications, such as image classification, object detection, object tracking, pose estimation and action recognition. One specific application that can greatly benefit from the recent advance of deep learning is robot vision-based obstacle avoidance. Vision-based obstacle avoidance systems are mostly based on classification algorithms. Most of these algorithms use either color images or depth images as the main source of information. In this paper, the aim is to investigate whether using information extracted from both types of images simultaneously would give better performance than using each one separately. To do this, we chose the convolutional neural network (CNN) as the classifier and HSV-based method to achieve the fusion. We tested this approach using two widely used pre-trained CNN architectures, namely Resnet-50 and GoogLeNet using a dataset locally collected. The results indicate that the image fusion-based classification algorithm achieve a higher accuracy (91.3%) than the one based on depth images (80.4%) but lower than the one based on color images (93.7%). These results can be partly explained by the fact that the used classifiers were pre-trained using color image datasets.

Chaymae El Mechal, Najiba El Amrani El Idrissi, Mostefa Mesbah

Arabic Handwriting Word Recognition Based on Convolutional Recurrent Neural Network

The success of any words-characters recognition system depends on board parameters such as the language (Arabic, Latin, Indi …), the document type (writing or typing), based or free-segmentation, pretreatment, features extraction and classification approaches. Within these fields, Building a robust and viable recognition system for Arabic handwritten has always been a challenging task since a long time. In this study, we propose an end-to-end system based on deep Convolutional Recurrent Neural Network CNN/RNN; we trained our system on IFN/ENIT extended database in order to improve our results.

Manal Boualam, Youssef Elfakir, Ghizlane Khaissidi, Mostafa Mrabti

Tuning Image Descriptors and Classifiers: The Case of Emotion Recognition

Greche, Latifa Taamouch, Abdelhak Akil, Mohamed Es-Sbai, NajiaIn general, the recognition involves several steps as follows: data acquisition, pre-processing, segmentation, feature extraction and classification. Automatic facial expression recognition has become a crucial technology in the computer vision field and its applications including identification and security, Medicine and Monitoring. The facial expression recognition system requires an algorithmic pipeline that involves two main blocks: feature extraction and classification. A large experimental session must lead to the adequate algorithmic pipeline, notably for identifying the best methods for feature extraction and classification to achieve robust facial expression recognition with high accuracy. Thus, it is essential to analyse data using multiple methods of feature extraction and classification. In this paper, we propose an approach to automate the analysis of data by repeating tests made to tune and compare feature extraction and classification methods. We evaluate our proposed data analysis approach using video sequences with fundamental emotion states: neutral expression, disgust, fear, happiness, sadness, anger and surprise. To transform the face images into vectors of features, we use shape, texture, and contour descriptors. This enables storing images in a table of vectors. Each table related to every descriptor is analysed with Five classifiers have been used, which are support vector machine, linear discriminant analysis, k-nearest neighbors, naïve Bayes, and binary tree classifiers. The techniques 10-fold and Leave-One-Out Cross-validations and the grid search have been used to tune the hyperparameters methods and compare them, computing the average recognition rate F-measure as evaluation metric. Experimentation on ChonKanade Image database (CK+) show that the proposed data analysis approach can find out the optimal combination to separate the data classes and identify the expression with a an F-score of 96.44%.

Latifa Greche, Abdelhak Taamouch, Mohamed Akil, Najia Es-Sbai

Prediction Potential Analysis of Arabic Diacritics and Punctuation Marks in Online Handwriting: A New Marker for Parkinson’s Disease

Parkinson’s disease (PD) is a progressive movement disorder characterized by tremors at rest, bradykinesia, and stifness. The alteration of handwriting (HW) faculties is one of the earliest motor symptoms in PD patients. This characteristic can be exploited to develop an automatic aid system for early detection of this pathologie. This article aims to assess the importance of diacritics and punctuation marks (DPM) in the PD patients and healthy controls (HCs) discrimination problem, by comparing the classification results obtained from three components: text carrying DPM, text without DPM, as well as only DPM. This work includes the Arabic manuscripts of 31 PD patients and 31 HCs. Furthermore, kinematic, mechanic, and inclination features were calculated for each component. Then, Adaboost models have been constructed on different feature sets, as well as on reduced sets formed in incremental manner using mRMR ranked-feature selection method. From the obtained results, it was concluded that the separating power of HW features in the classification problem of PD patients and HCs is present in all components of the Arabic text, but in varying degrees of importance. Despite the simple graphical nature of DPM, they are carrying of relevant diagnostic information, and effectively contributing to the improvement of PD detection performance. The highest accuracy of 93.54% was achieved for this component.

Alae Ammour, Ibtissame Aouraghe, Ghizlane Khaissidi, Mostafa Mrabti, Ghita Aboulem, Faouzi Belahsen

Development of an Ultra Wide Band Hybrid Coupler with Adjustable Phase Shifter for 5G Applications

The goal of this paper is to design and develop of an ultra-wideband (UWB) hybrid coupler with a variable phase shift (90°–180° and vice versa) which operates around the resonant frequency of 3.8 GHz required for 5G applications. The design approach is based on the microstrip waveguide technique with localized lumped elements to control the phase shift between outputs ports. The proposed coupler is planar etched onto a Rogers RT substrate with an overall size of 50 × 42 × 1.6 mm3 and dielectric constant εr = 2.2. The performance of the coupler can be seen from its important characteristics: Adaptation, Coupling, Isolation and Phase shifter.

Abdellatif Slimani, Saad Bennani Dosse, Ali El Alami, Mohammed Jorio, Abdelhafid Belmajdoub, Mohamed Amzi, Sudipta Das, Sghir Elmahjouby

WiMAX Throughput Maximization for MIMO-OFDM Systems via Cross-Layer Design

WiMAX system is based on the IEEE 802.16-2005 specification and provides wireless broadband to fixed and mobile users. It adopts MIMO technology to produce a high system capacity and spectral efficiency. By coupling a robust and efficient OFDM technique with MIMO antenna diversity systems, we offer a very compelling high-speed data downlink solution for future wireless communications systems. IEEE802.16e WiMAX system proposes an adaptive digital modulation and coding schemes according to the wireless link quality. In this work, we present a Cross-Layer approach based on the adaptive Modulation and Coding Scheme (MCS) mechanism which incorporates Network layer adaptive code with the PHY layer feedback information. Results show that this design achieves better performance of the WiMAX achievable throughput and transmission efficiency than adaptive modulation and coding used only with ARQ retransmissions through STBC and SM schemes.

Equivalent Circuit Modelling of a Cantor Multifractal Slots Antenna

The paper aims to study the effect of Cantor multifractal introducing in the radiating element of the reference antenna, then to propose an equivalent circuit model of the investigated antenna by means of two approaches. The First approach is based on the Vector fitting approximation of simulated data. In the second approach, the antenna input impedance is described by means of the first Foster canonical form. To ensure the validity of the suggested equivalent circuit model the results of the used methods are compared.

Fatima Ez-Zaki, Hassan Belahrach, Abdelilah Ghammaz

A Novel Two-Branch Dual-Band Rectifier for 2.45 GHz 5.8 GHz RFID Systems

In this paper, we present a novel design of high-efficiency, dual-band and two-branch rectifier for wireless energy harvesting in the 2.45 and 5.8 GHz frequency bands. The rectifier is based on a two conversion circuit; a dual-band voltage doubler for the first branch and a single-diode rectifier for the higher frequency band in order to further enhance in 5.8 GHz the efficiency of the resulting circuit. The rectifier achieves a maximum conversion efficiency of 59.683% for the primary band at 1dBm of incident power, while it reaches in the second band (5.8 GHz) at 3 dBm an efficiency of 51.521%. These results allowed the resulting rectifier to be the most efficient at these frequencies and in particular for the low incident powers compared to other reported designs with the same radio frequency conditions.

Sara El Mattar, Abdennaceur Baghdad, Abdelhakim Ballouk

A Survey of NOMA for 5G: Implementation Schemes and Energy Efficiency

The 5th generation (5G) Network technology must meet various stringent necessities such as high spectral efficiency, low energy consumption and immense connectivity. Current wireless communication system uses radio resources to transmit data with Orthogonal Multiple Access (OMA). But, as the number of user’s growths, OMA-based technique might not meet user requirements. That why, NOMA has been introduced to fulfil the challenges of 5G network. This article provides a comparison between NOMA and OMA systems. The main NOMA strategies are debated by dividing them into two classes, namely, the power domain and the NOMA code domain. In addition, benefits and limitations are discussed.

Jamal Mestoui, Mohammed El Ghzaoui

A Modified E-Shaped Compact Printed Antenna for 28 GHz 5G Applications

A single band printed antenna with a modified E-shaped structure is proposed for 5G applications at 28 GHz band. The suggested antenna has been constructed on a 5.5 $$\times$$ × 4.35 mm2 FR-4 substrate ( $${\varepsilon }_{r}=4.4$$ ε r = 4.4 and $$h=1.6 \mathrm{mm}$$ h = 1.6 mm ) through slot loading technique. The incorporation of rectangular and tiny square shaped slots has improved the resonance and radiation characteristics of the designed antenna. The wide operating bandwidth of the suggested antenna covers 25.99–29.88 GHz frequency spectrum to support 28 GHz 5G applications. The presented antenna depicts low mismatch loss and thus better impedance matching offering a reflection coefficient of about −40.88 dB at 28 GHz. The proposed antenna offers desired radiation patterns (E and H planes) with a peak gain of about 5.64 dBi at 28 GHz. Furthermore, an extensive analysis of VSWR, input impedance, and surface current distribution are also presented to explain the working functionality of the suggested antenna. The proposed structure is a preferable choice for 28 GHz 5G applications due to its compact size and high performance parameters and it covers the bandwidth requirements of 5G applications at 28 GHz.

Yousra Ghazaoui, Ali El Alami, Sudipta Das, Mohammed El Ghzaoui

Design of Microstrip Sierpinski Carpet Antenna Using a Circular Pattern with Improved Performance

In this work, we present the two first iterations design of the Sierpinski carpet fractal antenna by using a circular pattern. The proposed antenna is printed on FR4 substrate with a dielectric constant of 4.4. At the second iteration, the studied antenna has a multiband behavior with four resonant frequencies: 3.92, 4.89, 6.61 and 7.22 GHz with a good impedance matching. The simulated results performed by CADFEKO a Method of Moments (MoM) based Solver and measurement using Vector Network Analyzer (VNA) Anritsu MS2026C are in good agreement.

Abdelhakim Moutaouakil, Younes Jabrane, Abdelati Reha, Abdelaziz Koumina

Load Condition for Minimum Backscattering Antennas

The scattered fields of antennas can be decomposed into two terms; one depends only on the structure of the antenna while the other depends on the antenna radiation parameters and the loading conditions. This decomposition is used to investigate the antenna radar cross section (RCS) behavior with variable loadings. Then, a general analytical load condition for zero or minimum bistastic radar cross section is derived. The derived formula is used to obtain the regions of feasible passive loads for zero backscattering for three different antenna structures. The backscattering levels of these antennas are examined to validate this formula. The results demonstrate the validity of the procedure with almost zero backscattering for all structures with evident RCS reduction factors.

Zaed S. A. Abdulwali, Majeed A. S. Alkanhal

LTE-M Evolution Towards Massive MTC: Performance Evaluation for 5G mMTC

Abou El Hassan, Adil El Mehdi, Abdelmalek Saber, MohammedSince emerging fifth generation (5G) wireless network is expected to significantly revolutionize the field of communication, its standardization and design should regard the Internet of Things (IoT) among the main orientations. Long Term Evolution-Machine Type Communication (LTE-MTC/LTE-M) is among the emerging technologies that are promising candidates for 5G wireless network, enabling massive deployment of devices for massive Machine-Type Communication (mMTC). This paper describes LTE-M evolution in recent 3rd Generation Partnership Project (3GPP) releases. A complete evaluation of LTE-M performance against 5G massive MTC requirements is presented. The results analysis of this evaluation show that these requirements can be met but under certain conditions regarding the configuration and deployment of the system. The enhancements provided by the recent 3GPP releases are also discussed.

Adil Abou El Hassan, Abdelmalek El Mehdi, Mohammed Saber

Communication Optimization Approach for S-Band LEO CubeSat Link Budget

El Moukalafe, Mohammed Amine Minaoui, KhalidCubeSats has evolved a lot since their first appearance in 1999. Indeed, we see that small satellites have emerged in several areas that were exclusive to large satellites. However, this miniaturization is costly in term of size and energy which are crucial for high data rate communication systems. Hence, the need to optimize communication link parameters for energy efficiency. Previous works on link budget analysis adjust modulation and coding schemes on worst cases in order to guarantee a robust communication link between CubeSats and Grounds Stations. Thus, penalizing the data rate. In our work we start by establishing a classic link budget on S-band transceivers, then we study the impact of communication conditions improvement on data rate using Variable Coding and Modulation (VCM) approach. Our results have demonstrated a notable performance gain from applying VCM technique to a Low Earth Orbit (LEO) CubeSat link budget.

Mohammed Amine El Moukalafe, Khalid Minaoui

Ground Penetrating Radar Data Acquisition to Detect Imbalances and Underground Pipes

Past research shows that the Ground Penetrating Radar (GPR) can be an effective and efficient way to map buried pipeline systems. This paper presents a GPR data analysis technique, which is first generated by applying the Finite Differences Time Domain (FDTD) method to estimate the thickness of the subsurface layers and characterize the piping systems buried in the underground. In practical investigations, the GPR unit with a 400 MHz antenna was used to detect imbalances and underground pipes. The GPR profiles provided details on the shapes and nature of the target in the underground. These profiles can, therefore, detect water pipes, utility systems up to a depth than 2 m.

Tahar Bachiri, Gamil Alsharahi, Abdellatif Khamlichi, Mohammed Bezzazi, Ahmed Faize

Nash Equilibrium Based Pilot Decontamination for Multi-cell Massive MIMO Systems

Belhabib, Abdelfettah Boulouird, Mohamed Hassani, Moha MRabetDue to the limitation imposed by the scarcity of pilot resources, Massive multi-input multi-output ( $$M^{MIMO}$$ M MIMO ) technology suffers from the problem of pilot contamination (PC), which is due to the reuse of the same pilot sequences for the users of the adjacent cells. To deal with this constraint, this paper proposes a new decontaminating strategy, which is based on the Nash Equilibrium theory. Specifically, by exploiting the large-scale fading coefficients, the users are assigned with the available pilot sequences under the constraint to fulfill a good/fairness signal-to-interference-plus-noise ratio (SINR) to the users of the overall cells. Because, even though the users of the same cell are optimally assigned with the pilot sequences, the achieved SINR in some adjacent cells can be degraded, which is unfair; consequently, the proposed strategy assigns the available pilots to the users to reach an acceptable SINR for the users of the overall cells; in other words, to guarantee a high and similar quality of service for the users of all cells. Simulation results prove the effectiveness of our proposal on boosting the per-cell achievable rate.

Abdelfettah Belhabib, Mohamed Boulouird, Moha M’Rabet Hassani

Channel Estimation for Massive MIMO TDD Systems and Pilot Contamination with Uniformly Distributed Users

Amadid, Jamal Boulouird, Mohamed Hassani, Moha MRabetThis work introduces a straightforward and applicable channel estimator for Rayleigh Fading Channels (RFC) in Multi-Cell (M-C) Multi-User (M-U) Massive MIMO (m-MIMO), Time Division Duplex (TDD) systems with Pilot Contamination (P-C). The traditional Least Square (LS) channel estimator undergoes considerable performance degradation whereas the Ideal Minimum Mean Square Error (I-MMSE) presents a much better performance in comparison with the LS estimator. Nonetheless, in many studies, the I-MMSE estimator depends on a supposition unjustified, which is prior knowledge of channel statistics. The Maximum Likelihood (ML)-based estimator solves those problems of both current estimators and furnishes Channel Estimation (C-E) performance approaching that of the I-MMSE estimator.

Jamal Amadid, Mohamed Boulouird, Moha M’Rabet Hassani

Mapping the Geothermal Potential of the Jbel Saghro Massif by Airborne Magnetism (Anti-Atlas, Morocco)

The main objective of this research is to use the aeromagnetic data covering the Jbel Saghro Massif (Southern Morocco) to study the transition zone between magnetized and non-magnetized rocks by estimating the curie isotherm depth, to calculate the thermal flux in mW m−2. The processing and filtering of magnetic data has given the possibility to trace the lateral limits of hidden and magnetic geological sources, the latter have directions ENE–WSW, NNW–SSE, W–E and NW–SE inherited by the geodynamic evolution of the study area. Using the local wave number method, a curie depth isobaths map was generated, showing varying values from 6.9 to 25.9 km. The regional geothermal gradient map shows values between 22 and 85 °C km−1. The calculation of the thermal flux reveals several geothermal provinces (>190 mW m−2) with a total area of about 1500 km2 associated with inliers of Precambrian age.

Miftah Abdelhalim, El Azzab Driss

A Compact Flexible UWB Antenna for Biomedical Applications: Especially for Breast Cancer Detection

Worldwide, cancer remains a scary topic that cause many deaths in recent years such as breast cancer, which affects many women in many ways. UWB antennas play an important role in early detection. The design of UWB antennas for biomedical applications presents a promising challenge and requires more research to improve health care system. This paper proposes an UWB antenna printed on a flexible substrate with thickness of 0.254 mm and relative permittivity εr = 2.2, operating in the range of 3.64–12.11 GHz. For its realization, we applied some techniques of miniaturization to improve some antenna characteristics. In addition, the effects of the curvature and the biological tissues on the return loss and radiation pattern were investigated.

Nirmine Hammouch, Hassan Ammor, Mohamed Himdi

A Low Profile Frequency Reconfigurable Antenna for mmWave Applications

A compact frequency reconfigurable printed antenna for millimeter-wave applications is presented in this paper. This design is obtained by the merging of a half-arc and a right-angled triangle patch extracted from a rectangular radiator. The design approach is based on the use of two S-PIN diodes to obtain frequency reconfigurability. The antenna exhibits seven reconfigurable bands while showing good performances in terms of return loss, bandwidth and gain. The proposed antenna is well suited for future fifth-generation (5G) networks because of its notable features of small overall size (7.5 $$\times$$ × 5 $$\times$$ × 0.762 mm3), wide bandwidth, and frequency reconfigurability.

Wahaj Abbas Awan, Niamat Hussain, Adnan Ghaffar, SyedaIffat Naqvi, Abir Zaidi, Musa Hussain, Xue Jun Li

On-Demand Frequency Reconfigurable Flexible Antenna for 5Gsub-6-GHz and ISM Band Applications

This paper presents the design and characterization of on-demand frequency reconfigurable antenna for 5Gsub-6-GHz and ISM-band applications. The antenna comprises of compact overall size having dimensions of 0.23λ0 × 0.2λ0 × 0.002λ0. The octagonal patch was loaded with stub to enhance bandwidth along with improvement in return loss, afterword a slot and diode was utilized to achieve frequency reconfigurability. The numerical analysis of the proposed antenna was done using Higher Frequency Structural Simulator (HFSS). Furthermore, to demonstrate the potential of the proposed work it is compared with state of the art works for similar applications.

Musa Hussain, Syed Naheel Raza Rizvi, Wahaj Abbas Awan, Niamat Husain, Halima, Ahsan Hameed

Dual-Band BPF Based on a Single Dual-Mode Stepped-Impedance Resonator for 4G Systems

A dual-band bandpass filter (BPF) based on a dual-mode stepped-impedance resonator (SIR) operating at 1.8/2.6 GHz is proposed. It is shown that the dual-band BPF has a smaller area (12.6 × 15 mm2), lower Insertion loss (<0.5 dB) and good return loss (>15 dB) at both passbands. In addition, two transmission zeros are created by the input and output tapping to enhance the rejection band (>30 dB). The simulations are carried out using Ansoft’s HFSS and CST-MS software.

Mohamed Amzi, Jamal Zbitou, Saad Bennani Dosse

High Gain Cascaded GaAs-pHEMT Broadband Planar Low Noise Amplifier for WiMAX-802.16b Applications

Low noise amplifiers are essential structures in telecommunications systems. The problem of LNAs in planar technology is the difficulty of having a flat gain with a low noise figure in a wide bandwidth. In this work, we focus on the design of a broadband low noise amplifier for WiMAX-802.16d applications using GaAs-pHEMT ATF-35176 of Avago technologies. The three-stage cascade configuration mounted on an FR-4 substrate is used and provides a high gain, with an average value of 34.3 dB and a low noise figure of 0.95 ± 0.1 dB with a power consumption of 417 mW. The total die size is 9.5 × 2 cm2.

Moustapha El Bakkali, Naima Amar Touhami, Taj-Eddin Elhamadi

Application of Electrical Resistivity Soundings to Identify Unstable Areas, “Tghat-Oued Fez” District as a Case Study (Fez—Morocco)

Measurements of 72 vertical electrical soundings (VES) using a Wenner-Schlumberger electrode array configuration were conducted on “Tghat-Oued Fez” district of Fez city. The purpose of this study is to identify voids inside basement, to prevent collapse and landslides in urban areas. The natural parameters acting to accelerate the phenomenon are mainly the tectonic activity, topography, and lithology. The studied area is close to the major active south rifain fault, marked by a steep slope and heterogeneous lithology, which is composed of a glacis surface, marls, and conglomerate deposits. We suspect the presence of week zones inside the marls layer because of cracks and fissures which sometimes appear even on the surface and the surrounding building's wall. This is due to the presence of gypsum which left many voids and cavities due to dissolution. Thirteen iso-resistivity maps resulting from VES measurements were created. The electrical survey has accurately detected and located high resistivity anomalies inside the conductive marly formation between 7 and 15 m depth. The survey results suggest that VES is a viable geophysical tool to explore electrical anomalous linked to medium vulnerable to lands movement.

Jabrane Oussama, El Azzab Driss, El Mansouri Bouabid, Charroud Mohammed

A New Compact 1.0 GHz LPF Using Double Hi-Lo-Resonators and Cross Defected Ground Structure for Radar Applications

In this work, a novel miniaturized microstrip low pass filter using a double Hi-Lo and cross defected ground structure resonators is presented. The Hi-Lo resonator is placed on the top layer of the structure while the two identical cross DGS resonators are etched in the ground plane. Each DGS shape consists of two cross heads, which are connected with a channel slot. The both DGS resonators are electromagnetically coupled. The proposed filter has been designed simulated, optimized and manufactured. The filter topology is simulated using HFSS simulator and measured using Agilent Field Fox NA, N9918A VNA. The both results of the proposed LPF shown a sharp roll-off (ξ) of 84 dB/GHz and exhibit a very low insertion loss in the pass band of 0.4 dB from DC to 0.9 GHz and it achieves a wide rejection bandwidth with overall 20 dB attenuation from 1.2 GHz up to 3.2 GHz. The compact low pass structure occupies an area of (0.37λg × 0.37λg) where λg = 94 mm is the waveguide length at the cut-off frequency 1 GHz. The carried-out results confirm the effectiveness of the proposed method.

A. Boutejdar, H. Bishoy, Saad Bennani Dosse

Backmatter

Weitere Informationen