Department of Electronic Engineering

Hellenic Mediterranean University

Diploma Thesis

The Diploma Thesis at the Department of Electronic Engineering
According to the regulations applicable to the Department of Electronic Engineering, the preparation of a diploma thesis by students following the 5-year Undergraduate Program of Study is mandatory and is credited with 30 ECTS credits. The preparation of the thesis aims:

  • to help students delve deeper into topics of their interest, acquiring knowledge and skills that will support their future development and career;
  • to familiarize students with scientific thinking and methodology, the review of existing knowledge, and the research process;
  • to give students the opportunity to broaden their way of thinking and pursue innovation;
  • to give students the opportunity to engage with cutting-edge research topics, always according to their interests, which is strongly encouraged in the Department of Electronic Engineering.

Upon completing the diploma thesis, students will have developed and improved their skills in:

  • the structured search of scientific sources and the literature review;
  • the clear formulation of a scientific problem and the organization and design of the methodology to address it;
  • the collection of data, its processing and presentation, and the drawing of conclusions to interpret phenomena;
  • solving problems related to the field of Electronic Engineering;
  • writing scientific texts and presenting them to a wide audience.

 

Indicative Diploma Thesis Topics

~.~.~.~.~.~.~.~

1. Antonakakis Marios, Assistant Professor, Division of Informatics & Automation (mantonakakis@hmu.gr)

  • Section 1: Computational Neuroscience & Geometric AI
    • 1. Geometric DL for Uncertainty-Aware Head Modeling: Use of Bayesian GNNs for direct mesh deformation on MRI, avoiding slow voxel-based segmentation. Quantification of geometric uncertainty and study of its effect on source localization error.
    • 2. Deep-FEM & LaBraM Integration (Cam-CAN): Development of a 5-compartment segmentation pipeline with CNNs and solution of the Forward Problem via FEM. Integration into the LaBraM Foundation Model to extract physics-informed brain representations.
    • 3. Neural Granger Causality (NGC) in Source Networks: Application of Neural-GC (LSTMs/SRUs) to source time series to reveal non-linear directed interactions. Highlighting dynamic connectivity in large-scale networks, overcoming classical limitations.

    Section 2: Agentic AI & Computer Vision

    • 1. Agentic Visual Perception (YOLO & SAM): Design of an AI Agent (Microsoft AutoGen) that combines YOLO and SAM for interactive segmentation via natural language. Real-time Visual Reasoning for autonomous object handling.
    • 2. Brain-to-Image Reconstruction (Visual Transformers): Use of Mind-Vis to reconstruct visual stimuli from fMRI data during scanning. Decoding of visual cortex activity via ViTs and Masked Autoencoders.

    Section 3: Advanced Research Topics (LEMON Dataset)

    • 1. Topological Data Analysis (TDA) on Functional Connectivity: Use of Persistent Homology to detect Higher-Order Interactions beyond pairwise relationships. Identification of topological signatures that characterize subjects’ cognitive profiles.
    • 2. Self-Supervised Learning for Connectome Embeddings: Training of Transformer models (LaBraM-type) to extract embeddings representing functional connectivity. Automatic clustering of subjects based on their “Functional Fingerprint”.

~.~.~.~.~.~.~.~

2. Antonidakis Emmanouil, Professor, Division of Informatics & Automation (antonidakis@hmu.gr)

~.~.~.~.~.~.~.~

3. Vardiambasis Ioannis, Professor, Division of Telecommunications & Networks (ivardia@hmu.gr)

~.~.~.~.~.~.~.~

4. Giannakakis Georgios, Associate Professor, Division of Informatics & Automation (ggian@hmu.gr)

  • Detection of epileptic activity using electroencephalography and cardiography (Description: For the management of epilepsy, brain activity is widely recorded in clinical practice through the electroencephalogram (EEG). In parallel, biosignals such as the cardiogram (ECG), respiration pattern, and myogram (EMG) also provide useful information before or during epileptic seizures. In long-duration recordings, the dynamic state at each moment is revealed through characteristic indicators (features). These indicators are linear/non-linear, derived from temporal, spectral, and morphological characteristics. Division: Informatics, Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital Signal Processing, MATLAB or Python).
  • Recognition of the emotional state of stress through facial video processing (Description: The human face is in many cases an indicator of a person’s emotional state. To evaluate facial expressions, techniques based on facial muscle estimation, techniques based on 2D or 3D models, or feature-motion study techniques are used. The features are fed into an expression representation system (FACS, FAPs) and evaluated by the final system. The aim of this thesis is the extraction of facial features for the assessment of the emotional state of stress. Division: Informatics. Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Image/video analysis, Computer Vision, MATLAB or C++, Python).
  • Recognition of the emotional state of stress through biosignals (Description: It is known that stress affects the human body at many levels and the body’s physiological response to it is perceived through biological signals. The main biosignals that assess emotional state are the electroencephalogram (EEG), electrocardiogram (ECG), electromyogram (EMG), respiration pattern (RSP), and skin conductance (EDA). From these recordings, features will be extracted from each recording that describe as representatively as possible the quantity they measure. Subsequently, the selection of the most robust features will be carried out, i.e., features whose variation is characteristic of stress. The aim of this thesis is the development of a computational system for recognizing the emotional state of stress through biological signals. Division: Informatics, Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Recognition of the emotional state of stress through voice (Description: This thesis aims at the automatic detection of the psycho-emotional state through the acoustic characteristics of voice, using speech processing and machine learning techniques. The methodology includes collection/annotation of recordings, preprocessing (denoising, VAD), extraction of features such as prosodic (F0, intensity, speech rate, pauses), spectral/cepstral (MFCC), voice quality (jitter, shimmer, HNR), and formants. Stress/non-stress classification or regression on continuous dimensions (arousal/valence) is implemented with machine learning techniques or deep networks (CNN-BiLSTM/Transformers). Division: Informatics, Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Development of a web-based session system with automatic facial expression recognition and storage of facial video and biosignals (Description: The proposed system is a web-based session platform that integrates facial expression recognition and biosignal recording technologies. The platform will allow live sessions between users in a secure environment through a web browser, while simultaneously analyzing and storing data from the web camera and biosignal sensors. The data will be stored in a structured way for each participant, and visual presentation of the data will be enabled. Division: Informatics, Area: Web programming. Difficulty Level: A. Required Knowledge: Web technologies (web development), Databases).
  • Time-frequency representation of EEG signals (Description: The electrical activity and changes in the electrical potential of the human brain are depicted through the electroencephalogram (EEG), which is widely used in clinical practice. The EEG signal, particularly in the case of pathologies or during cognitive processes (evoked potentials), is characterized by non-stationarity. In these cases, traditional spectral analysis with the Fourier transform is not recommended. Time-frequency analysis can better reveal the dynamic behavior of these signals. In this thesis, various time-frequency transform algorithms (STFT, matching pursuit, Wigner-Ville) will be compared with respect to representation accuracy. The methods will be applied to real EEG signals from individuals with dyslexia, epilepsy, or schizophrenia. Division: Informatics, Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Analysis of EEG signals using entropy measures (Description: Entropy is a characteristic quantity that originates from thermodynamics and describes the degree of organization/disorder of a physical system. It is a useful tool for understanding brain dynamics in various cognitive processes or pathological conditions. Its advantage is that it does not require signal stationarity, so it can be applied beyond the resting-state EEG to evoked potentials where very rapid changes in signal variance are observed. In this thesis, quantitative entropy measures will be computed in a MATLAB environment, followed by application to signals of dyslexic and/or epileptic individuals. Division: Informatics, Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Study of directed brain connectivity through epileptic seizure propagation models (Description: Directed brain connectivity is a modern field in neuroscience and neurophysiology that examines the causal relationships and information flow between different brain regions. Unlike simple functional connectivity, directed connectivity seeks to reveal the causal relationship and directed coherence between brain structures, using methods such as the Directed Transfer Function (DTF) and Partial Directed Coherence (PDC), as well as more recent approaches with dynamic estimation of these. The study of these relationships is crucial for understanding information processing mechanisms in the brain, as well as for detecting dysfunctions associated with pathological conditions such as epilepsy or anxiety disorders. The aim of this thesis is the analysis of EEG signals using directed brain connectivity methods, on real data from epileptic patients, to detect patterns of epileptic seizure propagation. Division: Informatics, Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Emotional state detection through body posture with video analysis (Video-based posture detection and classification) (Description: The ability to detect human posture is very important for applications related to the analysis of human behavior. The development of techniques for detecting and classifying body posture is therefore particularly important, with applications in psychology, elderly care, etc. The basic idea is that emotional states (e.g., anxiety, sadness, anger, joy) are often manifested in characteristic ways in the body and posture (e.g., hunched shoulders, tension in the hands, nervous movements, open vs. closed posture). The aim of this thesis is the application and study (comparison) of existing vision-based approaches using depth cameras (3D depth cameras) to monitor human body posture in order to detect emotional state. Division: Informatics, Area: Digital signal processing, Machine Learning and AI, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Psychoacoustic analysis of emotions through musical stimuli and EEG (Description: This thesis examines the detection and categorization of psychoacoustic emotions through the interaction of musical stimuli and brain activity, as recorded via electroencephalography (EEG). Specifically, it investigates the effect of individual musical characteristics, such as rhythm, melody, harmony, and intensity, on eliciting emotional reactions and their correspondence to brain function patterns. For this purpose, advanced biosignal processing techniques, feature extraction, and machine learning algorithms are used, with the aim of developing a computational model for the detection and classification of emotional states. The study aspires to contribute to deepening scientific knowledge regarding the neurophysiological basis of music perception and emotional response, with applications in neuroscience, psychology, and music therapy. Division: Informatics, Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Early diagnosis of neurodevelopmental disorders (ADHD, learning difficulties, etc.) through computational gaze-direction estimation (Description: This thesis focuses on the detection and classification of eye-tracking patterns to support the early diagnosis of neurodevelopmental disorders (ADHD, learning difficulties, etc.). Data from eye-trackers will be used, and then temporal sequences of characteristic gaze movements (fixations, saccades, regressions) will be extracted. Subsequently, machine learning and deep learning algorithms (e.g., SVM, Random Forests, LSTM neural networks) are applied to recognize patterns associated with difficulties in reading, text comprehension, and attention retention in different student populations. The final goal is the development of an automated eye-tracking data analysis framework that can support specialists in the early detection of learning difficulties and contribute to personalized interventions. Division: Informatics, Area: Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Treatment of phobias through the development of a Virtual Reality serious game and biosignals (Description: This thesis focuses on the design and implementation of a Virtual Reality serious game for the treatment of psychological phobias and post-traumatic stress (PTSD). The system combines various VR scenarios with the simultaneous recording and analysis of biosignals (e.g., HRV, EDA), so that the user’s psycho-emotional state is monitored. In this way, the game adapts dynamically to the user’s stress level, offering a personalized and controlled exposure process. The aim is the creation of an innovative tool that can be used in psychotherapeutic settings, enhancing the effectiveness of traditional phobia treatment methods. Division: Informatics, Area: VR development, Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).
  • Natural Language Processing (NLP) and semantic text analysis for psychiatric applications (Description: This thesis aims at the development of Natural Language Processing (NLP) methods and semantic text analysis to support the monitoring of psychiatric disorders (depression, psychoses, post-traumatic stress). Semantic text analysis focuses on identifying linguistic patterns and extracting semantic features from clinical texts, medical records, and psychotherapy sessions. Based on these, Named Entity Recognition models will be built from specialized clinical lexicons (UMLS, SNOMED, etc.). Subsequently, structures for representing semantic content will be developed — Contextual embeddings (e.g., MentalBERT) and knowledge creation through semantic networks (Knowledge Graphs) (e.g., RDF/OWL). In this way, symptoms (e.g., depressive mood, sleep disorders, etc.), relevant stressors (e.g., significant life events, interpersonal conflicts), interventions (e.g., medication, psychotherapy), and outcomes (e.g., symptom improvement, response to treatment) will be analyzed. Finally, sentiment analysis (e.g., sadness, stress, etc.) will be performed through machine/deep learning techniques and AI to recognize psychopathological patterns. Division: Informatics, Area: Semantic text analysis, Psychiatry. Difficulty Level: A or B).
  • Development of a computational algorithm for the analysis of cardiac electrograms for the surgical treatment of atrial fibrillation (Description: Atrial fibrillation (AF) is the most common persistent arrhythmia worldwide. The maintenance of AF has been attributed either to self-sustaining electrical waves or to localized dominant regions that harbor “drivers” of electrical activity. Time- and frequency-domain analysis of local fragmented atrial electrograms has been used to reveal regions triggering or maintaining AF. In this thesis, time/frequency-domain analysis methods of atrial electrograms from the pulmonary veins and the left atrial appendage will be applied during AF cryoablation. Division: Informatics, Area: VR development, Digital signal processing, Biomedical Technology. Difficulty Level: A or B. Required Knowledge: Digital signal processing, MATLAB or C++ or Python).

~.~.~.~.~.~.~.~

5. Kakavelakis Georgios, Assistant Professor, Division of Electronics & Applications (kakavelakis@hmu.gr)

~.~.~.~.~.~.~.~

6. Kalderis Dimitrios, Professor, Division of Informatics & Automation (kalderis@hmu.gr)

~.~.~.~.~.~.~.~

7. Kapetanakis Eleftherios, Associate Professor, Division of Electronics & Applications (ekapetan@hmu.gr)

  • Electrical Characterization of MOS Transistors.
  • Electrical Characterization of MOS and MIS Capacitors.
  • An Overview of Inorganic and Organic Transistor Technology for the Detection and Determination of Biomarkers.

~.~.~.~.~.~.~.~

8. Katsivela Eleftheria, Associate Professor, Division of Electronics & Applications (katsivela@hmu.gr)

~.~.~.~.~.~.~.~

9. Kokkinos Evangelos, Associate Professor, Division of Telecommunications & Networks (ekokkinos@hmu.gr)

~.~.~.~.~.~.~.~

10. Kouli Maria, Associate Professor, Division of Informatics & Automation (mkouli@hmu.gr)

  • Mapping of the Urban Heat Island phenomenon with Remote Sensing and Geographic Information Systems.
  • Multi-criteria analysis and use of GIS and remote-sensing methods for the creation of a flood hazard map.
  • Multi-criteria analysis and use of GIS and remote-sensing methods for the creation of a landslide hazard map.
  • Site selection for a wind farm installation.
  • Contribution of Sentinel satellite data to the assessment of areas for the installation of offshore wind farms.
  • Investigating the ability of Sentinel, ASTER, and Landsat satellite data to detect mineralization zones.
  • Monitoring of urbanization with night-light remote sensing.

~.~.~.~.~.~.~.~

11. Konstantaras Antonios, Professor, Division of Informatics & Automation (akonstantaras@hmu.gr)

  • 3D Design in Studio, Implementation and Programming in Scratch and Python of SPIKE educational robots with flexible-learning scenarios for K6-K12 students (Description: This thesis concerns the design in the Studio 3D modeling software and the construction of SPIKE educational robots, accompanied by the development of flexible educational scenarios for K6-K12 students to solve, with code development in Scratch and Python. A total of 30 complete exercises will be developed (5 per month)).
  • 3D Design in Studio, Implementation and Programming in Scratch and Python of NEZHA educational robots with flexible-learning scenarios for K6-K12 students (Description: This thesis concerns the design in the Studio 3D modeling software and the construction of NEZHA educational robots, accompanied by the development of flexible educational scenarios for K6-K12 students to solve, with code development in Scratch and Python. A total of 30 complete exercises will be developed (5 per month)).
  • 3D Design in LDD, Implementation and Programming in Scratch and Python of WeDo2 educational robots with flexible-learning scenarios for K6-K12 students (Description: This thesis concerns the design in the LDD 3D modeling software and the construction of WeDo2 educational robots, accompanied by the development of flexible educational scenarios for K6-K12 students to solve, with code development in Scratch and Python. A total of 30 complete exercises will be developed (5 per month)).

~.~.~.~.~.~.~.~

12. Kotti Spyridoula-Melina, Associate Professor, Division of Electronics & Applications (kotti@hmu.gr)

~.~.~.~.~.~.~.~

13. Liodakis Georgios, Assistant Professor, Division of Telecommunications & Networks (gsl@hmu.gr)

  • Study of electromagnetic scattering topics in Reconfigurable Intelligent Surfaces (RIS)-assisted wireless networks.
  • Performance of optical communication systems with coherent transmission.
  • Techno-economic analysis for 6G networks.
  • Study of telecommunications legislation and the regulatory framework of digital services.
  • Achieving Quality of Experience (QoE) in cellular communication networks.
  • Application of Canvas Business Modeling for IoT applications.
  • Application of machine learning techniques in cellular communication systems.

~.~.~.~.~.~.~.~

14. Makris Ioannis, Professor, Division of Electronics & Applications (jpmakris@hmu.gr)

~.~.~.~.~.~.~.~

15. Maravelakis Emmanouil, Professor, Division of Informatics & Automation (marvel@hmu.gr)

  • Design and construction of a multi-sensor electronic board for small-scale underwater vehicles.
  • Development of software for underwater-environment monitoring.
  • Feasibility study of a GPS-localization architecture for monitoring underwater gliders.
  • Scan to BIM – Application of a portable terrestrial scanning system for Building Information Modeling.
  • Scan to CAD – Study of reverse-engineering techniques for the creation of mechanical CAD models.
  • Comparison and evaluation of state-of-the-art technologies for 3D building capture.
  • 3D reconstruction of ancient monuments using 3D data from drones.
  • Design and Implementation of a Smart Enclosure for Laser Cutter/Engraver Systems with Air-Pollutant Measurement and Filter Evaluation via ESP32.

~.~.~.~.~.~.~.~

16. Baklezos Anargyros, Assistant Professor, Division of Telecommunications & Networks (abaklezos@hmu.gr)

  • Development of a magnetic-signature compensation device for a point source.
  • Development of a magnetic-signature compensation device for a distributed source.
  • Development of passive RADAR for DRONE/UAV detection.
  • Development of countermeasure algorithms for combating false-positive targets.
  • Comparison of different neural-network architectures for optimal medical-image segmentation.

~.~.~.~.~.~.~.~

17. Barbounakis Ioannis, Assistant Professor, Division of Telecommunications & Networks (i.barbounakis@hmu.gr)

~.~.~.~.~.~.~.~

18. Bikakis Nikolaos, Assistant Professor, Division of Informatics & Automation (bikakis@hmu.gr)

  • Explainability methods and algorithms.
  • Fairness in Recommender systems.
  • Design and Development of Interactive Data Visualization Tools.
  • Trajectory analysis methods and algorithms for moving objects.
  • Data structures and algorithms for visual data exploration.

~.~.~.~.~.~.~.~

19. Nikolopoulos Christos, Assistant Professor, Division of Telecommunications & Networks (cnikolo@hmu.gr)

  • Characterization of ferromagnetic nanoparticles using machine learning techniques for solving inverse electromagnetic problems.
  • Study of optimal 5G-network target tracking through Reconfigurable Intelligent Surfaces (RIS) using computer vision and machine learning techniques.
  • Study and implementation of 3D electromagnetic-field visualization through AR/VR glasses for electromagnetic-compatibility purposes.
  • Development of vehicle-motion prediction algorithms in multi-parameter environments.
  • Modeling and simulation of wearable, sewn, or knitted antennas.
  • Study of metal-object detection through microwave technology.
  • Autonomous robotic system for antenna radiation-pattern measurements.

~.~.~.~.~.~.~.~

20. Petrakis Nikolaos, Assistant Professor, Division of Informatics & Automation (nik.s.petrakis@hmu.gr)

~.~.~.~.~.~.~.~

21. Petridis Konstantinos, Professor, Division of Electronics & Applications (cpetridis@hmu.gr)

~.~.~.~.~.~.~.~

22. Saltas Vasileios, Associate Professor, Division of Electronics & Applications (saltas@hmu.gr)

  • Signal analysis and pattern recognition in acoustic-emission recordings during the unconfined uniaxial compression of rock specimens.
  • Effect of temperature on the electrical-dielectric characterization of composite plastic materials.
  • Broadband dielectric spectroscopy on conductive composite plastic materials for electromagnetic-shielding applications.

~.~.~.~.~.~.~.~

23. Spanoudakis Nikolaos, Assistant Professor, Division of Informatics & Automation (nspan@hmu.gr)

  • ASEME code generator for ROS – the turtlebot 3 case.
  • ASEME code generator for the game Starcraft. Use of argumentation for decision-making.
  • A spider application for Skroutz for real-time product pricing using argumentation.
  • Formal models for ethical agents and robots. An argumentation-based approach.
  • Upgrade of course outlines in the student information system.
  • Data analysis on course outlines.
  • ASEME code generator for an IoT platform – the “sinaisthisi” case.
  • A recruiter application using argumentation and LLMs.
  • Use of the Eclipse Modeling Framework for the analysis of ASEME transition expressions.
  • Web-based ASEME modeling.

~.~.~.~.~.~.~.~

24. Stavroulakis Georgios, Professor, Division of Electronics & Applications (gstav@hmu.gr)

  • Environmental protection and sustainable development.
  • Environment and new technologies.
  • Renewable Energy Sources and the environment.
  • Circular economy and the environment.
  • Those interested in preparing a diploma thesis can be informed in detail by enrolling at https://eclass.hmu.gr/modules/course_info/index.php?course=EE257.

~.~.~.~.~.~.~.~

25. Tatarakis Michail, Professor, Division of Electronics & Applications (mictat@hmu.gr)

~.~.~.~.~.~.~.~

26. Fouskitakis Georgios, Associate Professor, Division of Informatics & Automation (fouskit@hmu.gr)

~.~.~.~.~.~.~.~

27. Fragkiadakis Nikolaos, Lecturer, Division of Informatics & Automation (nfrag@hmu.gr)

~.~.~.~.~.~.~.~

28. Fytilis Ioannis, Associate Professor, Division of Electronics & Applications (fitilis@hmu.gr)

~.~.~.~.~.~.~.~

29. Chatzakis Ioannis, Professor, Division of Electronics & Applications (jchatzakis@hmu.gr)

  • Class-E power transfer.
  • PWM amplifier of MMC topology.
  • Inductive charging system.

 

Assignment and preparation procedures
The preparation of a diploma thesis requires that the student has successfully completed 36 of the 48 courses of the first 8 semesters. In the 10th semester, typically, the student selects and prepares the diploma thesis, with a standard preparation period of six months.
In the Department of Electronic Engineering, and for the convenience of students, it is possible and recommended to select a thesis topic as early as the 9th semester of studies. All diploma theses in the Department are, to a lesser or greater extent, research-oriented, and are prepared under the supervision and research guidance of a faculty member of the Department.
After each examination period, the teaching staff (all Faculty and EDIP members, as well as any Adjunct Lecturers who wish to) suggest diploma thesis topics. The Department Assembly ensures that a sufficient number of topics are announced, so that students have the option to choose. Interested students, after consultation with the topic’s proposer, request the assignment of a thesis on a specific topic through an application to the Secretariat, countersigned by the proposer. Provided there is a joint decision between the Departments, thesis supervision may be assigned to faculty members or appointed lecturers of another Department of the same or another School, or to faculty members of other Universities.
Typically, the thesis topic is assigned to a single student. In the event that two or more students express interest in the same thesis topic, it is at the discretion of the supervising faculty member who proposed the topic to select the student to whom it will be assigned. However, in special cases where there is a clear distinction between the individual tasks, and following the supervisor’s recommendation, the same thesis may be assigned to two or more students.
Templates of the documents that must be completed to begin the diploma thesis are provided below. Upon their submission, the final decision and approval is given by the Department Assembly. The official start date of the thesis is defined as the date on which the Department Assembly approved the submitted assignment proposal. A student who has undertaken a diploma thesis under the supervision of a specific faculty member may, with a justified request, apply (once) to change the topic. Also, again following clear justification and a written request, the subject/title of the thesis may be modified, provided the original description is not substantially altered. In case substantial changes become necessary during the preparation of the thesis, a written request must be submitted in a timely manner by the supervisor in consultation with the student.
To facilitate the writing of the thesis, the Department provides a template as an electronic word-processing file, posted on the Department’s website after approval by the Assembly. In addition, explicit instructions for writing the thesis, as well as a guide for citing bibliographic references, are posted on the Department’s website.

Supervision – Obligations of supervisors and students
All members of the teaching staff who independently teach courses in the 5-year Undergraduate Program of Study may undertake the supervision of a diploma thesis on topics related to their subject and research interests. The supervisor undertakes to guide the student throughout the preparation of the thesis on the basis of a clear schedule, and to assist at all stages: (i) the initial planning and approach to the topic, (ii) the literature review, (iii) solving the specific problem, (iv) processing the data, (v) drawing the relevant conclusions, and (vi) writing and presenting the thesis.
Students must cooperate with their supervisors at regular intervals, keep them informed of any problems and difficulties that arise, adhere to the timetables set, and not exert undue pressure.
In general, the role of the supervisor is advisory, and students themselves should take the necessary initiatives, since, regardless of the degree of the supervisor’s contribution, responsibility for the quality of the thesis lies primarily with the student who prepares it, and secondarily with the supervisor who proposes and oversees it.

 

Completion procedures – Evaluation
After completing the thesis, and with the supervisor’s approval, the student submits it to the Secretariat in electronic form. The Assembly appoints a three-member Examination Committee from members of the Department’s permanent and temporary teaching staff for its evaluation. The evaluation process includes the presentation of the thesis by the student to a wide audience. Thesis presentations may be organized in groups on fixed dates at least 4 times a year, and may be attended by all members of the Department’s academic community (teaching staff and students), as well as visitors.
Depending on the topic and the particularities of each thesis, the main evaluation criteria are indicatively:

  • familiarity with existing knowledge through appropriate literature research;
  • the correct presentation of bibliographic sources;
  • the formulation and solution or simulation of the problem;
  • the acquisition of data as a result of theoretical calculation, experimental process, or mining/collection;
  • the computational processing of the collected data;
  • the tests, applications, and evaluation of results;
  • its structure and written presentation (coherence of the text, correct use of terminology and language, precise formulation of concepts, compliance with writing standards);
  • the scientifically sound documentation of the conclusions;
  • the diligence and initiative shown by the student during the preparation; and
  • the oral presentation of the thesis.

The weighting of the above indicative criteria varies according to the nature of the topic and the assessments of the Examination Committee, whose members judge and grade independently at their discretion. The grade of the thesis results from the average score of the Committee members. The minimum passing grade is 5 (five). The Committee members complete and submit to the Secretariat a relevant evaluation record, which includes the student’s name, the title of the thesis, and the final grade. If a thesis is deemed incomplete, it is referred back for additional work and returned for examination.
All completed theses are mandatorily deposited in the University’s institutional repository in accordance with the institutional repository policy, which is approved by the Senate. The submission of theses by students is mandatory and is monitored by the Department Secretariat.
Templates of the aforementioned documents are provided below:

 

Ethics – Protection of rights – Plagiarism
The copyright of the thesis, according to applicable provisions, belongs to those who contributed to its preparation (Institution, Department, supervisor, student). In the event that the scientific results of the work are protected by a patent, current legislation applies regarding the holders of the rights arising from it. What is stated in the text of the thesis does not necessarily mean that it is identified with or accepted by the Department. In any case, reference should be made to the authors of the work, the title of the work, and the Department. All parties involved in the preparation of a thesis are committed to complying with the Rules of Academic Conduct, as established by the Institution’s Ethics Committee.
Plagiarism, in the sense of appropriating the intellectual property of others, often in the form of presenting, intentionally or unintentionally, another person’s work as one’s own, is condemned by the Department of Electronic Engineering, and constitutes not only a moral offense but also a violation of copyright law, which may result in legal penalties. Sources should always be clearly cited, whether text or unpublished ideas and opinions.
Other actions that violate the codes of academic ethics and integrity, and are likewise reprehensible, include delegating part or all of a thesis to someone else, wholesale copying of the work of others, and collaboration among students on a thesis without the supervisor’s knowledge. In connection with the above, each student should make clear in the text of the thesis what is the product of their own work and what is the product of another person’s work, making systematic reference to citations and placing them where appropriate.

Skip to content