Internet of Things: Artificial Intelligence and Next Generation Networks

Machine Learning

Supervised learning. Decision trees. Artificial neural networks and deep learning architectures. Support vector machines. Bayesian learning. Instance-based learning. Ensemble methods for classifiers (bagging, boosting, random forests). Semi-supervised and unsupervised learning methods. Reinforcement learning. Basic principles of experimental evaluation. Compression of machine learning models and their deployment on IoT devices.

Design, Development, and Performance Evaluation of Next-Generation Networks

Presentation of the most advanced network architectures, protocols, and services. Topics on active services with capabilities such as self-organization, environmental intelligence, and adaptation to underlying network infrastructures, spatial awareness, and extraction of multimodal interfaces for algorithms related to cost management, security, mobility, and quality of service assurance. Methods and tools for performance evaluation of modern heterogeneous systems. IoT architectures and services. Key characteristics of cloud technologies, types of offered services, development models, management technologies, and security issues. Integration of IoT into existing network infrastructures and ROI.

Pervasive Computing Systems

Introduction to Pervasive Computing (PC) and Pervasive Computing Systems (PCS), with emphasis on the intelligent interconnection of people, processes, data, and objects. Topics on the design and development of PCS. Architectures, models, and programming frameworks for context-aware systems. Location recognition and management of sequential data from sensors. Applications involving the collaboration of smart objects through PC. User interfaces and IoT application development by end users.

Modern Networks and IoT Interconnection

Management of network resources in wireless networks, with emphasis on low-power wide-area networks such as NB-IoT, LTE-M, EC-GSM-IoT, and LoRa. Quality differentiation of data flows between H2H (Human-to-Human) and M2M (Machine-to-Machine) communications. Principles of incoming network traffic management, Quality of Service (QoS), and flow scheduling. Emerging techniques for network resource management through virtualization technologies such as Software-Defined Networking (SDN), Network Function Virtualization (NFV), and Network Slicing. Improvement of network access through increased density of interconnection points and unified use of heterogeneous networking technologies. Applications in smart power grids and urban infrastructure.

Big Data Analytics

Algorithms for big data analysis and hyperlink analysis. Supervised and unsupervised learning for big data. Recommender systems: Collaborative & Hybrid Filtering, Graph-based approaches, Reinforcement Learning, LLMs for content recommendation. Dimensionality reduction in large-scale data (SVD, UV decomposition, Autoencoders, PCA). Large-scale data and behavior recognition. Affective Computing and Big Data. Technologies for managing large-scale data.

Topics in IoT Technology Applications

IoT Design and Architectures. Abstract Information Models and Middleware Technologies. Large-scale Distributed Experimental Infrastructures for IoT. Spectrum (TVWS), Energy, and Handover Management in IoT. Business Accelerators for IoT with emphasis on agriculture. Comparison of IoT Applications using intelligence and energy consumption metrics. Examples of commercial applications such as FarmBeats. Telecommunication Specifications and new broadband network technologies for supporting large-scale IoT networks. Neuromorphic and Distributed Computing Systems in IoT. Security and device authentication issues in large-scale IoT networks.

Embedded Systems and IoT

The aim of the course is to familiarize students with issues related to design, interfacing, and hardware/software interaction, the design of realistic microprocessor-based systems using modern tools, as well as the implementation of embedded digital systems to be used as nodes in a sensor network of the Internet of Things (IoT) type.

Specifically, the course content includes: introduction to embedded systems, arithmetic operations and peripheral device access, hardware/software interfaces, interrupts, buses, memory and its use in embedded systems, microprocessors, microcontrollers, FPGA and ASIC integrated circuits, programming embedded systems without an operating system (bare metal), programming embedded systems with an operating system for network applications, microprocessor and sensor interfacing, serial communication, use of messaging protocols in IoT networks.

Advanced IoT Communication Systems

The course focuses on modern wireless technologies and their applications in IoT, with an emphasis on experimental research. The topics covered include: analysis of communication requirements (energy consumption, range, reliability, density, delay, throughput, etc.). Radio channel analysis and characterization. IoT communication protocols and waveforms (with emphasis on LoRaWAN, 5G, and beyond 5G solutions). IoT radio network dimensioning. IoT and ultra-reliable low-latency communications.

Emerging technologies beyond 5G: beamforming/beam selection, massive MIMO, Reconfigurable Intelligent Surfaces, integration of artificial intelligence, and edge computing. The course modules include both analysis and experimentation through remote use of testbed platforms. Key experimentation tools are utilized, including programmable 5G network infrastructures, Software Defined Radios, IoT wireless communication nodes (LoRa, Bluetooth, Wi-Fi, and NB-IoT), simulation/development tools (Matlab, NS3, GNURadio), IoT development toolkits, edge infrastructure devices, and cloud platforms.

Robotics and Computer Vision

Sensors and Perception in Space: Foundations of Computer Vision. Basic Principles of Computer Vision. Computer Vision Systems. Stereo Vision. Representing Robot Position & Orientation. Kinematics. Navigation. Image Processing. Object Recognition. Sensor Fusion. Foundation Models.

Artificial Intelligence in IoT

Basic principles of IoT and AI. IoT and knowledge representation (XML, RDF, RDFS, OWL). Ontology engineering, semantic interoperability, and modeling in IoT. Bayesian Networks. Sensor data processing and analysis, and multimodal signal processing. Pretrained models, transformers, encoders, decoders, AI-generated reports. Deep Learning and edge AI. IoT data analysis and prediction using AI (time series forecasting, anomaly detection). Security and ethical challenges in the use of AI in IoT. Smart IoT devices for Health and other domains (e.g. Internet of Medical Things, Semantic Web of Things).

Master's Thesis Preparation

The MSc Thesis can be:

1. Research/Theoretical: This type focuses on developing a new theoretical model or extending an existing one, and applying it to problem-solving.
2. Research/Development: This type concentrates on the development of a 'new' system, based on existing theory. The development usually forms the dominant part of the work, enabling the demonstration of its application.
3. Application: This type focuses on developing a large application useful in a specific area of interest, utilizing one or more software/hardware packages.

The thesis text should be at least 20,000 words and indicatively include the following sections:

• Introduction to the Topic
• Literature Review
• Theoretical Development (including Theoretical Background and Theoretical Proposal)
• Methodology/Research
• Results, Conclusions, Future Research, Bibliography
• Other Deliverables, Applications (if applicable)

Machine Learning

Supervised learning. Decision trees. Artificial neural networks and deep learning architectures. Support vector machines. Bayesian learning. Instance-based learning. Ensemble methods for classifiers (bagging, boosting, random forests). Semi-supervised and unsupervised learning methods. Reinforcement learning. Basic principles of experimental evaluation. Compression of machine learning models and their deployment on IoT devices.

Design, Development, and Performance Evaluation of Next-Generation Networks

Presentation of the most advanced network architectures, protocols, and services. Topics on active services with capabilities such as self-organization, environmental intelligence, and adaptation to underlying network infrastructures, spatial awareness, and extraction of multimodal interfaces for algorithms related to cost management, security, mobility, and quality of service assurance. Methods and tools for performance evaluation of modern heterogeneous systems. IoT architectures and services. Key characteristics of cloud technologies, types of offered services, development models, management technologies, and security issues. Integration of IoT into existing network infrastructures and ROI.

Topics in IoT Technology Applications

IoT Design and Architectures. Abstract Information Models and Middleware Technologies. Large-scale Distributed Experimental Infrastructures for IoT. Spectrum (TVWS), Energy, and Handover Management in IoT. Business Accelerators for IoT with emphasis on agriculture. Comparison of IoT Applications using intelligence and energy consumption metrics. Examples of commercial applications such as FarmBeats. Telecommunication Specifications and new broadband network technologies for supporting large-scale IoT networks. Neuromorphic and Distributed Computing Systems in IoT. Security and device authentication issues in large-scale IoT networks.

Embedded Systems and IoT

The aim of the course is to familiarize students with issues related to design, interfacing, and hardware/software interaction, the design of realistic microprocessor-based systems using modern tools, as well as the implementation of embedded digital systems to be used as nodes in a sensor network of the Internet of Things (IoT) type.

Specifically, the course content includes: introduction to embedded systems, arithmetic operations and peripheral device access, hardware/software interfaces, interrupts, buses, memory and its use in embedded systems, microprocessors, microcontrollers, FPGA and ASIC integrated circuits, programming embedded systems without an operating system (bare metal), programming embedded systems with an operating system for network applications, microprocessor and sensor interfacing, serial communication, use of messaging protocols in IoT networks.

Advanced IoT Communication Systems

The course focuses on modern wireless technologies and their applications in IoT, with an emphasis on experimental research. The topics covered include: analysis of communication requirements (energy consumption, range, reliability, density, delay, throughput, etc.). Radio channel analysis and characterization. IoT communication protocols and waveforms (with emphasis on LoRaWAN, 5G, and beyond 5G solutions). IoT radio network dimensioning. IoT and ultra-reliable low-latency communications.

Emerging technologies beyond 5G: beamforming/beam selection, massive MIMO, Reconfigurable Intelligent Surfaces, integration of artificial intelligence, and edge computing. The course modules include both analysis and experimentation through remote use of testbed platforms. Key experimentation tools are utilized, including programmable 5G network infrastructures, Software Defined Radios, IoT wireless communication nodes (LoRa, Bluetooth, Wi-Fi, and NB-IoT), simulation/development tools (Matlab, NS3, GNURadio), IoT development toolkits, edge infrastructure devices, and cloud platforms.

Pervasive Computing Systems

Introduction to Pervasive Computing (PC) and Pervasive Computing Systems (PCS), with emphasis on the intelligent interconnection of people, processes, data, and objects. Topics on the design and development of PCS. Architectures, models, and programming frameworks for context-aware systems. Location recognition and management of sequential data from sensors. Applications involving the collaboration of smart objects through PC. User interfaces and IoT application development by end users.

Modern Networks and IoT Interconnection

Management of network resources in wireless networks, with emphasis on low-power wide-area networks such as NB-IoT, LTE-M, EC-GSM-IoT, and LoRa. Quality differentiation of data flows between H2H (Human-to-Human) and M2M (Machine-to-Machine) communications. Principles of incoming network traffic management, Quality of Service (QoS), and flow scheduling. Emerging techniques for network resource management through virtualization technologies such as Software-Defined Networking (SDN), Network Function Virtualization (NFV), and Network Slicing. Improvement of network access through increased density of interconnection points and unified use of heterogeneous networking technologies. Applications in smart power grids and urban infrastructure.

Big Data Analytics

Algorithms for big data analysis and hyperlink analysis. Supervised and unsupervised learning for big data. Recommender systems: Collaborative & Hybrid Filtering, Graph-based approaches, Reinforcement Learning, LLMs for content recommendation. Dimensionality reduction in large-scale data (SVD, UV decomposition, Autoencoders, PCA). Large-scale data and behavior recognition. Affective Computing and Big Data. Technologies for managing large-scale data.

Robotics and Computer Vision

Sensors and Perception in Space: Foundations of Computer Vision. Basic Principles of Computer Vision. Computer Vision Systems. Stereo Vision. Representing Robot Position & Orientation. Kinematics. Navigation. Image Processing. Object Recognition. Sensor Fusion. Foundation Models.

Artificial Intelligence in IoT

Basic principles of IoT and AI. IoT and knowledge representation (XML, RDF, RDFS, OWL). Ontology engineering, semantic interoperability, and modeling in IoT. Bayesian Networks. Sensor data processing and analysis, and multimodal signal processing. Pretrained models, transformers, encoders, decoders, AI-generated reports. Deep Learning and edge AI. IoT data analysis and prediction using AI (time series forecasting, anomaly detection). Security and ethical challenges in the use of AI in IoT. Smart IoT devices for Health and other domains (e.g. Internet of Medical Things, Semantic Web of Things).

Master's Thesis Preparation

The MSc Thesis can be:

1. Research/Theoretical: This type focuses on developing a new theoretical model or extending an existing one, and applying it to problem-solving.
2. Research/Development: This type concentrates on the development of a 'new' system, based on existing theory. The development usually forms the dominant part of the work, enabling the demonstration of its application.
3. Application: This type focuses on developing a large application useful in a specific area of interest, utilizing one or more software/hardware packages.

The thesis text should be at least 20,000 words and indicatively include the following sections:

• Introduction to the Topic
• Literature Review
• Theoretical Development (including Theoretical Background and Theoretical Proposal)
• Methodology/Research
• Results, Conclusions, Future Research, Bibliography
• Other Deliverables, Applications (if applicable)

Master's Thesis Preparation

The MSc Thesis can be:

1. Research/Theoretical: This type focuses on developing a new theoretical model or extending an existing one, and applying it to problem-solving.
2. Research/Development: This type concentrates on the development of a 'new' system, based on existing theory. The development usually forms the dominant part of the work, enabling the demonstration of its application.
3. Application: This type focuses on developing a large application useful in a specific area of interest, utilizing one or more software/hardware packages.

The thesis text should be at least 20,000 words and indicatively include the following sections:

• Introduction to the Topic
• Literature Review
• Theoretical Development (including Theoretical Background and Theoretical Proposal)
• Methodology/Research
• Results, Conclusions, Future Research, Bibliography
• Other Deliverables, Applications (if applicable)