AI-Driven Approaches to System Requirements, Test Case, and Code Generation: A New Paradigm in Software Engineering
Luay Tahat, Gulf University for Science and Technology, Kuwait
|
|---|
|
|---|
| Dr. Luay Tahat is an Associate Professor in the Department of Computer Science and Management Information Systems at Gulf University for Science and Technology (GUST), Kuwait, where he has served since 2008. Prior to joining GUST, he worked as Lead Mobile Network Solution Architect at Alcatel-Lucent (now Nokia) in Naperville, USA, and as a part-time professor at the Illinois Institute of Technology (IIT), Chicago, USA.
In addition to his academic career, Dr. Tahat brings over 15 years of professional and industrial experience with leading organizations such as AT&T, Lucent Technologies/Bell Labs, Alcatel-Lucent (Nokia), and IBM, spanning both mobile and fixed networks. During his time in the industry, he held several key roles, including software developer for advanced GSM features, systems engineer for VoIP, systems architect for 3G Mobile Switching Centers, lead solution architect for mobile solutions, and business analyst for consumer VoIP (cVoIP) and IPTV services. He was also part of the team that designed and architected Alcatel-Lucent’s 3rd Generation Mobile VoIP switches and Dual Mode Service Solutions.
Dr. Tahat has an extensive research record, with more than 50 publications in international conferences and journals. His work primarily focuses on software engineering and wireless technologies.
.
He holds a master’s degree in computer science from Northeastern Illinois University, Chicago, and a Ph.D. in Computer Science from the Illinois Institute of Technology (IIT), Chicago (2007). His research interests include software engineering, model-based testing, test suite prioritization, mobile network solutions, system impact analysis, and network management architecture.
.
|
| System modeling is a crucial practice in software engineering, widely applied in sectors such as telecommunications, healthcare, and automotive to design complex state-based systems. In some domains, system models are provided by standardization bodies, for example, in telecommunications, to help companies build systems that comply with established standards. Traditional methods for deriving specifications from these models are labor-intensive and error-prone, creating major bottlenecks in development and introducing risks that are unacceptable in high-reliability industries.
This talk introduces an AI-driven approach to enhance these processes by automating the generation of requirements, validation artifacts, and code from system models. With recent advances in artificial intelligence, generative models such as ChatGPT can perform these translations directly, ensuring adherence to industry standards while optimizing the development lifecycle. Case studies demonstrate how AI streamlines the path from modeling to implementation, showing substantial gains in efficiency and product quality.
The effectiveness of this approach is evaluated against key metrics such as development speed, error reduction, compliance with quality standards, and system robustness, with results indicating significant improvements. Rather than replacing established practices, AI supports engineers by accelerating workflows and helping manage increasing system complexity.
The transformative role of AI in system modeling reflects a paradigm shift in software engineering, advocating a dynamic partnership between human expertise and AI capabilities. In conclusion, we recommend further research into AI integration, highlighting potential areas for deeper investigation and future technological advancements.
|
Introduction to Quantum Computing
Walid Gomaa, Egypt Japan University of Science and Technology, Egypt
|
|---|
|
|---|
Walid Gomaa is a distinguished Professor of Computer Science and Engineering at the Egypt-Japan University of Science and Technology (E-JUST), where he has been a driving force in research, education, and innovation since 2010. With a Ph.D. in Computer Science from the University of Maryland, College Park (2007), and a postdoctoral fellowship at INRIA Loria Lab in France, his academic journey spans theoretical computer science, machine learning, and applied AI. His research interests are diverse, encompassing human activity recognition from sensory data (RGB, thermal, depth imaging, and inertial motion streaming), Arabic natural language processing (including social media and lyrics analysis), and applications of machine learning in video analytics, data analysis, and time series analysis. He also delves into theoretical computer science, focusing on computable analysis, descriptive complexity, and computation over metric and topological spaces.
Walid’s professional experience is marked by leadership in both academia and industry. He has held visiting and research positions at prestigious institutions such as Loria Lab (France), Osaka University (Japan), and Rice University (USA). His contributions extend beyond research, as he has played a pivotal role in mentoring over 14 Ph.D. students, 7 M.Sc. students, and numerous undergraduates, fostering the next generation of researchers and innovators. His dedication to education is evident through his extensive teaching portfolio, covering advanced topics like machine learning, natural language processing, cryptography, and quantum computation at both undergraduate and postgraduate levels.
Walid is also a prolific researcher, with over 90 publications in top-tier journals and conferences, and several books and patents to his name. His work has been recognized with multiple awards, including the Best Ph.D. Dissertation Award for his student and grants from international agencies such as JICA, STDF, and ITIDA. Beyond academia, he is actively involved in public outreach, creating educational content on platforms like YouTube to democratize knowledge in machine learning, cryptography, and quantum computing.
|
| Quantum computing represents a paradigm shift in the way information is processed, moving beyond the deterministic limits of classical computation. This talk introduces the fundamental principles underlying quantum computation, beginning with the nature of quantum states and the four postulates of quantum mechanics. Core concepts such as superposition, entanglement, and measurement are illustrated through the qubit model and Bloch sphere representation. The gate model is then explored as the quantum analogue of classical logic circuits, highlighting how unitary transformations enable quantum parallelism and novel algorithmic behaviors. Key examples include entangled states, teleportation protocols, and the exponential speedup in prime factorization afforded by Shor’s algorithm. By examining the architecture of quantum circuits and the characteristics of quantum algorithms, the talk emphasizes both the theoretical foundations and transformative potential of quantum computing in cryptography, optimization, and beyond.
|
Towards Sustainable 6G: The Role of Non-Terrestrial Networks in Intelligent Green Communication Systems
Metin Ozturk, Ankara Yıldırım Beyazıt University, Turkey
|
|---|
|
|---|
Metin Ozturk (Senior Member, IEEE) is an Assistant Professor of Telecommunications at Ankara Yıldırım Beyazıt University (AYBU), Türkiye, where he also serves as the Director of the AYBU Non-Terrestrial Networks Laboratory (AYBU NTN-Lab). From 2023 to 2024, he served as a Visiting Professor at the Non-Terrestrial Networks Lab at Carleton University (Carleton NTN-Lab), Canada. He received the B.Sc. degree in Electrical and Electronics Engineering from Eskisehir Osmangazi University, Türkiye, in 2013; the M.Sc. degree in Electronics and Communication Engineering from Ankara Yıldırım Beyazıt University, Türkiye, in 2016; and the Ph.D. degree from the Communication, Sensing, and Imaging (CSI) Research Group in the James Watt School of Engineering, University of Glasgow, UK, in 2020. His research interests include wireless communications, with a particular focus on artificial intelligence-driven mobile networking, non-terrestrial networks (NTN), and sustainable wireless network design. He has co-authored nearly 60 publications in leading journals, conferences, and book chapters. He has delivered more than 15 talks, including keynotes, seminars, tutorials, and invited talks, at national and international events such as the IEEE International Smart Cities Conference, the IEEE Signal Processing and Communication Applications Conference, and the IEEE International Conference on Communications (ICC). He has also organized a workshop at flagship IEEE conference IEEE WCNC and served as a guest editor for special issues in leading international journals.
|
| The tutorial addresses the role of non-terrestrial networks (NTN) in advancing intelligent and green communication systems for 6G and beyond. It covers the integration of satellites, high-altitude platform stations (HAPS), and uncrewed aerial vehicles (UAVs) with terrestrial networks to achieve sustainable, resilient, and inclusive connectivity. The content includes the concept of multi-layer cell switching, where users and traffic demands are dynamically distributed across terrestrial and non-terrestrial layers to enhance energy efficiency, manage latency, and maintain service continuity in diverse mobility scenarios.
The tutorial also includes the contribution of artificial intelligence and generative AI in improving the adaptability of these systems through intelligent resource allocation, data compression, mobility prediction, and context-aware orchestration. It highlights how NTN elements powered by renewable energy can reduce carbon emissions, expand coverage to underserved regions, and support critical applications such as disaster recovery and green mobility services.
Illustrative examples and case studies are incorporated to demonstrate the potential of NTN-enabled architectures in reducing network power consumption, optimizing quality of service, and enabling sustainable operations. The tutorial provides a broad perspective on NTN as an enabler of intelligent green mobility systems, combining technical advances with sustainability objectives to shape the design of future communication infrastructures.
|
The Contour-FFT
Shambhu Nath Jha, Thales Group, Belgium
|
|---|
|
|---|
Dr. Jha Shambhu Nath (S'06-M'14-SM'20) holds a PhD Degree in Engineering Sciences from UCLouvain, Belgium (2014), an MSc Management (Great Distinction) from VUB, Belgium (2016), Dual master’s degree: Master’s degree in Electrical Engineering (Distinction) from UCLouvain (2009), European Master of Research on Information and Communication Technology from UPC, Spain (2008) and a BSc in EE (Distinction) from Institute of Engineering, Tribhuvan University, Nepal (2001). Dr. Jha is currently working as an Electromagnetics Specialist at Thales Belgium SA. His current interests include metasurfaces, surface-based electromagnetics, optimized antenna and radio system development and EMC/EMI studies for various applications, and the value-creation for customer dynamic requirements. He has published around 25 papers in the leading journal and conference proceedings. Dr. Jha was the recipient of the European Commission's Erasmus Mundus scholarship for the year 2007-2009 and has been awarded with other 15 national and international academic/research/travel grants, scholarships, honors and fellowships.
| | A singular integral that appears to have a form of the Fast Fourier Transform (FFT) structure, although it looks like on which FFT could be applied, is not readily the case. Basically, in various numerical conditions, a singularity pole can appear in the spectral domain interactions having the Green's function-based reaction integrals. Which then treated by a contour deformation in the complex plane integration brings the numerical structure that generally does not really allow the efficient use of the FFT. It can be noted that the application of FFT can transform spectral-domain reaction integrals to the space-domain parameters on-the-fly, no matter how large an antenna array is, as long as the dielectric and antenna types remain the same. But this needs a specialized method called "contour-FFT" that makes such powerful numerical tool (FFT) possible and accessible for such numerical conditions. The purpose of this talk on the Contour-FFT is centered around the followings:
- What are the spectral integrals that are used for the computation of the interactions in antenna/EM analysis?
- What is a singularity that appears in an integrand and how is it treated using the contour deformation to be able to integrate such function?
- Why is the direct use of FFT not readily possible on such function with high accuracy?
- How then Contour-FFT came to existence and how has it been used to efficiently and rapidly solve a long-time challenge in numerical computation with singularity?
|
Free Vibration and Stability Analyses of Structures Using the Spectral Collocation Method
Ma’en Sari, German Jordanian University, Jordan
|
|---|
Dr. Ma’en Sari earned B.S. (2002) and M. S. (2005) degrees in Mechanical engineering from Jordan University of science and Technology, and a Ph.D. (2011) in Mechanical Engineering from New Mexico State University. He is currently a professor in the Mechanical and Maintenance Engineering Department at the German Jordanian University in Jordan.
His research interests include free vibration analysis of continuous systems, stability and dynamic analyses of continuous systems, non-linear vibrations, hydrodynamics and thermal behavior of Micro electromechanical systems (MEMS), non-local behavior of different continuous structures, and structural health monitoring.
. |
| The spectral collocation method is used to investigate the free vibration and stability analyses of one- and two-dimensional structures. The spectral collocation method is a powerful numerical technique for solving differential equations by approximating functions globally rather than locally. Unlike finite difference or finite element methods, the spectral collocation method utilizes global basis functions, such as orthogonal polynomials or trigonometric functions, to represent the solution. The governing equations are imposed exactly at defined collocation points, which converts the differential problem into a system of algebraic equations. This method provides very high accuracy for smooth problems, often with far fewer collocation points compared to traditional low-order methods. Its efficiency and precision make it suitable for a wide range of applications, including fluid dynamics, structural mechanics, and wave propagation.
Among the various spectral collocation techniques, the Chebyshev collocation method has proven to be one of the most efficient approaches, especially for tackling partial differential equations defined on simple domains. By using Chebyshev polynomials and collocation at Chebyshev–Gauss–Lobatto nodes, this method provides excellent accuracy due to the clustering of nodes near boundaries. The Chebyshev spectral differentiation matrices enable efficient computation of derivatives with spectral (often exponential) convergence for smooth solutions. Due to its stability and ability to achieve high precision with relatively few collocation points, the Chebyshev collocation method has become a standard tool in modern numerical analysis of partial differential equations.
|
Intelligent Green Mobility Systems: AI, Autonomous Vehicles, MaaS, and V2X for Smart Cities
Manuel Cabral Reis, Universidade de Trás-os-Montes e Alto Douro, Portugal
|
|---|
Manuel J. Cabral S. Reis received his PhD in Electrical Engineering from the University of Aveiro, in 2001. He is currently Associate Professor with “Agregação” at the Department of Engineering, School of Science and Technology at UTAD. His main areas of interest include: study and development of devices and systems for smart friendly environments; signal and image processing and applications; development of multimedia methods and tools applicable to teaching/learning, in particular the use of educational resources on the Internet. He is an integrated researcher at the Institute of Electronics and Informatics Engineering of Aveiro (IEETA), University of Aveiro. He has published more than 190 papers in journals, conference proceedings, and book chapters, among others, and supervised 1 Post-Doc, 6 PhD and 33 Master students. He participated in more than 70 events around the world, and served as research member in more than 31 projects, including 21 projects where he was the lead researcher. He owns 12 patents and utility models registrations and has won 5 awards.
.
|
| The convergence of Artificial Intelligence (AI), electrification, and connectivity is reshaping the future of urban mobility. This tutorial provides an integrated perspective on how AI can enable sustainable, intelligent, and user-centric transportation systems for smart cities.
The talk begins with AI-driven approaches to optimize urban traffic management, electric vehicle (EV) charging infrastructure, and public transport scheduling, highlighting how predictive analytics and machine learning enhance efficiency and reduce environmental impact. It then explores the role of autonomous and electric vehicles, focusing on their integration within urban ecosystems and their interaction with intelligent infrastructure.
The session will also cover Mobility-as-a-Service (MaaS) as an AI-enabled framework that seamlessly integrates multiple transport modes, offering greener and more convenient mobility solutions. In addition, the tutorial will address energy-efficient Vehicle-to-Everything (V2X) communications, examining how AI can minimize communication overhead while improving safety, reliability, and energy efficiency.
Through this multi-level perspective — spanning vehicles, services, and networks — participants will gain a comprehensive understanding of how AI is driving the transition to sustainable and intelligent green mobility systems. The talk concludes with a forward-looking discussion on open challenges, cooperative intelligence, and policy implications for greener urban futures.
|
Teaching and Learning in the Era of AI
Pavel Loskot, ZJU-UIUC Institute, China
|
|---|
| Pavel Loskot joined the ZJU-UIUC Institute in January 2021 as an Associate Professor after being 14 years with the College of Engineering at Swansea University in the UK. He received his PhD degree in Wireless Communications from the University of Alberta in Canada, and the MSc and BSc degrees in Radioelectronics and Biomedical Electronics, respectively, from the Czech Technical University of Prague in the Czech Republic. In the past nearly 30 years, he had diverse teaching assignments, and was involved in numerous collaborative research and development projects concerning mostly wireless and optical communication systems, but also network science, computational molecular biology, renewable energy, and air transport management. He also held a number of paid consultancy contracts with industry. In 2012, he outlined the concept of interactive information discovery in a large corpora of documents akin to searching the web, which he shared to a key industry player in early 2014.
Pavel Loskot is the Senior Member of the IEEE, Member of the ACM, Fellow of the Higher Education Academy in the UK, and was in the first cohort to be awarded the Recognized Research Supervisor title by the UK Council for Graduate Education. Last year, he was elected the IARIA Fellow. In the past 20 years, he delivered over 120 talks including keynotes and plenary talks, and 18 tutorials at various international conferences. He currently serves as a Review Editor in Frontiers in Genetics, and an Editor in ICT Express. At present, he is teaching undergraduate courses in computer engineering including C and X86 assembly programming, operating systems, advanced mathematics, distributed systems, and introduction to theoretical computer science. Previously, he taught courses in digital communications, statistical inference, convex optimization, numerical methods and algorithms, and computer networks. His current research interests focus on mathematical and probabilistic modeling, and statistical signal processing and machine learning for multi-sensor and longitudinal data.
.
|  | Our classrooms have not changed for hundreds of years. After information and communication systems emerged three decades ago, knowledge started to become ubiquitous, and the technologies affordable. This had a substantial impact on how teaching and learning (T&L) have been provided in recent years. Novel T&L concepts such as flipped classroom and blended learning, and various digital systems to support and manage T&L appeared. Then, less than three years ago, large language models (LLMs) arrived, and immediately and quietly and without any warning started to disrupt our long-established T&L processes. The corresponding changes to our education are likely to be profound and permanent. Banning AI completely from the schools is a short-term solution just to gain more time to allow developing new education platforms where the AI is going to play a dominant role. At the same time, we can expect any developments from continued exponentially improving capabilities of AI systems to the AI hype fading away in a near future.
In this talk, I will touch upon a number of important topics how the current AI systems are affecting T&L. In particular, I will discuss how to strengthen the AI literacy for both teachers and students. I will explain how to develop AI-friendly mindset and digital citizenship. Other topics to be covered will concern how to use AI tools for various T&L tasks including enhancing the student engagement and classroom management, planning and preparing the lectures, and designing the AI-proof assessments. It is also important to consider how to define the AI classroom policies, how to reduce the risks of AI misuse, and how to use AI in classrooms responsibly. I may lightly touch on essential pedagogy, but it is more crucial to discuss what skills and knowledge to focus on in today’s rapidly changing world increasingly involving AI. The T&L should likely maintain the focus on skills that AI cannot replicate such as genuine creativity, critical thinking and deep social interactions. I will also mention the implications of AI-driven cognitive outsourcing versus AI assisted thinking. I will outline how to discover the right AI tools and relevant resources, particularly for teachers. Finally, I will provide a few examples how to use ChatGPT as a teacher and as a student. One of the aims of my talk is to examine whether using AI in T&L is worth it. I will also share my experiences from teaching undergraduate computer engineering students, and my experiences from guiding the final year research projects and supervising research students in the era of AI.
|
OptiSystem Software Enabling Machine Learning Capabilities for Free Space Optical Communication Systems
Ahmad Atieh, Optiwave Systems Inc., Canada
|
|---|
Ahmad
Atieh had received his Ph.D. degree in Electrical Engineering from University of Ottawa Canada in 1997, M.Sc. degree in Electrical Engineering from Jordan University of Science and Technology in 1987 and B.Sc. degree in Electrical Engineering from Yarmouk University Jordan in 1985. His current research interests are in the fields of optical fiber communication systems including optical fiber characterization, optical amplifiers, nonlinear fiber optics, and optical communication transmission systems. As well as in Free-space optical communication systems.
He has contributed more than 200 technical papers in different refereed journals and conferences. He holds over 32 issued patents and patents pending. He worked at the National Research Council of Canada, JDS Uniphase Inc, BTI Systems Inc in Canada, Taibah University in Madinah Saudi Arabia, Jordan University in Amman Jordan. He is currently VP at Optiwave Systems Inc. Canada and Adjunct Professor at University of Ottawa. |  | OptiSystem software offers a comprehensive simulation tool that can be used for photonic design automation and customized engineering design services. It offers a distinct competitive advantage through vastly shortening the development time to market while dramatically improving quality, productivity, and cost-effectiveness. OptiSystem has over 610 components and visualizers that can be used in designing free-space optic (FSO) systems using different models for data and transmission and quantum communication in air and underwater including unmanned aerial vehicles (UAV). The models used in the FSO components address light transmission in all weather and cloud conditions in addition they consider temporal and spatial scintillation effects. A new tool that has been added recently to the software enables machine learning based on eye diagram or performance metric parameters such as quality factor (QF) and optical signal to noise ratio (OSNR).
The tutorial will discuss the different aforementioned topics and provide guidelines for using the software to simulate the different systems. |
AI-Enabled Engineered Systems
Mohammad Abuzayyad, MathWorks, United States
|
|---|
Mohammad Abuzayyad holds an M.Sc. in Mechatronics Engineering and is pursuing his PhD at Asia Pacific University in the field of AI. He is currently the Technical Manager at MathWorks (makers of MATLAB), focusing on Electrification, AI, and Robotics.
|  | The MathWorks Academic Team will present the latest MATLAB tools for teaching and research. The session will cover the AI workflow in MATLAB and demonstrate how engineers and researchers implement complex solutions within a single environment. Highlights include:
-
Getting Started with MATLAB and Simulink – Discover powerful tools for modeling, simulation, and algorithm development.
-
Building AI-Enabled Systems – Learn how to integrate AI into engineered systems using an end-to-end workflow, with robotics as a key example.
-
Global Success Stories – Explore how leading institutions such as the
German Aerospace Research Center (DLR) are accelerating innovation with MATLAB and Simulink.
|
|
|---|
