Monday  Jan 27th, 2020 ADVANCE 2020 Workshop (Cancun)	Tuesday Jan 28th, 2020 ADVANCE 2020 Workshop (Cancun)	Tuesday Jan 30th, 2020 ADVANCE 2020 Networking (Merida)
08:30-09:00	Registration/welcome campus Universidad Caribe	Registration/welcome campus Universidad Caribe
09:00-09:30	Opening from General Chair and TPC chairs	Opening from General Chair and TPC chairs	Welcome Campus Universidad Autónoma de Yucatán
09:30-10:10	Keynote  Talk 1	Keynote Talk 2	Research activities in Brazil/Canada/France
10:10-10:30	Break/Posters	Break/Posters	Break
10:30-11:50	Technical session 1	Technical session 6	Research activities in Mexico
11:50-12:20	Technical session 2 (SP)	Technical session 7 (SP)	Open collaboration discussions
12:20-13:30	Lunch break	Final remarks and closing	Lunch break
		Lunch break
13:30-14:50	Technical session 3	Short Professional  Courses	Visit of Campus
14:50-15:20	Technical session 4 (SP)
15:20-16:00	Break/Posters
16:00-17:10	Technical session 5	Visit of Campus
19:30-22:30	ADVANCE Gala Diner

KEYNOTES TALKS 1 and 2

Keynote Talk 1 : Wavelet-based information tolos for the analysis of computer communications traffic
By Professor Julio César Ramirez Pachero University of Caribe, Mexico

Abstract: In this talk, we take a brief overview of some wavelet-based information entropies which can be used to analyze self-similar traffic generated from computer communications networks. First, we present the stochastic processes which can be used to model computer traffic and then present the advantages of representing these processes with wavelets and wavelet transforms. We present some well known results of the wavelet analysis of self-similar signals and then construct various generalization of these entropies in the wavelet domain. Later, based on the wavelet spectrum of these processes we construct wavelet information planes from which several applications for the analysis of these processes can be achieved. We present some results in the detection of level-shift detection and self-similar processes classification

Bio: Prof. Julio César Ramìrez-Pacheco received his PhD Degree from CINVESTAV-IPN Unidad Guadalajara in 2013 where his research topics included Telecommunications software design and performance evaluation in computer networks. Current research of Prof. Ramirez-Pacheco is mainly focused on the generalization of complexity metrics to the wavelet domain for the analysis of scale-invariant stochastic processes. Other research topics include time series analysis and heavy-.tail processes.

Keynote Talk 2 : Blockchain  for AI
By Professor Hakim Abdelhafid, University of Montreal, Canada

Abstract: Blockchain can be simply defined as a distributed digital ledger that keeps track of all the transactions that have taken place in a secure, chronological and immutable way using peer-to-peer networking technology. The most known blockchain application is bitcoin that supports transactions of bitcoin transfer. However, blockchain is finding many uses in financial and non-financial applications; indeed, it is believed that blockchain will transform how we live,work, and interact. This talk will start with an introduction to Blockchain technology. It will show how Blockchain works. It will present the concept of smart contracts, which are critical in developing blockchain applications. It will also identify industry segments where blockchain can play a key role in solving existing issues. Then, the talk will address the integration of blockchain and AI; in particular, it will show how Blockchain can help advancing and democratizing AI. The talk will conclude by briefly presenting our work at Montreal Blockchain Lab and our view on the future of blockchain and its intersection with AI.

Bio: Dr. Hafid is Full Professor at the University of Montreal, where he founded the Network Research Lab (NRL) in 2005. He is also research fellow at CIRRELT, Montreal, Canada. Dr. Hafid published over 230 journal and conference papers; he also holds 3 US patents. Prior to joining U. of Montreal, he spent several years, as senior research scientist, at  Telcordia Technologies (formerly Bell Communications Research), NJ, US working in the context of major research projects on the management of next generation networks. Dr. Hafid has extensive academic and industrial research experience in the area of the management of next generation networks. His research interests include also cloud/mobile cloud/fog, IoT, blockchain and intelligent transport systems.

TECHNICAL SESSIONS

Technical session 1 - Monday, January 27^th, 2020 (Full Papers)
Topics: SDN/Cloud/Fog

Paper 1.1 Managing IEC 61850 Message Exchange for SDN-Controlled Cognitive Communication Resource Allocation in the Smart Grid
Yanny Moscovits, Eliseu Torres and Joberto S. B. Martins
Salvador University – UNIFACS, Brazil 

Abstract: The IEC 61850 standard is being largely used in the Smart Grid (SG) context mainly due to its ability to address communication, interoperability and migration issues. IEC 61850 currently aims at internal substation communication. Nevertheless, there is a demand to generalize its use for distributed SG systems like Home Energy Management Systems (HEMS) and Advanced Monitoring Infrastructure (AMI) Communication which potentially involves distributed substations or distributed SG components. IEC 61850-based systems require constrained timing requirements for communication and the common approach is to allocate static link bandwidth resources leading in some cases to over dimensioning. This paper presents the Substation Cognitive Communication Resource Management (IC2RM) that aims the management of bandwidth allocation for IEC messages using a cognitive approach for its provisioning and the SDN/OpenFlow for its deployment. By dynamically deploying bandwidth for IEC messages, IC2RM optimizes links between SG substations and systems and potentially reduces the operational costs (OPEX).

Paper 1.2  Inspiring from SDN to Efficiently Deploy IoT Applications in the Cloud/Fog/IoT ecosystem

Nada Chendeb¹, Nazim Agoulmine² and Mohammad El-Assaad^2,1
1Lebanese University, Faculty of Engineering, Lebanon
²University of Evry Val d'Essonne - Paris Saclay University, France

Abstract: Billions of devices are connected to the internet nowadays, more are coming in the future. Cisco assumes that 50 billion devices will be connected by 2020. It is quite difficult to manage and control the exchange of the huge amount of data generated from those connected devices. Thus the need for a new more intelligent Internet of Things (IoT) architecture for large scale networks. Based on the above needs and challenges of the IoT, Software Defined Networking seems to be an excellent key solution. SDN is a networking structure that reduces the complexity, and increases the flexibility and manageability of the networks by splitting data plane and control plane. Thus, using the concepts of SDN, we aim to reduce the complexity of traditional IoT networks, and form a more manageable network so networks can be programmed according to the network operator’s needs. This means that recent IoT networks will be more programmable, managed and controlled by software. In this work, we provide an architectural model combining SDN and IoT, and a mathematical formulation of the controller placement problem as a linear integer program optimization. At the end, we present same simulation results to show the advantages of this integration.

Paper 1.3 Performance analysis of computational offloading on embedded platforms using the gRPC framework
Mateus Araujo, Marcio Maia, Paulo Rego and José Neuman
Universidade Federal do Ceará (UFC), Brazil

Abstract: Embedded systems are becoming increasingly accessible to the Internet allowing the creation of new services and applications. Such systems need to communicate in a structured form in a way that uses standardized technologies for better results. Mobile systems also have a number of limited features like battery life, internal storage and processing performance. Such restrictions can be mitigated by the use of computational offloading since algorithms or applications can be executed in the cloud or other networked devices. This article is intended for the analysis of emerging technologies in cross-process communication between Linux and Android-based multiplatform using the gRPC framework. Applications have been developed in various object-oriented programming languages for performing remote procedure calls between a single-board computer and a personal-use smartphone for processing higher order arrays and applying filters to images. Then, a series of analyzes were performed on the transferred data and the computational offloading performance of the algorithms in each platform.

 

Technical session 2 - Monday, January 27^th, 2019 (Short WiP Papers)
Topics: Profiling/ML/Data integration/eHealth

Paper 2.1 A proposal methodological to Assure Quality using Data Profiling Techniques

César Guerra¹, Hector Perez-Gonzalez¹, Victor Menendez-Dominguez² and Reyes Juárez Ramírez^3 
1Universidad Autonoma de San Luis Potosi, Mexico
²Universidad Autónoma de Yucatán, Mexico
³University of Baja California, Mexico

Abstract: For any organization it is necessary to satisfy their business objectives besides using data to implement organizational processes, for that reason, it is indispensable to have knowledge of how these data satisfy the preset quality requirements. Thus, these requirements could be expressed by means of some data quality dimensions. In some scenarios, models and methodologies of data quality assessment require of mechanisms to control and monitor the level of quality of data. Thus, proposing a methodology with a qualitative diagnosis of the data quality dimensions and using data profiling techniques to measure some of these dimensions, will have a significant impact on the processes of appropriate use of the data. The main contribution of this paper is a methodology that assesses the data quality, by diagnosing its dimensions through a survey proposed and data profiling techniques.

Paper 2.2 Document Database for Scientific Production
Jared D.T. Guerrero-Sosa, Víctor Hugo Menéndez-Domínguez, María-Enriqueta Castellanos-Bolaños and Francisco Moo-Mena
Universidad Autónoma de Yucatán, Mexico

Abstract: The goal of this work is to present the use of a document database for the storage of information related to scientific production that can be retrieved through various digital repositories. The introduction briefly presents what the digital repositories of scientific production consist of, indexed publications, some tools for the recovery of the information of the products stored in the repositories and a description of the NoSQL databases. Subsequently, the MongoDB database manager and its differences with a conventional database are presented. Then the methodology is presented, which is based on the discovery of knowledge through data sets. Then the implementation of the methodology with the production stored in Scopus in a Mexican university is described. An overview of results are also presented, which are found in the proposed databases. Finally, the conclusions of the work are presented.

Paper 2.3 Machine Learning Supporting Brazilian Public Health Care policies
Raimundo Valter Costa Filho¹, Silas Santiago Lopes Pereira¹, Ronaldo F. Ramos¹, Luiz Odorico Monteiro de Andrade², Ivana Barreto² and Mauro Oliveira¹. 
¹Federal Institute of Ceara (IFCE), Brazil
²Department of Public Health - Fiocruz, Brazil

Abstract: Health data monitoring is a crucial activity to reduce maternal, neonatal and infant mortality rates by supporting public health policies decisions. Available data in Brazilian health databases point that It is possible to predict death risk in the early stages of gestation and infant development. In this research, we consider the information availability still in the gestational period to propose different death risk prediction models for this public of interest. We also detail the data mining process to apply machine learning-based techniques in death risk classification for maternal, neonatal and infant patients. We present an experiment pipeline to estimate average performance and evaluated machine learning models with different features combinations. Additionally, we show a web service which provides multiple predictive models by information availability. Results show Random Forest obtaining better performance when compared to the other machine learning methods.

Technical session 3 - Monday, January 27^th, 2020 (Full Papers)
Topics: Distributed control/Security/Blockchain

Paper 3.1 Plante: An Intelligent Agent-based Platform for Monitoring and Controlling of Agricultural Environments
Renato Alves de Oliveira, Reinaldo Bezerra Braga and Carina Teixeira De Oliveira
LAR/Federal Institute of Education, Science and Technology of Ceará (IFCE), Brazil 

Abstract: Several studies estimate a population increase for the coming decades that will result in a rise in global food demand. In this scenario, the primary sector, especially agriculture, must take a series of measures to modernize its production processes, aiming at reducing waste and increasing production levels. To effect this modernization, the sector must solve a series of problems related to its production processes, ranging from planning to quality control. Given the circumstances, this paper introduces Plante, which consists of a platform with a hardware module responsible for collecting environmental data and irrigating agricultural environments, and a mobile application, responsible for monitoring climate aspects, providing weather forecast, and alert the farmer when something is wrong with the plantation.

Paper 3.2   Mitigating Man in the Middle attacks within Context-based SDNs

Robson Gonzaga Silva and Paulo Nazareno Maia Sampaio. 
UNIFACS - Universidade Salvador, Brazil

Abstract: The application of experimental networks through hybrid proposals such as the combination of Context-Sensitive Networks and Software-Defined Networks (SDNs) is a promising approach due to the combination of its key benefits such as scalability, dynamism, flexibility, easy management, function control, programming and others. However, despite the many advantages of hybrid approaches and new network paradigms, the study of the literature reveals some existing security challenges, since in addition to introducing new vulnerabilities to the context of computer networks, they end up potentially leveraging existing vulnerabilities, which are more critical, increasing the risks of exploiting vulnerabilities by malicious users and cyber criminals. This paper aims at presenting a study about potential man-in-the-middle attacks in the context of hybrid networks, based on the CAARF-SDN project, with the objective of identifying and assessing risks related to Information Security aspects (Confidentiality, Integrity and Availability). Moreover, solutions are also conceived through the implementation and validation of security mechanisms in order to mitigate these attacks, enhancing network security and guaranteeing its features and services.

Paper 3.3 Smart Contract modeling and verification techniques: A survey
Adnan Imeri¹,², Nazim Agoulmine² and Djamel Khadraoui¹
1LIST - Luxembourg Institute of Science and Technology, Luxembourg
2University of Evry, Paris Saclay University, France

Abstract: The capabilities of smart contracts for supporting and enhancing business processes in distributed-decentralized environments have affected the technological transformation of numerous industries. Designing and developing blockchain-based solutions requires model checking and verification of the components of the system such as smart contracts, for well-behave, correct execution and fulfilling of the business process requirements. Certainly, there are concerns about the execution of smart contracts in such distributed environments. This study shows the research results about model checking of smart contracts, performing a deep analysis of current approaches on modeling and verifying smart contracts and reviewing available tools for such practices. Modeling and verifying smart contracts are addressed at the levels of programming and run time execution.

Technical session 4 – Monday, January 27^th, 2020 (Short WiP Papers)
Topics: Data management technologies/Interoperability

Paper 4.1 Heimdall: An Authorization Framework Based on Blockchain for Sensitive Data Access
Bruno Batista¹, José Neuman² and Joaquim Celestino^3
1Univerasidade de Fortaleza (UF), Brazil
²Universidade Federal do Ceará (UFC), Brazil
³Universidade Estadual do Ceará (UECE), Brazil

Abstract: The Internet of Things (IoT) is transforming the way that we interconnect the devices, collect data and make computation over these data. But to achieve the widespread adoption of IoT it‘s necessary to improve the security of the IoT. In this paper, we propose Heimdall, a distributed smart contract-based framework that provides access control for sensitive or personal data collected by IoT devices that follow the prerogatives of GDPR and LGPD laws.

Paper 4.2 Quality of Health Service, an architecture to optimizing an IoT solution with Diffserv and health protocol EWS.
David Viana¹, Raimundo Valter Filho¹, Wendell Oliveira¹, Odorico Monteiro² and Mauro Oliveira^1
1Federal Institute of Ceara (IFCE), Brazil
²Universidade Federal do Ceará (UFC), Brazil

Abstract: This work presents the Quality of Health Service, an IoT solution for a health patient monitoring environment and proposes an optimization mechanism with the Diffserv and EWS protocols. A mobile application is implemented for the specific healthcare team to have access to the system for viewing and modifying patient information. The analysis of vital signs in QhS (Quality of Health Service) solution took into consideration the network paradigm and IoT service, as well as the risk of the patient based on the EWS protocol. Thus, the QhS combines the Diffserv (network level) and EWS (application level) protocols for the optimization of data traffic in the system monitoring information and alerts. Additionally, It results in energy saving, still a vital resource in IoT devices.

Paper 4.3 Analysis of Interoperability in Public Health Systems
Leonardo Nascimento¹, Renato Freitas¹, Cesar Olavo¹, Ivana Holanda², Odorico Monteiro² and Mauro Oliveira¹. 
¹Federal Institute of Ceara (IFCE), Brazil
²Department of Public Health - Fiocruz, Brazil

Abstract: Health information systems (HIS) are constantly evolving in both complexity and data volume. Nevertheless, not every HIS share the same data standard or are otherwise able to communicate with one another. The need for communication among these systems has resulted in a worldwide effort to develop interoperability standards for healthcare systems. Brazil also has adopted measures to foster interoperability in public health services, notably with the issue of the ordinance 2.073 of August 31, 2011 by the Ministry of Health, which regulates the adoption of international standards. This article presents an analysis of the different types of interoperability in public health systems, using GISSA, an intelligent system to support decision making in maternal and child health, to showcase them. A prototype was implemented that addresses the problem of interoperability from a structural viewpoint, by aggregating new services to the GISSA legacy version and also from a semantic viewpoint, by enabling the coexistence, in one system, of the two main electronic health record standards, i.e., FHIR and OpenEHR.

Technical session 5 - Monday, January 27^th, 2020 (Full Papers)
Topics: Monitoring/Classification/Prediction

Paper 5.1 Kidney Failure Detection Using Machine Learning Techniques
Kílvia L. De A. Almeida, Lucília Lessa, Anny B. S. Peixoto, Rafael L. Gomes and Joaquim Celestino Júnior
State University of Ceará (UECE), Brazil

Abstract: Renal insufficiency is the loss of kidney function, which is responsible for the filtration of residues, salts and liquids present in the blood. Being considered a silent disease, as the symptoms are often only detected in advanced stages of the disease. Loss of kidney function can be measured by the Glomerular Filtration Rate, which represents the volume of fluid that is filtered into the Browman capsule, located in the glomerulus, per unit of time. It is an important indicator for detection, evaluation and treatment of kidney failure. According to a census conducted in 2018 by the Brazilian Society of Nephrology (SBN), the number of chronic dialysis patients in Brazil from 2002 to 2017 increased by 159.4%, while Acute Kidney failure has been showing a mortality rate around 50%. Early detection is a goal to be pursued by those dealing with public health. In this work, a system was developed to detect kidney failure early and thus increase the chances of treatment for these patients. For this, several Machine Learning techniques were used, where through these techniques were calculated their respective accuracy. The data were using the MIMIC-II database and it was shown that techniques such as decision trees and random forests provided good results and could be important life saving strategies.

Paper 5.2 The CAARF approach towards Monitoring and analysis of Contextual data within SDN networks
Constantino Jacob Miguel, Francisco Badaró Neto and Paulo N. M. Sampaio
UNIFACS, Brazil

Abstract: Software-Defined Networks (SDN) modified the way network traffic can be monitored and analyzed, allowing for new approaches for network optimization. This paper presents an approach for monitoring and analyzing context-based data flow considering the global state of the computational environment (users, presentation/communication devices, and network infra-structure). The proposed solution relays on the development and deployment of an API called Context-Aware Adaptive Routing Framework (CAARF-SDN). CAARF-SDN is a context-based service that provides the monitoring and optimization of the SDN traffic. In this paper we present the CAARF-SDN based monitoring through the implementation of a virtual network using Mininet in order to analyze contextual information such as Quality of Service (QoS), Quality of Device (QoD) and Quality of Experience (QoE). For this purpose, different scenarios are proposed in order to validate the so-called network Quality of Context (QoC) which is related to the global state of the computational environment and allows for the optimization of the network.

Paper 5.3 AuFa - Automatic Detection and Classification of Fake News Using Neural networks
Vinícius Nunes Barbosa, Carina Teixeira de Oliveira and Reinaldo Bezerra Braga
LAR/Federal Institute of Education, Science and Technology of Ceará (IFCE), Brazil 

Abstract: The increased proliferation of fake news on social networks has a significant impact on the information received by the society. The malicious use of information can compromise the democracy, as well as manipulate the opinions of people exposed to such news. These impacts have boosted new search directions in an attempt to classify and identify this news. Therefore, we propose AuFa, a solution to automatic detect and classify fake news through neural networks algorithms. Preprocessing techniques are used in the collected database to analyze the patterns of news, as well as to reduce noise and eliminate null information. The results obtained showed that the supervised neural network algorithm (MLPClassifier), obtained satisfactory performance to be used in the proposed solution.

Technical session 6 - Tuesday, January 28^th, 2020 (Full Papers)
Topics: Education Robotics/Machine Learning/IoT-BC

Paper 6.1 Educational robotics at K-12 schools in the southeast of Mexico
Cinhtia Gonzalez-Segura, Michel Garcia-Garcia, Jorge Rios-Martinez, Sergio Gonzalez-Segura and Luis Basto-Diaz
Universidad Autónoma de Yucatán, Mexico

Abstract: This paper presents a descriptive study about the use of robots in the teaching of mathematics for elementary children and the use of different models of educational robotic kits which were programmed to address issues such as the Cartesian plane, fractions, polygons and areas; performing operations of addition, subtraction, multiplication and division. During four semesters, fourteen elementary schools located in the eastern part of the State of Yucatan, were visited to perform activities with robots. More than two thousand children attending the six grades of the elementary education in Mexico, and their respective teachers, who number about one hundred participated. It was observed that the use of robots as new pedagogical tool greatly contributes to motivate students during the teaching-learning process of mathematic subjects.

Paper 6.2 Machine Learning models for the prediction of Wi-Fi links performance using a CityLab testbed
Paulo Marques
Instituto Politecnico de Castelo Branco, Portugal

Abstract: The Wi-Fi links performance depends in a highly complex way on the actual topology, channel qualities, spectral configurations, etc. Existing Wi-Fi radio link performance models usually adopt explicit and bottom-up approaches in order to predict throughput figures based on Markov chains and SINR levels. In this work we have validated a new approach for predicting the performance of Wi-Fi networks. Based on data measurements from the outdoor Wi-Fi CityLab testbed in Antwerp we have tested four different supervised learning algorithms. We observed that abstract “black box” models built using supervised machine learning techniques – without any deep knowledge of the complex interference dynamics of IEEE 802.11 networks – can estimate the link throughput with very good accuracy, reaching a value of R2-score of 90% for the case of the Gradient Boosting Regressor.

Paper 6.3 Integrating Blockchain with IoT for a Secure Healthcare Digital System
Nada Chendeb¹, Nour Khaled¹ and Nazim Agoulmine^2 
1Lebanese University, Faculty of Engineering, Lebanon
²University of Evry Val d'Essonne - Paris Saclay University, France

Abstract: In the past few years, the number of wireless devices connected to the Internet has increased to a number that could reach billions in the next few years. While cloud computing is being seen as the solution to process this data, security and scalability challenges could not be addressed solely with this technology. Blockchain technology is the technology that underpins Bitcoin, it introduces manners to provide a fully autonomous secure system, by using smart contracts. Multi-layer BC is a very powerful solution to overcome many IoT challenges. This paper illustrates how Blockchain technology works, what are the IoT challenges, and how it can be integrated with Blockchain. We proposed in this work a multi-layer IoT/blockchain based architecture customized and designed to be used in the medical field. With this information interact many parties including the doctors, healthcare service providers, insurance companies and pharmacies. The ultimate goal being to solve the problem of scalability and performance.

 

Technical session 7 – Tuesday, January 28^th, 2020 (Short WiP Papers)
Topics: QoS/QoE/Security/identification

Paper 7.1 Gesture Recognition in Sign Languages: Methods and Approaches
Allan Ojeda-Pat and Francisco Moo-Mena
Universidad Autónoma de Yucatán, Mexico

Abstract: In the world of computing, computer vision is a highly studied due to the great advantages it provides. For example, the recognition of objects by means of a photograph or using a camera in real time allows to extract important information that may be useful to the user. Using sign language is an effective way in which people with hearing troubles can communicate with any other person, either by emergency or by social inclusion. A problem arises due to the gap and the lack of interaction between deaf and hearing people, and the lack of effective systems capable of helping to minimize this gap. The main goal of this work is to analyze the methods and work done to recognize the sign language using some input source with information in video or images.

Paper 7.2 Deploying-in-Production of a connected object-user identification approach by recognizing physical activity in a Big Data environment
Hamdi Amroun¹ and Mehdi Ammi^2
1LIMSI, CNRS, France
²University of Paris 8, France

Abstract: In this paper, a method for identifying users of a connected object based on the recognition of the physical activities induced by the manipulation of these connected objects has been presented. Data from connected objects were first classified to recognize four types of activities: standing, sitting, lying and walking. These four activities thus classified generate four databases. Then, within each activity database, a classification model was trained to identify the different users. The model thus obtained is put into production on a Big data platform. User recognition results reached 73% accuracy in production. This study has provided a non-intrusive approach to recognizing users of a connected object in a "Big Data" environment.

Paper 7.3 Security Issues of Healthcare IoT Devices
Sadry Fievet and Karima Boudaoud 
Laboratoire I3S - Université Côte d’Azur/CNRS, France

Abstract: The recent adoption of connected objects in the health sector raises the question of security of personal data that are collected and transmitted by these objects. Actually, cybercriminals show an increasing interest in stealing and reselling personal data thanks to the illegal revenue they can benefit from. To evaluate the security measures implemented by these objects, we have conducted pentests on healthcare IoT devices targeting the mobile application designed for the connected objects and the network communications between the connected object and the smartphone of health professionals. The goal of this work in progress paper is to present the preliminary results of the pentests related specifically to network attacks.

Paper 7.4 Supporting security in a MCC framework
Francisco Gomes, Paulo Rego, Fernando Trinta, Windson Viana, José Macedo and José Souza 
GREat/Universidade Federal do Ceará (UFC), Brazil

Abstract: Mobile Cloud Computing (MCC) unites two complementary paradigms by allowing the migration of tasks and data from resource-constrained devices into remote servers with higher processing capabilities in an approach known as offloading. An essential aspect of any offloading solution is the privacy support of information transferred between mobile devices and remote servers. A common solution to address privacy issues in data transmission is the use of encryption. Nevertheless, encryption algorithms impose additional processing tasks that impact both the offloading performance and the power consumption of mobile devices. This paper discusses how we extended the frameworks CAOS and CAOS D2D to supporting encryption and presents initial results on the impact caused by encryption algorithms on the execution time of offloaded methods.

SHORT PROFESSIONAL COURSES:
 
Short Professional Course 1 : Introduction to IoT and MEC (Mobile Edge Computing)Technologies
By Professor Nazim Agoulmine, University of Evry – Paris Saclay University, France

Abstract: This talks aims to introduce the technologies that are changing the game namely IoT, Edge Computing (a.k.a. Fog Computing) and Cloud Computing and how these technologies are glued together along with novel data mining techniques to allow collecting and reasoning to better collect, interpret and predict information, make diagnosis and detect anomalies in large populations or geographic area. Applications are in all verticals of the society such as health, transportation, business, etc.