Adnan Shaout, Dhanush Mysuru, and Karthik Raghupathy

The Electrical and Computer Engineering Department, the University of Michigan, USA

Abstract: modern vehicles nowadays have many Electronic Control Units (ECU) and sensors. Information (data) within an automobile is transferred via the Control Area Network (CAN) of a vehicle. This data may not be important to the driver, but if the CAN messages are analyzed, then driver behavior and vehicle conditions can be determined. This data can determine ahead of time any possible future failure and the driver can be alerted about it. This paper presents standalone real time embedded system that could analyze the CAN data were valuable lives may be saved in real time. The proposed hardware system acquires CAN data from a vehicle (CAN sniffing). The data is then processed and an appropriate message is then displayed for the driver. The proposed system also stores the CAN messages on to an industrial grade SD card (CAN logging) for future analysis. The proposed system can also perform driver behavior and driving terrain analysis.

Keywords: vehicle condition analysis, CAN sniffing, CAN logging, CANoe 9.0, driver behavior.

Abstract: Fighting poverty is one of the main objectives of sustainable development program. In a country like Lebanon, where poverty is a real threat and hidden under a good living looking, the situation should be explored in depth. This paper aims to evaluate the position of Lebanon compared to other Middle East countries in sustainable development. Furthermore, our goal is to reveal the power and weaknesses of resources management, based on income and non-income indicators retrieved from World data bank. For this purpose, we adopted a combination of data mining techniques as tools to study the relationship between these indicators. The K-means clustering technique is used to define the different levels of living. In order to extract the most relevant non-income indicators to our study, information gain as feature selection technique was applied. Finally, k-Nearest Neighbor (KNN) classification technique was used for the predicting model.

The Impact of Natural Language Preprocessing on Big Data Sentiment Analysis

Abstract: The sentiment analysis determines peoples’ opinions, sentiments and emotions by classifying their written text into positive or negative polarity. The sentiment analysis is important for many critical applications such as decision making and products evaluation. Social networks are one of the main sources of sentiment analysis. However, the huge volume of data produced by social networks requires efficient and scalable analysis techniques to be applied. The MapReduce proved its efficiency and scalability in handling big data, thus attracted many researchers to use the MapReduce as a processing framework. In this paper, a sentiment analysis method for big data is studied. The method uses the Naïve Bayes algorithm for classifying texts into positive and negative polarity. Several linguistic and Natural Language Processing (NLP)preprocessing techniques are applied on a Twitter data set, to study their impact on the accuracy of big data classification. The preformed experiments indicates that the accuracy of the sentiment analysis is enhanced by 5%, yielding an accuracy of 73% on the Stanford Sentiment data set.

Temporal Neural System Applied to Arabic Online Characters Recognition

Khadidja Belbachir¹ and Redouane Tlemsani²

¹Department of computer sciences, University of Science and Technologies of Oran Mohamed Boudiaf, Algeria

²LaRATIC Laboratory, National Institute of Telecommunication an ICTs of Oran, Algeria

Abstract: This work presents survey, implementation and test for a neural network: Time Delay Neural Network (TDNN), applied to on-line handwritten recognition characters. In this work, we present a recognizer conception for on-line Arabic handwriting. On-line handwriting recognition of Arabic script is a complex problem, since it is naturally both cursive and unconstrained. This system permits to interpret a script represented by the pen trajectory. This technique is used notably in the electronic tablets. We will construct a data base with several scripters. Afterwards, and before attacking the recognition phase, there is a constructional samples phase of Arabic characters acquired from an electronic tablet to digitize Noun Database. Obtained scores shows an effectiveness of the proposed approach based on convolutional neural networks.

Keywords: Isolated handwritten characters recognition, on-line recognition, convolution neural network, TDNN.

Mesh HDR WPAN Resource Allocation

Optimization Approaches

Samar Sindian¹, Abed Ellatif Samhat³, Matthieu Crussière², Jean-François Hélard², and Ayman Khalil¹

¹CCE Department, Faculty of Engineering, IUL, Lebanon

²INSA Rennes, University Rennes, France

³Faculty of Engineering, Lebanese University, Lebanon

Abstract: In the multihop IEEE 802.15.5 networks, all the devices compete to the resources of a shared superframe. In order to distribute these resources in a fair and satisfactory manner among the competing devices, we propose a distributed optimization framework for resource allocation scheme in an IEEE 802.15.5 hop-1. For this purpose, we introduce in this paper a suite of optimization problems for the hop-1 IEEE 802.15.5 resource allocation to optimize fairness and satisfaction without exceeding the superframe size and respecting the demanded channel time size sent by the requesting devices. Simulation results studied and compared the satisfaction factor and fairness index of the different proposed optimization problems. Consequently, a trade-off between satisfaction and fairness should be conducted for choosing the optimal solution.

Keywords: IEEE 802.15.5; resource allocation; optimization; fairness; NUM.

Design of an Automated Extraoral Photogrammetry 3D Scanner

Musa Alyaman¹, Alaa Abd-Raheem², and Farah AlDeiri³

¹Department of Mechatronics Engineering, the University of Jordan, Jordan

²Nanolab, German Jordanian University, Jordan,

³Levant Engineering Company, Jordan

Abstract: Dental anatomy is a field of anatomy dedicated to the study of tooth structure, it uses 3D physical model for teeth and jaw. Creating a computer aided design model from an existing teeth or jaw is called 3D scanning. This can be accomplished using different techniques like: LASER, camera, and many advanced optical techniques. In this work, Photogrammetry will be used, it is based on camera scanning technique. The input is a set of photographs taken by a camera from a predefined positions and orientations, and the output is a 3D model of a real-world teeth or jaw. Structure from Motion algorithm is used to 3D reconstruction from 2D images, it produces the point cloud and eventually the complete textured mesh. Our system is closed, where it controls scanning conditions in terms of lighting and angles of captured images.

Keywords: 3D scanning, photogrammetry, Structure from Motion, 3D reconstruction, dental anatomy, teeth impression.

Received September 29, 2018; accepted January 23, 2019

Full Text   

Realistic Heterogeneous Genetic-based

RSU Placement Solution for V2I Networks

Mahmoud Al Shareeda, Ayman Khalil, and Walid Fahs

Faculty of Engineering, Islamic University of Lebanon, Lebanon

Abstract: The main elements of a Vehicular Ad Hoc Network (VANET), besides VANET-enabled vehicles, are Roadside Units (RSUs). The effectiveness of a VANET in general depends on the density and location of these RSUs. Throughout the primary tiers of VANET, it will not be possible to install a big number of RSUs either due to the low market penetration of VANET enabled vehicles or due to the deployment fee of RSUs. There is, therefore, a need to optimally select a restricted number of RSUs in a special region in order to accomplish maximum performance. In this article, we use the well known genetic algorithm primarily based on RSU region to locate the most appropriate or near optimal solution. We supply the fundamental simulation environment of this work by OpenStreetMap (OSM) to download actual map data, Grupo de Arquitectura y Tecnología de COMputadores (Gatcom) to generate car mobility, Software Update Monitor (SUMO) to simulate street traffic, Veins model framework for walking vehicular network simulation, OMNET++ to simulate practical network and Matlab to build the algorithm in order to analyze the results. The simulation scenario is primarily based on Hamra district of Beirut, Lebanon. Based on the genetic algorithm, our proposed RSU placement model demonstrates that a most appropriate RSU position that can enhance the reception of Basic Safety Message (BSM) delivered from the vehicles, can be performed in a exact roadmap layout.

Keywords: Vehicular model, RSU, genetic algorithm, optimization.

Enhanced Performance and Faster Response

using New IoT LiteTechnique

Kassem Ahmad^1,2, Omar Mohammad¹, Mirna Atieh³, and Hussein Ramadan⁴

¹Department of Computer Science and IT, Lebanese International University, Lebanon

²Department of Computer Engineering, Islamic University of Lebanon, Lebanon

³Department of Economic Sciences and Administration, Lebanese University, Lebanon

⁴Olayan School of Business, American University of Beirut, Lebanon

Abstract: Internet of Things (IoT) as a concept wasn’t officially named until 1999 where it still used by big computer and communication companies. It is the connection objects with each other anywhere, anytime, via internet communication without human intervention. Communication is the main part of IoT. With the development of technology and the revolution in a smart cell phone, the connected devices reach billions which lead to a fast increase in the transmitted data through the network. This rapid increase results in a heavy load on servers which need more processing and routing time. Fog Computing and Cloud computing paradigm extend the edge of the network, thus enabling a new variety of applications and services. In this paper, we focus on the modeling of the fog computing architecture and compare its performance with the traditional model. We present a comparative study with traditional IoT architecture based on classifying applications,define a priority for each application, and use the cell operator as the main fog center to store data. Then we givea solution to decrease data transmission time, reduce routing processes, increase response speed, reduce internet usages, and enhance the overall performance of IoT systems.

Keywords: Fog computing revolution, processing time, speed, reliability, and bandwidth drop.

Applications of Logistic Regression and Artificial Neural Network for ICSI Prediction

Zeinab Abbas¹, Ali Saad¹, Mohammad Ayache¹, and Chadi Fakih²

¹Department of Biomedical Engineering, Islamic University of Lebanon, Lebanon

²Department of medicine, Lebanese University and Saint Joseph University, Lebanon

Abstract: The third most serious disease estimated by Word Wide Organization after cancer and cardiovascular disease is the infertility. The advanced treatment techniques is the Intra-Cytoplasmic Sperm Injection (ICSI) procedure, it represents the best chance to have a baby for couples having an infertility problem. ICSI treatment is expensive, and there are many factors affecting the success of the treatment, including male and female factors. The paper aims to classify and predict the ICSI treatment results using logistic regression and artificial neural network. For this purpose, data are extracted from real patients and contain parameters such as age, endometrial receptivity, endometrial and myometrial vascularity index, number of embryo transfer, day of transfer, and quality of embryo transferred. Overall, the logistic regression predicts the output of the ICSI outcome with an accuracy of 75%. In other parts, the neural network managed to achieve an accuracy of 79.5% with all parameters and 75% with only the significant parameters.

Keywords: Artificial Neural Network (ANN), Assisted Reproductive Treatment (ART), In-Vitro Fertilization (IVF), ICSI, logistic regression.

An Ontology-based i* Goal-Oriented Referential

Integrity Model in Systems of Systems Context

²King Hussein Cancer Centre (KHCC), Jordan

Abstract: System of Systems (SoS) results from the integration of a set of independent Constituent Systems (CS) that could be socio or technical, in order to offer unique functionalities. SoS is largely driven by stakeholders’ needs and goals taking into consideration SoS-level global goals and CS-level individual goals. It’s challenging to manage the satisfaction of these goals in such complex SoS arrangements, where links between these goals may not be clearly known or specified, and competing goals establish a complex stakeholder environment. In this research the i* goal-oriented framework has been utilised in SoS context to identify, model and manage goals of the overall SoS and its constituent systems. This paper discusses a novel Goals Referential Integrity (GRI) model that is intended to maintain the integrity and consistency of both the SoS-level and the CS-level goals, in an attempt to address the current challenges of managing goals in an SoS arrangement. Furthermore, an ontology-based model has been developed to support the GRI model and semantically annotate goals’ levels in SoS context, specify the relationships and linkages between the SoS organisation, its constituent systems, global and local goals, and strategic and policy documents. Together the GRI model and its associated ontology model form the Semantic Goals Referential Integrity (SGRI) applied in SoS context, where conflicts between goals at the SoS and the CS-levels can be discovered in an attempt to maintain the semantic integrity of the SoS and CS goals.

Keywords: SoS, goal-oriented modelling, i* framework, GRI, ontology, semantic i* enrichment, cancer care informatics.

Deriving Object-Based Business Process

Architecture using Knowledge Management

Enablers

Abstract: This paper discusses a semantic-driven approach to deriving an object-based Business Process Architecture (BPA) using Knowledge Management Enablers (KMEs). The semantic enriched Riva BPA (srBPA) ontology has been selected as an object and ontology based BPA to be derived by the Abstract Knowledge Management Enablers’ Ontology (aKMEOnt). The aKMEOnt includes six KMEs: information technology, leadership, organisation structure, culture, business repository and knowledge context. The aKMEOnt has been utilised in order to generate the Essential Business Entities (EBEs) of the srBPA ontology. A link between these two artefacts, i.e., the srBPA ontology and aKMEOnt, is demonstrated using a typical example of the deposits department in banking. In conclusion, this new ontology-based approach between KMEs and BPA has informed the effectiveness of using semantic KMEs and Semantic Web Rule Language (SWRL) rules in the semi-automatic identification of representative EBEs. These EBEs characterise the business of deposits in banking and constitute the first essential building block of the Riva BPA method which drives the development of Units of Work and the subsequent 1st and 2nd cut Riva process architectures.

Keywords: Business process architecture, knowledge management enablers, srBPA, riva method.

Received October 6 2018; accepted January 22 2019

Full Text   

Autonomous Track and Follow UAV for Aerodynamic Analysis of Vehicles

Ahmad Drak and Alexander Asteroth

Department of Computer Science, Bonn-Rhein-Sieg University, Germany

Abstract: This work addresses the issue of finding an optimal flight zone for a side-by-side tracking and following Unmanned Aerial Vehicle(UAV) adhering to space-restricting factors brought upon by a dynamic Vector Field Extraction (VFE) algorithm. The VFE algorithm demands a relatively perpendicular field of view of the UAV to the tracked vehicle, thereby enforcing the space-restricting factors which are distance, angle and altitude. The objective of the UAV is to perform side-by-side tracking and following of a lightweight ground vehicle while acquiring high quality video of tufts attached to the side of the tracked vehicle. The recorded video is supplied to the VFE algorithm that produces the positions and deformations of the tufts over time as they interact with the surrounding air, resulting in an airflow model of the tracked vehicle. The present limitations of wind tunnel tests and computational fluid dynamics simulation suggest the use of a UAV for real world evaluation of the aerodynamic properties of the vehicle’s exterior. The novelty of the proposed approach is alluded to defining the specific flight zone restricting factors while adhering to the VFE algorithm, where as a result we were capable of formalizing a locally-static and a globally-dynamic geofence attached to the tracked vehicle and enclosing the UAV.

Keywords: UAV, flight zone, geofence, dynamic vector fields, aerodynamics.

Vertical Shuffle Scheduling-Based Decoder for

Joint MIMO Detection and Channel Decoding

Ali Haroun¹, Hussein Sharafeddin², and Ali Al-Ghouwayel¹

¹Computer and Communications Engineering Department, International University of Beirut, Lebanon

²Faculty of Sciences, Lebanese University, Lebanon

Abstract: This paper presents a novel architecture of a soft Non-Binary Low Density Parity Check (NB-LDPC) decoder for joint iterative Multiple-Input Multiple-Output (MIMO) receivers. The proposed architecture implements a single variable node processor where the Log Likelihood Ratio (LLR) computation block is removed. It also implements a single Check Node (CN) processor that is composed of six Elementary Check Nodes. The architecture is able to decode the rate R=1/2 with frame length N= 384Low Density Parity Check (LDPC) code using a 64 QAM modulation. To our knowledge, it is the first soft decoder architecture that implements the belief propagation algorithm based on vertical shuffle schedule. Synthesis results show that the proposed architecture consumes 6.476 K slices and runs at a maximum clock frequency of 70 MHz. Taking only the decoding process part alone, 188 clock cycles are required to perform decoding iterations.

Keywords: MIMO, Iterative Belief Propagation (BP), Joint Factor Graph, NB-LDPC, Vertical Shuffle Schedule (VSS).

Received October 7, 2018; accepted January 21, 2019

Full Text   

Oussama Zahwe¹, Hassan Harb², and Hussein Nasrallah¹

Abstract: This paper takes place in the field of passive microwave components. Coming from the requirements of mobile communication devices, miniaturization of microwave components is needed. Aimed at this objective, for several years we have been working on the development of a miniature planar circulator. The main aim of this paper is to show the circulator with different ferrite thickness and then to present a method to reduce the insertion loss of the circulator in function of the thickness of the ferrite film. The circulator is designed with new topology coplanar/microstrip. The analytical structure based on stripline circulator and analysed by using a three dimensional finite-element method. The circulator is then fabricated, and its properties in the microwave range are characterised using a network analyser and a probing system. An additional part for the isolator with non-symmetrical was designed in order to reduce the insertion loss in the conductor and to obtain a large bandwidth compared to the existing devices.

Keywords: Coplanar circulator/isolator, non-reciprocal passive component, ferrite film, miniaturization, insertion loss.