Belkacem Athamena

RESEARCH, TEACHING, or OTHER INTERESTS

Management Information Systems, Information Systems, Artificial Intelligence, Software

Scopus Publications

An Optimized SDN Framework for the Internet of Things
Zina Houhamdi, Mohamed Raid Athamena, Belkacem Athamena, and Shorouq Eletter
IEEE
Low-power wireless networks (LPWN) have traditionally been central to the discussion of the Internet of Things (IoT). Nevertheless, as these networks grow more complex, their control architectures and protocols reveal significant limitations, particularly when dealing with multi-hop topologies and lossy channels. To tackle these challenges, there has been growing interest in adopting Software Defined Networking (SDN), which has revolutionized data center and campus network management over the past decade by moving away from traditional vertical infrastructure. Despite its advantages, the centralized SDN model encounters substantial difficulties in the restricted settings of LPWN. The current study investigates the leverage of SDN concepts to control Industrial IoT dynamically and flexibly while focusing on minimizing and managing SDN overhead. This paper designs and implements a novel SDN architecture tailored for LPWN: Optimized SDN (OSDN), in addition to simulated, experimental, and analytical findings. The results highlight that OSDN meets the diverse and complex traffic demands of Industrial IoT applications throughout LPWN and that challenges in integrating SDN in limited IoT networks can be successfully addressed.


Big Data Potentials to Reduce Uncertainty in Strategic Decision-Making Process
Zina Houhamdi and Belkacem Athamena
IEEE
Decision–making under uncertainty has been investigated in different domains during the past few years to examine how people make decisions. Recently, big data emerged and is recognized as a promising approach to assist companies in improving the quality of their choices. Several organizations face problems in taking advantage of big data tools. For this reason, it is important to study this topic further. Accordingly, this paper investigates how big data tools can help decision-makers reduce uncertainty in strategic decisions.This study reviews the literature on Strategic decision–making processes (SDMP) and big data and provides a complete view of how these concepts are connected and the impact of big data tools on the quality of the decision. After a comprehensive literature analysis, a model is proposed to solve this problem by analyzing the SDMP-making process to clarify the uncertainty in the decision–making process steps and assess how big data tools can be used by administrators in the decision-making process.The analysis's result discloses the potential for improving the strategic decision–making process under uncertainty. It also reveals that the uncertainty inherent to technology usage influences decision-makers' choices. Accordingly, a framework is proposed to assess the incorporation of big data analytics tools in companies to fully benefit from these tools' power.


Decision Support in Goal-Oriented Requirements Engineering
Zina Houhamdi, Mohamed Raid Athamena, Belkacem Athamena, and Ghaleb ElRefae
IEEE
The requirement engineering process starts with an abstract description of the user needs or problems and ends with a detailed and precise specification of these user needs. Therefore, the analyst needs to decide several times during the requirements engineering. The incorrect or inadequate decisions cause software failure or software that partially fulfills the user requirements. Consequently, it is mandatory to integrate the decision-making process at the beginning of software development. In other words, the tradeoffs can be discovered in the software requirements phase, and the alternatives can be identified. Furthermore, the decision-making process requires special attention since it applies to all requirements engineering activities, such as requirements collection, negotiation, prioritization, and planning. This paper emphasizes the advantages of integrating decision-making in Goal-Oriented Requirements Engineering (GORE) and mapping the decision-making framework to GORE. Also, the paper discusses whether the decision-making process should be included in one of the GORE phases or an ongoing process covering all GORE phases.


AI in Diagnostic Imaging: An Overview
Zina Houhamdi, Mohamed Raid Athamena, and Belkacem Athamena
IEEE
Incorporating Artificial Intelligence (AI) into medical image diagnoses transforms medical diagnostics by improving efficiency, accuracy, and patient outcomes. AI claims huge assurance for promoting medical image diagnoses. However, its effective implementation necessitates attentive attention to practical and ethical issues. This paper provides an overview of the revolutionary influence of AI technologies on pathology and radiology, emphasizing the main improvements, advantages, and challenges that are considered to achieve their full potential.


Impact of religious tourism on the economic development, energy consumption and environmental degradation: evidence from the Kingdom of Saudi Arabia
Mosab I. Tabash, Umar Farooq, Ghaleb A. El Refae, Mamdouh Abdulaziz Saleh Al-Faryan, and Belkacem Athamena
Emerald
Purpose Saudi Arabia is the main destination of religious tourism, as it has many spiritual places. With the passage of years, the figures for pilgrim visits are increasing, which is contributing to the economic growth of the Kingdom of Saudi Arabia (KSA). However, pilgrims’ visits can create strong opportunity costs in the form of environmental degradation. Owing to these notions, this study aims to discover the impact of religious tourism on the quality of the natural environment of Saudi Arabia. Design/methodology/approach This study develops the empirical relationship between the variables by sampling the data from 35 years ranging from 1986 to 2020. The regression among variables was checked by using fully modified ordinary least square and dynamic ordinary least square models. Findings This analysis proves that religious tourism has a direct impact on the environmental degradation of KSA. The unceasing visits of pilgrims accelerate various economic operations and activities, e.g. assimilation and digestion of industrial products, that necessarily hamper the environmental quality. In addition, this analysis indicates a negative impact on financial development, foreign investment and renewable energy consumption while the positive impact of fossil fuels assimilation and economic expansion on the secretion of CO2. The statistical findings are robust and verify the pollution halo hypothesis while rejecting the Environmental Kuznets Curve model in this region. Research limitations/implications This analysis recommends restructuring the policies on hajj and Umrah visits. KSA Government should ensure green consumption by pilgrims. The limitation on pilgrims’ visits and the introduction of quotas are alternative policies to impede the pollution in this region. Originality/value By controlling the routine determinants, this study offers innovative thoughts regarding the consequences of religious tourism on environmental quality.


Retention contracts with asymmetric information: optimistic approach vs pessimistic approach
Belkacem Athamena, Zina Houhamdi, and Ghaleb A. ElRefae
Emerald
Purpose This paper aims to focus on the utilization of retention contracts to screen and discipline managers in a context in which the council, board of directors, possesses incomplete information about the consequences of managers’ decisions. The analysis enlightens us on empire building, on the slight connection between achievement and firing, and describes concerns about the belief that low achievements result from bad managers. Design/methodology/approach This paper analyzes a basic model to show the resulting dilemmas. The desire to screen managers to enhance the organization's future well-being motivates managers to show their credentials by becoming excessively active. The council can address this bias by firing a manager whose project is proven to ruin value. Moreover, the council can replace the manager if he has implemented a project but its outcomes remain unobservable. Both decisions decrease the attraction to develop loss-generating projects. However, the dismissing decision on either ground will affect the council deduction that the expected competence of the incoming manager is lower than that of the dismissed manager. Findings This study shows in which situation the selection option is preferred over the disciplining option using two different retention contracts: optimistic contract and pessimistic contract. Originality/value This study shows in which situation the selection option is preferred over the disciplining option using two different retention contracts: optimistic contract and pessimistic contract.


Load Balancing Algorithms for Software-Defined Networks
Zina Houhamdi and Belkacem Athamena
IEEE
This paper proposes two different algorithms for load balancing in the data plane of the Software Defined Networking (SDN) framework. The first algorithm emphasizes the Flexible Connection Handoff (FCH) that considers the maximum load and the difference of loads between the Access Points. The load represents the number of connected stations to a particular Access Point. The proposed algorithm optimizes the load between Access Points by considering the fairness index. The second algorithm describes a Quality of Service Load Balancing approach (QLB) for a High-Density software-defined Wi-Fi network by designing a module at the Access Points side and another module at the controller side. The Access Points are the OpenFlow Access Points (OFAPs) that push the capacity and load information of the OFAPs toward the controller side. The OFAPs module controls the OFAPs connection and reconnection with minimum controller intervention. The OFAPs module dynamically sets the load level based on the network's current state. The destination Access Point for reconnection is selected based on three metrics, for instance, Throughput, Received Signal Strength Indicator, and Packet Loss Rate. After receiving the OFAPs reports, the controller module makes decisions about load balancing by getting a complete view of the network condition. The integration of SDN into Wireless networks allows the development of load-balancing algorithms for control and data planes.


Smart Contracts and Blockchain-Based Tools for Privacy-Preservation
Zina Houhamdi, Belkacem Athamena, and Ghaleb ElRefae
Springer Nature Switzerland



Flexible Connection Hand-off: A Software-Defined Networks Load Balancing Algorithm
Belkacem Athamena and Zina Houhamdi
IEEE
This paper proposes a new load-balancing algorithm for Software Defined Networking (SDN) framework. The algorithm emphasizes the Flexible Connection Hand-off (FCH) that considers the maximum load and the difference of loads between the Access Points. The load represents the number of connected stations to a particular Access Point. The proposed algorithm optimizes the load between Access Points by considering the fairness index.


Formal Approach to Data Accuracy Evaluation
Athamena Belkacem and Zina Houhamdi
Slovenian Association Informatika



A Smart Approach using Multi-agent System for Big Data Security
Dounya Kassimi, Okba Kazar, Ezedin Barka, Abdelhak Merizig, Zina Houhamdi, Belkacem Athamena, and Meftah Zaoui
IEEE
With the evaluation of technology and the appearance of new tools that help store the information we create, especially in Banking, business intelligence, and even Education as Datawarehouse, Big data, and cloud computing. Those new tools create another obstacle: how we can secure and protect the information and the data stored in them. This paper treats the problem of Big Data security and privacy using mobile and stationary agents’ technologies. The main important security proprieties in big data are integrity, authentication, privacy, and access control. For integrity, the problem lies in checking the integrity of data and if this data is good and can be used since the Big Data is receiving data from different sours and different formats (structured, semi–structured, and non-structured). In access control, we need to consider the users’ secrecy, monitor the authorities, and properly apply the confidentiality requirement. The problem with authentication is to authenticate each user over the network, while big data collect sensitive data from trusted or untrusted users. While the privacy policy problem revolves around data collection and the use of transparency. To answer the previous needs for security in big data, we have proposed a smart approach using multi–agent systems (MAS) as a model for our solution.


A New Approach Based on a Multi-Agent System for IDS in Cloud Computing
Dounya Kassimi, Okba Kazar, Ezedin Barka, Abdelhak Merizig, Zina Houhamdi, Belkacem Athamena, and Meftah Zaoui
IEEE
Cloud computing is a recent innovation in the IT industry that is expanding quickly. Furthermore, this technology is widely used to provide computation, data storage, and other resources remotely through the web on a pay-per-usage basis. It is now the favored option for any IT firm since it increases its capacity to satisfy the computing requirements of its everyday operations through scalability, mobility, and flexibility at a low price. But there are two key problems with cloud computing. The biggest issue is storage-related, and Google has addressed it by adding a new layer to the cloud dubbed “Big data as a service (BDaaS).” The second problem is security and privacy. The Intrusion Detection System (IDS) has become the most widely utilized component of computer systems, security, and compliance processes, safeguarding network-accessible Cloud resources and services from various threats and assaults. This study examines IDS approaches in Cloud Computing and big data sets. To identify anomalous data in BDaaS, we also suggested a sensible intrusion detection system (SIDS) based on the autonomic system. The agent contributes the most to introducing more proprieties, particularly the autonomy element.


Cloud Ubiquitous Learning Approach Based on Multi-agents System
Manel Guettala, Saad Harous, Samir Bourekkache, Belkacem Athamena, Okba Kazar, and Zina Houhamdi
IEEE
Information technology (IT) plays an important role in the education field, with ubiquitous computing technology the ubiquitous learning (U-Iearning) is now a practical possibility. Cloud computing can offer many benefits to the learning field, as it provides all resources such as software (SaaS), platforms (PaaS), and infrastructure (IaaS) as a service. The multi-agent systems approach is an excellent way to implement flexible systems whose structure evolves dynamically. Based on the discussion of the benefits of cloud computing and the multi-agent systems approach to support adaptive and personalized learning content, the paper presents a hybrid approach based on cloud computing that uses the multi-tenant architecture where tenants share the system software. Multi-agent system is used in this approach to support a ubiquitous learning system. A multi-tenant SaaS application addresses the unique requirements of each tenant by enabling customization at the tenant level with the adoption of the concept ofvirtualization (virtual machines). This paper explores the challenges of achieving performance isolation in the context of multi-tenant SaaS applications. As an output of this study, U- learning on the cloud-based multi-agent system architecture has been proposed. To achieve lower cost of resources and server allocation and provide isolation among learners in the ubiquitous learning environment.


Two-Sided Matching under Incomplete Information
Zina Houhamdi, Belkacem Athamena, and Ghaleb ElRefae
IEEE
In many contexts, stakeholders' preferences are exploited in decision-making. Because of its countless applications in business and the huge number of involved questions, such context has received substantial attention in different domains such as economics, political science, philosophy, and in recent years, computer science. Despite a considerable literature body that studied this kind of context, most efforts assume the availability of precise and complete information about the stakeholders' preferences needed by the decision-making process. Nevertheless, this assumption is invalid because of the confidentiality issues and immense cognitive burden. The target of this study is to formally discuss these restrictions by focusing on prior studies that look at dealing with partial information and proposing solution notions and concepts that assist the development of methods and algorithms that work with inaccurate and partial information in multiple contexts. The paper focuses on the decision-making process under partial information. At the begging, the study address informally the following question: under partial information about the stakeholder preferences, how can we develop an algorithm that is ‘good’, in other words, an algorithm that produces “good” results regarding the complete intrinsic preferences. The paper looks at this problem in a modified version of the two-sided matching problem and shows how to design an approximately-powerful algorithm in such contexts.


Open-SBS: Smart Building Simulator
Houssem Eddine Degha, Fatima Zohra Laallam, Okba Kazar, Issam Khelfaoui, Belkacem Athamena, and Zina Houhamdi
IEEE
With the rise of machine learning and deep learning techniques in recent years, a representative dataset has become an inspiring source of prediction and knowledge extraction. Furthermore, ambient intelligence environments have emerged as one of the most important research areas. Many complex tasks and challenges face this area to validate and evaluate field research. One of the major challenges in the smart building domain is the lack of a simulation tool capable of simulating complex contexts, executing semantic rules, and collecting representative datasets for experiments. To address these issues, we present in this paper a cross-platform, open-source Smart Building Simulator (Open-SBS). Open-SBS offer an opportunity for researchers in the field of Ambient Intelligence environments to simulate complex contexts and collect representative datasets, save and share their models of experiments. We propose a new simulation process that is divided into four distinct phases: designing, scenario creation, semantic rule addition, and data-set generation. On the System Usability Scale, we conducted a study to assess the ease of use of Open-SBS (SUS).


Blockchain Technology for Secure Shared Medical Data
Mhamed Mancer, Khelili Mohamed Akram, Ezedin Barka, Kazar Okba, Slatnia Sihem, Saad Harous, Belkacem Athamena, and Zina Houhamdi
IEEE
A blockchain is a shared database allowing trust to be created between individuals without using intermediaries. The architecture here is decentralized, the data is distributed among the users, and therefore the information can never be erased. Today, many fields are interested in the development of products and technical solutions based on Blockchain technology. In this paper, we investigate blockchain in developing a Secure Shared Medical Record system (SSMR) for health data management. The proposed system aims to collect, store and share electronic medical records securely. It also makes it possible to provide doctors, for example, with medical information from other healthcare professionals, by defining a medical profile for each patient.


Deep Learning for Single-Channel EEG-Based Driver Drowsiness: Comparative Study
Imene Latreche, Sihem Slatnia, Okba Kazar, Saad Harous, Belkacem Athamena, and Zina Houhamdi
IEEE
To prevent car accidents caused by drowsiness, a transitional state between alertness and sleepiness, numerous authors have conducted studies employing various monitoring approaches and detecting techniques. However, the Electroencephalogram (EEG) was the most used monitoring approach due to its advantages compared to other techniques. Also, Deep Learning (DL) techniques have been widely used in this context and have given prominent results. In this paper, we have established a comparative study between four deep learning methods. The aim of this comparison is to determine the optimal DL model for detecting EEG-Based Drowsiness using a small balanced dataset. We evaluated the performance of the models using the hold-out method and the Leave-One-Subject-Out Cross-Validation (LOSO CV) techniques. With an accuracy of 77% with the Hold-Out method and 70.31% with the LOSO CV method, the GRU model outperforms other deep learning and machine learning models, followed by the BI-LSTM with 74.88% and 68.07, respectively.


Fuzzy Logic and Deep learning Techniques for Covid-19 Detection
Belkis Hassani, Khelili Mohamed Akram, Kazar Okba, Slatnia Sihem, Saad Harous, Belkacem Athamena, and Zina Houhamdi
IEEE
With the development of ICT and its adoption in various domains, it gained remarkable intention in the healthcare sector which introduce the telemedicine term. The coronavirus pandemic has created several challenges for researchers to develop an accurate and fast detection system. In this paper, we present a new telemedicine application to predict Covid-19 using CNN and Fuzzy set techniques. The evaluation of the system indicates high performance with a 98% F1 score, 99% of recall, 98% for precision, and 97% of accuracy.


Optimistic Retention Contract Evaluation
Belkacem Athamena, Zina Houhamdi, and Ghaleb El Refae
Springer International Publishing



Retention Contracts under Partial Information Electoral Competition Case Study
Zina Houhamdi, Belkacem Athamena, and Ghaleb El Refae
Zarqa University
This study copes with a class of principal-agent problems where information asymmetry represents an important characteristic. The paper examines the relationship between the principal and agents. The principal has to perform two agents’ screening and discipline tasks. To complete his duties, the principal lacks complete information concerning the agents’ behavior and rarely has partial information regarding the failure or success of launched tactics, alliances, rationalization, etc. We analyze the type of retention contracts (implicit) used by the principal to replace or retain agents. Consistent with literature findings, we demonstrated that agents could be extremely active in showing their competencies; the relationship between dismissal and bad performance is invalid; and occasionally, the principal dismisses qualified agents. Then we determined the rules under which electorates urge political parties to acquire information and choose optimal policies from the voter’s viewpoint.


Cognitive and Autonomic IoT System Design
Belkacem Athamena and Zina Houhamdi
IEEE
Currently, the Internet of Things (IoT) usage observes a drastic growth in several areas and participates in the rapid universe digitalization. Henceforward, the IoT systems next generation will be more difficult to develop and monitor. Gathering real-time data created by IoT triggers some novel opportunities for businesses to take at the right time more accurate and precise decisions. However, several challenges (such as IoT systems complexity and heterogeneous data management, and IoT system scalability) restrain the elaboration of IoT systems that are smart and impel business decision-making. This paper proposes to automatize IoT systems management using an autonomic computing approach. Nevertheless, autonomic computing is insufficient for smart IoT systems development. Actually, a smart IoT system implements cognitive abilities that allow its learning and decision-making in real-time. Therefore, this study proposes a model for designing smart IoT systems. It defines a set of cognitive design patterns that delineate the dynamiccooperation between management processes (MPs) (to handle the requirements evolvement and the system's environment unpredictability) and add cognitive capabilities to IoT systems (to generate new insights, perceive big data, and communicate with users). The study's primary goal is to support the developer in designing smart IoT systems that are flexible by choosing an appropriate pattern (or a set of patterns) to meet complex system requirements.


IoT Framework for Effective and Fine-Grain Access Control
Zina Houhamdi and Belkacem Athamena
IEEE
The standardized protocols and new generation of hardware for communications increased interoperability more than ever. However, the increasing interoperability causes a rise in offensives from vicious people and hardware. Therefore, there is a necessity to apply encryption algorithms to guard the communications between clients and servers. Nevertheless, the current encryption techniques do not protect the service access at a fine-grain level. Furthermore, in wireless sensors and actuators, each network endpoint is integrated constrained–resource; thus, the interoperability increase necessitates a high computation rate. On the other hand, the endpoints inherent to processing and memory restrictions negatively affect communication delays and power consumption, leading to a shorter battery lifetime. Consequently, there is a need for new methods to increase interoperability, dependability, scalability, security, and energy efficiency. This study proposes a theoretical design of a new and effective IoT model that supports authentication, authorization, and fine grain access control with no network configuration and dynamic reconfiguration. The proposed framework demonstrates the possibility of the integration of IoT devices powered by batteries and a functional System of Systems.


Retention contracts under hidden information
Belkacem Athamena, Zina Houhamdi, and Ghaleb El Refae
IEEE
This paper focuses on the utilization of retention contracts to screen and discipline managers in a context in which the council, board of directors, possesses incomplete information about the consequences of managers’ decisions. The analysis enlightens us on empire building, on the slight connection between achievement and firing, and describes concerns about the belief that low achievements result from bad managers. This paper analyzes a basic model to show the resulting dilemmas. The desire to screen managers to enhance the organization's future wellbeing motivates managers to show their credentials by becoming excessively active. The council can address this bias by firing a manager whose project is proven to ruin value. Moreover, the council can replace the manager if he has implemented a project, but its outcomes remain unobservable. Both decisions decrease the attraction to develop loss-generating projects. However, the dismissing decision on either ground will affect the council deduction that the expected competence of the incoming manager is lower than that of the dismissed manager. This study shows in which situation the selection option is preferred over the disciplining option using a pessimistic contract.


A website application for IoT devices management
Zina Houhamdi and Belkacem Athamena
IEEE
The Internet of Things (IoT) technology connects multiple types of physical devices with different functionalities, communication protocols, and capabilities. With the advance progresses in IoT technology, the provision of a platform that allows the users of IoT to communicate and interact directly with their connected devices and manage and control them easily and smoothly through that platform from anywhere at any time becomes a necessity. This will protect the users’ privacy when they use their IoT devices. This paper aims to develop a website application that addresses these challenges and provides real-time data management. As a result, we designed a user-interface prototype, which demonstrates the idea of IoT website manager and provides a trial version of the implementation. Since the suggested platform is aimed to improve the user’s insight about the IoT devices, the proposed prototype creates a platform that enables the user to control different types of IoT devices remotely. Also, the website prototype indicates that it is a user-friendly platform, can be used effortlessly, and does not need any technical experience. Users can access and collect information concerning their connected IoT devices and monitor, manage and control them.


Multi-agents collaboration in open system
Zina Houhamdi and Belkacem Athamena
Zarqa University
Share constrained resources, accomplish complex tasks and achieve shared or individual goals are examples requiring collaboration between agents in multi-agent systems. The collaboration necessitates an effective team composed of a set of agents that do not have conflicting goals and express their willingness to cooperate. In such a team, the complex task is split into simple tasks, and each agent performs its assigned task to contribute to the fulfilment of the complex task. Nevertheless, team formation is challenging, especially in an open system that consists of self-interested agents performing tasks to achieve several simultaneous goals, usually clashing, by sharing constrained resources. The clashing goals obstruct the collaboration's success since the self-interested agent prefers its individual goals to the team’s shared goal. In open systems, the collaboration team construction process is impacted by the Multi-Agent System (MAS) model, the collaboration’s target, and dependencies between agents’ goals. This study investigates how to allow agents to build collaborative teams to realize a set of goals concurrently in open systems with constrained resources. This paper proposes a fully distributed approach to model the Collaborative Team Construction Model (CTCM). CTCM modifies the social reasoning model to allow agents to achieve their individual and shared goals concurrently by sharing resources in an open MAS by constructing collaborative teams. Each agent shares partial information (to preserve privacy) and models its goal relationships. The proposed team construction approach supports a distributed decision-making process. In CTCM, the agent adapts its self-interest level and adjusts its willingness to form an effective collaborative team.