Electrical Engineering, Computer Engineering and Informatics

PhD positions, starting September 2018

The last date to apply for postgraduate studies is Friday, 13th of April 2018.

For applications, click  here

  • One (1) post in the following topic: “Neural attention mechanisms for learning to detect subtlefinancial crime patterns in big data”

Description: Financial crime, including tax evasion and financial fraud, is a major problem that plagues European economies. Typically, local European authorities employ various heuristic criteria to detect it, which require extensive human labor.

This thesis envisages the examination and development of novel machine learning techniques to facilitate automatic detection and characterization of patterns of such illicit activity. Specifically, we will develop statistical machine learning models that learn to extract salient temporal dynamics in vast amounts of data. We will focus on neural attention mechanisms that entail differentiable memory modules; these offer an unprecedented capacity to learn over long temporal horizons, cope with abrupt as well as slower pattern shifts, and learn to distinguish between salient and less important dynamics. Their success in hard problems, such as speech recognition and natural language understanding, provide strong promise   for their success in this context, whereby it is needed that they correlate across millions of individuals or corporates over long time horizons, in a way that humans cannot easily do.

Research Advisors:  Sotiris Chatzis, Assistant Professor, sotirios.chatzis@cut.ac.cy

 

  • One (1) post in the following topic: “Novel deep learning models for neural machine summarization”

Description: This thesis will develop novel neural statistical models for abstractive document summarization. Our research will focus on: 

(i) Policy learning algorithms for model training, which may enable us to ameliorate the “exposure bias” problem of the current state-of-the-art paradigm;

(ii) Novel statistical formulations of attention mechanisms and differentiable memory modules that may give rise to more potent mechanisms for capturing and modelling sequential dynamics.

Required Qualifications: We expect that the candidates will have some rudimentary affinity with linear algebra concepts, and the Python programming language. We will definitely consider a plus some experience with fundamental machine learning techniques, especially maximum-likelihood training and regularization techniques.  

Research Advisors:  Sotiris Chatzis, Assistant Professor, sotirios.chatzis@cut.ac.cy

 

  • One (1) post in the following topic: “Complex Networks: Geometry, Dynamics and Prediction”

Description: Real-world complex networked systems (technological, biological, social, financial, etc.) can be mapped to geometric spaces that lie hidden beneath their observable topologies. These geometric spaces are called “hidden”, as they play the role of an underlying coordinate system, not readily observable by examining the network topology. Nodes closer in the underlying space are connected in the observable network topology with higher probability.

The PhD candidate will focus in studying the properties of these underlying geometric spaces and the spatial dynamics of network nodes in these spaces. It is anticipated that important fundamental and practical questions will be addressed through this PhD thesis such as: (i) what are the laws governing the “motion” of network nodes in these spaces? (ii) Can this motion be modeled using classical mechanics laws, e.g., Newton’s laws of motion or stochastic versions of it? (iii) Is this motion chaotic or can be predicted? (iv) Given that we can predict this motion, can we predict the future structure and evolution of real complex networks?

This position falls under the general scientific areas of Network Science, Data Science, and Predictive Analytics. The ideal candidate should like networks. He/she should also like mathematics and probability, and should have excellent computer programming skills.

Research Advisor: Fragkiskos Papadopoulos, Assistant Professor, f.papadopoulos@eecei.cut.ac.cy

 

  • One (1) post in the following topic:  “Geometric Analysis and Dynamics of Brain Networks”

Description: Mapping the structural and functional connections of the human brain is one of the great scientific challenges of the 21st century, and real data with unprecedented resolution in space and time are being made publicly available for the first time (http://www.humanconnectome.org/). In this context, a great deal of recent research studies brain dynamics; the dynamics of the functional brain connectivity, i.e., the functional connections and disconnections taking place in the brain, at rest, during various tasks, or during abnormal behaviors, such as epileptic seizures. Furthermore, it has been recently recognized that the brain’s structural and functional systems have features common to other complex networks found in nature and society.

The PhD candidate will focus on: (i) data extraction and graph-theoretic analysis of brain network data from the human connectome project; (ii) mapping of these network data into different geometric spaces; (iii) studying the spatial dynamics of network nodes in these spaces; (iv) identifying laws/processes that can potentially describe these spatial dynamics; and (v) use the discovered laws to predict brain network dynamics.

This position falls under the general scientific areas of Network Science, Data Science, Brain Science, and Predictive Analytics. The ideal candidate should like networks. He/she should also like mathematics (especially statistics), and aspects of neuroscience. He should also have excellent computer programming skills. The research will take place in collaboration with researchers from the Department of Bioengineering at McGill University, Canada.

Research Advisor: Fragkiskos Papadopoulos, Assistant Professor, f.papadopoulos@eecei.cut.ac.cy

 

  • One (1) post in the following topic:  “Evaluation of a Magnetic Resonance Imaging (MRI) Guided Focused Ultrasound System for Ablation in the Abdominal Area”

Description: Focused ultrasound is a modality that can be used to treat various diseases in the area of oncology using thermal protocols. The thermal effects of Focused ultrasound can be monitored with excellent contrast using Magnetic resonance imaging (MRI).

The offered position will concentrate on the evaluation of an existing 4-D MR compatible robotic system. A major task is to design an agar-based prostate phantom.  Simulations will be performed in order to optimize the focused ultrasound therapeutic protocols.  A transducer design dedicated for ablation in the abdominal area will be performed.  The successful applicant is expected to extensively evaluate the system in the developed phantom in the laboratory setting and inside an MRI scanner.  MRI sequences will be optimized in order to monitor the thermal effects of ultrasound.

Required qualifications: MSc in Electrical Engineering, or Mechanical Engineering, or Physics.

Research Advisor: Christakis Damianou, Professor, christakis.damianou@cut.ac.cy

 

  • One (1) position in the following field: “Distributed Data-Driven Multiprocessing”

Description: High-Performance Computing (HPC) is seen as the only way so solve mankind’s pending big problems that require computational capability measured in terms of at least a million trillion of computations per second (ten to the power of 18), i.e., exascale. Such problems involve reverse engineering the human brain, creating medicine to eradicate diseases such as cancer, and simulating weather phenomena to predict climate change. This Doctoral Thesis concerns the research and development of a novel distributed multiprocessor architecture to address the power and concurrency challenges of future HPC/exascale systems. The system will be based on a Hybrid Data-Flow model, the Data-Driven Multithreading (DDM) model of execution. The multiprocessor architecture will be implemented and evaluated on a large capacity Field-Programmable Gate Array (FPGA) and will consist of low-power and low-complexity non-coherent processing elements and hardware support for the DDM model.  Additionally, it will incorporate a lightweight, mainly cache-based memory hierarchy, augmented with automated deterministic prefetching into scratchpad memories. Last, an Application Programming Interface (API) in C++ will be implemented to allow programmers to develop DDM applications for rapid architectural prototyping and evaluation. This Doctoral Thesis will build upon existing infrastructure in terms of both hardware and software, and extensive know-how that the Research Team has built over the years.

Required Qualifications: Candidates should possess a Bachelor’s Degree and a Master’s-level postgraduate degree from accredited universities in the field of Computer Science, or Electrical Engineering, or Computer Engineering with a preferred specialization in either computer architecture, distributed systems and networks, embedded systems, or related.  The candidate should have 2+ years of experience in object-oriented programming and be fluent in C++ programming and/or Python, and also possess experience in parallel and distributed computing (i.e., PThreads, OpenMP and MPI). Next, the candidate should have 2+ years of working experience with hardware description languages such as VHDL or Verilog. Excellent command of the English language is a must. Any research experience with data-flow/data-driven models (e.g., TBB, OmpSs, etc.) and command in Xilinx HDL tools (ISE or Vivado Design Suites) will be considered as an advantage.

Advisor:  Vassos Soteriou, Associate Professor, vassos.soteriou@cut.ac.cy

 

  • One (1) post in the following field: “Resilient Wear-Aware Computer Architectures”

Description: Moore’s Law scaling continues to yield higher transistor density with each succeeding process generation, leading to today’s many-core chip multiprocessors (CMPs) with hundreds of interconnected cores or tiles. Unfortunately, deep submicron CMOS process technology is marred by increasing susceptibility to wear. Prolonged operational stress gives rise to accelerated wear out and failure due to several physical failure mechanisms, including hot-carrier injection (HCI), electro-migration (EM), and negative-bias temperature instability (NBTI). Unfortunately such wear can prove catastrophic to the reliable operation of CMPs, as various chip components may introduce errors and/or timing violations during computation and data transportation across the chip, deeming it inoperable. To avoid such detrimental effects this Doctoral Thesis will deal with the development of wear out-decelerating techniques so as to slow-down wear in CMP components and improve their resilience, including processors, memory and on-chip interconnect. Such techniques will be incorporated seamlessly into the existing CMP architecture to work online during chip operation without any intervention from the programming stack or the user. Since wear in CMOS transistors is usage-based, and correlates heavily on how workloads utilize them over time, a key drive of this Thesis will be to understand the usage patterns of applications so as to adopt appropriate wear-aware policies to them for maximum positive lifetime extending effect. As such, wear-aware policies may be based on artificial neutral network techniques or algorithms which are very good in recognizing patterns and adapting to them. Other pattern recognition schemes will also be examined to further explore the design space of wear-reducing architectural-level policies. Wear-aware hardware augmentations to the base CMP architecture will be implemented using hardware description languages (e.g., VHDL) to prove their feasibility. This Doctoral Thesis will build upon existing and extensive know-how that the Research Team has developed over the years, and will utilize and substantially extent existing methods from the field of wear-aware multiprocessor architectures.

Required Qualifications: Candidates should possess a Bachelor’s Degree and a Master’s-level postgraduate degree from accredited universities in the field of Computer Science, or Computer Engineering, or Electrical Engineering with a preferred specialization in either computer architecture, distributed systems, interconnection/computer networks, embedded systems, artificial neural network architectures and algorithms, or related.  The candidate should have 2+ years of experience in object-oriented programming and be fluent in C++ programming and/or Python, and also have a good command in calculus. Last, the candidate should have 2+ years of working experience with hardware description languages such as VHDL or Verilog. Excellent command of the English language is a must. Good knowledge in using Xilinx HDL tools (ISE or Vivado Design Suites) will be considered as an advantage.

Research Advisor:  Vassos Soteriou, Associate Professor, vassos.soteriou@cut.ac.cy

 

  • Two (2) Posts  in any of the following subjects:

“Software Reliability”

Description: Methods, techniques, models and algorithms for studying software reliability. Software Reliability Growth Models (SRGM) based on mathematical and statistical approaches. Use of empirical data measured from real world software systems. Application of non-linear dynamics and time-series analysis for revealing the nature of software reliability in various application types (classic, web-based, mobile, etc.) Use of Computational Intelligence or/and of other sub-areas of Computer Science and Engineering for improving SRGM.

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc in Software Engineering) will be considered as advantage.

Funding: The candidates with the appropriate qualifications can be funded as participants in research projects or as teaching assistants.

 

“Automated Software Testing”

Description: Methods, techniques, models and algorithms for performing software testing in an automated way, with little or no human intervention. Use of Computational Intelligence or/and of other sub-areas of Computer Science for performing black-box (specifications-based) and glass-box (source code-based) testing for classic software systems, web applications and mobile software.

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc in Software Engineering) will be considered as advantage

 

“Software Engineering for the Cloud”

Description: The research to be conducted will revolve around issues of software development for the Cloud environment. This new environment poses several restrictions to the way we usually follow to develop classic software and necessitates the study of parameters to help raising the quality of software systems. In addition, the Cloud requires elasticity and automation of the development process to speed up release times and satisfy clients’ requirements for fast change. Finally, this thesis will investigate DevOps as there is strong need to bridge the two teams involved, the one that develops the software (Dev) and the one that operates (manages) it after its delivery (Ops). In this context, new life cycle models tailored to the need of the Cloud will be proposed, along with software development methodologies and techniques that will address issues like automatic detection of Service Level Agreements (SLA) violations, automatic software testing, reduction of cycle time and release time, etc. The research will utilize Computational Intelligence notions which will be combined with core software engineering subjects like Agile Processes, software testing, project management, team organization etc. This research will be supported by collaboration activities with Politecnico di Milano and University of Tilburg under a Horizon 2020 Twinning project that was recently awarded to our group.

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc in Software Engineering) or any involvement with research in the past will be considered as advantage.

 

“Automatic Resource Management for the Cloud”

Description: This research topic will concentrate on algorithms, methods and techniques for automating certain process in the Cloud environment dealing with how resources are managed. More specifically, Computational Intelligence – CI approaches will be utilized to tackle issues and solve problems related to optimizing the way resources are managed (e.g. physical servers, virtual machines, etc.) in such a way so that clients are serviced according to their Service Level Agreements – SLA, with high quality and performance, but at the same time energy and cost preservation is taken into consideration. Fog computing will also be investigated as the paradigm that pushes processing intelligence and data down to the local area network level of network architecture and a fog node to avoid latencies.

In this context different CI models will be investigated and apply in single- and multi-objective optimization of Cloud resources. This research will be supported by collaboration activities with Politecnico di Milano and University of Tilburg under a Horizon 2020 Twinning project that was recently awarded to our group.

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc in Software Engineering) or any involvement with research in the past will be considered as advantage.

 

“Software Engineering and Intelligent Data Processing in Distributed Blockchain Processing Systems”

Research will focus on methodologies and techniques for the development of software systems for distributed processing in the new computational model of Blockchain. Particular emphasis will be placed on the automation of smart data processes in this environment, using models and algorithms of Computational and Artificial Intelligence, as well as single and multi-objective optimization methods. This subject will apply research results to real-world systems that are already running or are going to be developed in the Blockchain environment, mainly from the Financial Sector, and may be supported by local private organizations by means of data provision and/or funding (currently in discussions with positive reactions).

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc) in Software Engineering and/or Computational or Artificial Intelligence, or any involvement with research in the past will be considered as advantage.

Research Advisor: Andreas S. Andreou, Professor, andreas.andreou@cut.ac.cy

 

  • One (1) post in the following field: “Event Detection, Localization and Tracking using Wireless Sensor Networks”

Description: Wireless Sensor Networks (WSNs) are a fairly new technology that can potentially provide an interface between the physical world and computers allowing the latter to vanish into the background. They have a wide variety of applications including military sensing, infrastructure security, environment and habitat monitoring, industrial sensing, building and structure monitoring, and traffic control.  The proposed research is expected to be based on ideas and techniques from a variety of different fields including Wireless Communication Systems, Computer Networks, Collaborative Signal and Information Processing and Computational Intelligence. The offered positions will concentrate on the development of new algorithms and techniques for detecting, localizing and tracking an event. The developed algorithms should feature low computational complexity, distributed implementation and fault tolerance in order to address the limitations of WSNs in terms of energy and bandwidth and the harsh conditions of operation. The successful applicants are expected to perform real–time experiments in order to verify the performance of their algorithms using the WSN platform at the Cyprus University of Technology.    

Required qualifications: BSc (required) and MSc (preferably) in Electrical Engineering and/or Computer Science. Prior research experience or specialization in related topics will be considered an advantage.

Research Advisor: Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

 

  • One (1) position in the following field: “Contaminant Event Monitoring in Intelligent Buildings”

Description: An Intelligent Building is a system that incorporates computer technology to autonomously govern and adapt the building environment in order to enhance operational and energy efficiency, cost effectiveness, improve user’s comfort, productivity and safety, and increase system robustness and reliability. The dispersion of contaminants from sources (events) inside a building can compromise the indoor air quality and influence the occupants' comfort, health, productivity and safety. These events could be the result of an accident, faulty equipment or a planned attack. Under these safety-critical conditions, immediate event detection should be guaranteed and the proper actions should be taken to ensure the safety of the people. The proposed research will investigate and produce solutions for the problem of monitoring the indoor building environment against the presence of contaminant events.  Distributed sensor networks have been widely used in buildings to monitor indoor environmental conditions such as air temperature, humidity and contaminant concentrations (e.g. CO, CO2). The goal of this research will be the development of methods for interpreting the real-time-collected data coming from the sensors in order to ensure the accurate and prompt identification of contaminant sources. The results can help determine appropriate control solutions such as: (i) indicating safe rescue pathways and/or refugee spaces, (ii) isolating contaminated spaces and (iii) cleaning contaminant spaces by removing sources, ventilating and filtering air.  

Required qualifications: BSc (required) and MSc (preferably) in Electrical Engineering and/or Computer Science. Prior research experience or specialization in related topics will be considered an advantage.

Research Advisor: Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

 

  • One (1) position in the following field: “Air Quality Monitoring in Smart Cities using Wireless Sensor Networks”

Description: Currently, there is a lack of sufficient infrastructure for environmental monitoring, both spatially (in multiple points) and temporally (in regular time intervals). The proposed wireless sensor network can constitute an economical and reliable solution to the problem of sufficient monitoring and control of the city air quality. The proposed research will focus on the development of innovative algorithms and techniques for detecting, identifying and tracking the release of pollutants in an urban environment using a wireless sensor network. More specifically, the successful candidate is expected to use signal processing and machine learning methods to analyze the collected data from the sensors in order to: (i) construct a fine-grained pollution map of the city, (ii) identify the main sources of pollution and estimate their locations, (iii) develop models for predicting the pollution levels in the near future. These results are expected to provide the necessary information for reducing the pollution levels through appropriate actions and policies, leading to a cleaner and safer city environment.  The successful applicants are expected to work with real data in order to verify the performance of their algorithms using the established WSN platform at the Cyprus University of Technology.

Required qualifications: BSc (required) and MSc (preferably) in Electrical Engineering and/or Computer Science. Prior research experience or specialization in related topics will be considered an advantage.

Research Advisor: Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

 

  • One (1) position in the following field: “Wireless Sensor Networks for Smart Ports”

Description: Ports play a crucial role in connecting the European markets with island areas, such as Cyprus, as 74% of goods imported or exported from Europe travel through sea transport. Wireless Sensor Networks (WSNs) are a fairly new technology that can potentially improve the efficiency of the various port operations and services via enabling real-time situation awareness to all participants involved in maritime activities in the ports of Cyprus. The proposed research will investigate the implementation of such a WSN at the Port of Limassol for collecting real-time information related to ship movements, the environment, and the tracking of cargo by exploiting innovative technological solutions such as buoys, UAVs, and RFIDs. The collected information, will be further processed using advanced data analytics for ensuring high quality data, calculating KPIs, and creating new decision-support tools and services for maritime stakeholders. The successful applicants are expected to perform real-time experiments in order to verify the performance of the developed algorithms and solutions using the WSN platform at the Port of Limassol.

Required Qualifications:  Undergraduate (BSc) and postgraduate (MSc) degrees in Electrical Engineering and/or Computer Science or related field. Prior research experience or specialization in related topics will be considered an advantage.

Funding: There is possibility of funding through involvement in a Research Program for full-time qualified applicants.

Research Advisors:

Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

Herodotos Herodotou Assistant Professor, herodotos.herodotou@cut.ac.cy

 

  • One (1) post in the following field: “Big Data Infrastructure and Analytics for the Maritime Industry”

Description: The maritime industry has been essential towards European economic growth and prosperity, as 74% of goods imported or exported from Europe travel through sea transport and ports. The digitization of port information and the introduction of new technologies such as AIS, sensors on buoys and UAVs, and RFIDs, has led to an explosion of maritime-related data that hold the key to improving the efficiency of the various port operations and services. The proposed research involves designing and developing the necessary big data infrastructure, which will form a comprehensive solution for (i) collecting and integrating data efficiently from various sources; (ii) enabling stream processing in real time; (iii) storing the data in a fault-tolerant way; and (iv) supporting machine learning and advanced analytical processing of data with the goal of extracting new meaningful insights and supporting decision-making activities for the maritime stakeholders. The proposed platform will be validated via processing and storing real-time data related to ship movements, the environment, and the tracking of cargo at the Port of Limassol.

Required Qualifications: Undergraduate (BSc) and postgraduate (MSc) degrees in Computer Science or related field. The ideal candidate should enjoy working on cutting-edge systems research problems and have good software development skills. Prior research experience or specialization in related topics will be considered an advantage.

Funding: There is possibility of funding through involvement in a Research Program for full-time qualified applicants.

Research Advisors:

Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

Herodotos Herodotou Assistant Professor, herodotos.herodotou@cut.ac.cy

 

  • One (1) position in the following field: “Smart Resource Management in Distributed Stream Data Processing Systems”

Description: The need to reduce the gap between the generation of data and extraction of insights from these data has led to significant innovations for distributed stream data processing engines (DSPEs). Such systems are driven by a data-centric model that allows for near real-time consumption and analysis of data. However, ensuring good and robust system performance at large scale for streaming applications poses several new challenges, including the distributed management of system resources, the heterogeneity of computing clusters, and the increased complexity of streaming applications. The goal is to develop automated resource management techniques for addressing the aforementioned challenges in order to make DSPEs more robust in their operating characteristics. Specifically, the PhD candidate will design and implement new algorithms and tools for optimizing the allocation of compute resources (e.g., CPU cores, memory) to both the underlying streaming engines and the applications processing the data streams. In addition, the tools will be responsible for monitoring the cluster utilization, finding and eliminating rogue usage, and defining automatic actions in response to anomalies and inefficiencies with the dual goal of maximizing resource utilization and minimizing application delays.

Required Qualifications: Undergraduate (BSc) and postgraduate (MSc) degrees in Computer Science or related field. The ideal candidate should enjoy working on cutting-edge systems research problems and have good software development skills. Prior research experience or specialization in related topics will be considered an advantage.

Funding: There is possibility of funding through involvement in a Research Program for full-time qualified applicants or as teaching assistants.

Research Advisor: Herodotos Herodotou Assistant Professor, herodotos.herodotou@cut.ac.cy

 

  • One (1) post in the following topic: “Development of Optical Fiber Plasmonic Sensors and Nanoantennas Using Femtosecond Laser Pulses”

Required Qualifications:  BSc and/or MSc in Electrical Engineering or Physics, or any other related subject. Strong mathematical background will be considered an advantage.

The PhD will focus on the development of photonic (bio)chemical sensing platforms, using custom sensors developed in-house with a femtosecond laser system. The PhD will focus on tilted fibre Bragg gratings surrounded by nanoscale coatings of metal layers and nanoparticles that will be studied and optimized to exploit the plasmonic enhancement of the sensing transduction mechanisms.

Research Advisors:   Kyriacos Kalli, Professor, kyriacos.kalli@cut.ac.cy

 

  • One (1) post in the following topic:  “Optical Fibre Sensors for Biomedical Applications”

Required Qualifications: BSc and/or MSc in Electrical Engineering or Physics, or any other related subject. Strong mathematical background will be considered an advantage.

Research Advisors:   Kyriacos Kalli, Professor, kyriacos.kalli@cut.ac.cy

 

  • One (1) post in the following topic: “Predictive Learning Algorithms for Distributed Acoustic Sensor (DAS) Networks for Oil and Gas Pipelines”

Description: The goal of this thesis is to develop novel machine learning algorithms, suitable for performing data-driven predictive tasks in the context of Distributed Acoustic Sensor (DAS) networks. DAS environments generate data that entail great deals of epistemic uncertainty, due to several hard to model artifacts, such as skewness, heavy tails, non-stationarity, and measurement noise. These data properties call for the development of deep generative models with novel statistical assumptions, that are not yet reported in the related literature. In addition, the very nature of DAS networks necessitates the development of novel distributed inference algorithms, as well as sensor hardware that effectively facilitates the operation of such algorithms. This thesis will address these challenges in a comprehensive way. We will develop in-house sensor networks for our experimentations, and will leverage state-of-the-art machine learning software, such as TensorFlow. There is also the strong prospect of real-world deployment and validation of our novel solutions, in the context of our existing collaboration with a world leader in Fiber-optic sensing technology. This thesis requires some basic affinity with DAS and statistical modelling.

Research Advisors:

Kyriacos Kalli, Associate Professor, kyriacos.kalli@cut.ac.cy

Sotirios Chatzis, Assistant Professor, sotirios.chatzis@eecei.cut.ac.cy .

 

  • One (1) post in the following topic:  “Night Cooling Systems: Modeling and monitoring systems”

Required Qualifications: BSc and/or MSc in Electrical Engineering or Physics, or any other related subject. Strong mathematical background will be considered an advantage.

Research Advisor:  Paul Christodoulides, Assistant Professor, paul.christodoulides@cut.ac.cy

 

  • One (1) post in the following topic:  Application of Structural Equation Models and Satellite orbits

Required Qualifications: BSc and/or MSc in Electrical Engineering or Physics, or any other related subject. Strong mathematical background will be considered an advantage.

Research Advisor:  Paul Christodoulides, Assistant Professor, paul.christodoulides@cut.ac.cy

  • One (1) position in the following field: “Mathematical modeling and performance analysis of micro-network router architectures and traffic flows”

Description: This research topic concentrates on elaborate detailed mathematical models to capture in detail the behavior of the architecture of pipelined micro-routers utilized in micro-interconnect networks found in today’s multi-core processors and embedded systems. Various state-of-the-art architectures will be considered. The interaction of the underlying router organization will be considered in tandem with various flow-control protocols and routing algorithms, along with numerous traffic flow spatio-temporal behaviors in order to determine throughput and network latency levels that will act as indicators of architectural router performance. Results obtained from software simulations of equivalent architectures will be carried out to confirm the validity and the accuracy of the mathematically-modeled micro-router architectures.

Required Qualifications: Candidates should possess a Bachelor’s Degree and a Master’s-level postgraduate degree from accredited universities in the field of Computer Science, or Electrical Engineering, or Computer Engineering or Mathematics with a preferred specialization in either computer architecture, computer networks, discrete mathematics, statistics, or related.  The candidate should have 2+ years of experience in object-oriented programming and be fluent in C++ programming and/or Python. A strong mathematical background is desired. Excellent command of the English language is a must.

Research Advisors:  

Vassos Soteriou, Associate Professor, vassos.soteriou@cut.ac.cy

Paul Christodoulides, Assistant Professor, paul.christodoulides@cut.ac.cy

 

  • One (1) position in the following field: “New techniques for data storage and archiving of massive and complex amounts of 2D/3D/4D Cultural assets”

Description: Cultural Heritage (CH) is an integral element of Europe and vital for the creation of a common European identity. The rapid growth of technology has led to mass digitization of cultural assets, requiring for their cost–effective preservation, documentation, protection and presentation in online digital libraries. The aim is to shed light, through technological innovation and digital media, on all aspects of cultural heritage, both tangible (books, newspapers, photographs, drawings, manuscripts, costumes, maps, objects, archaeological sites, monuments) and intangible (eg, music, performing arts, folklore, theater), as well as their semantic interrelations, and finally enhancing their added value by reusing them in the fields of education, tourism industry, advertising and art

The proposed research will focus on (a) the study and analysis of massive and complex amounts of multimedia 3D/4D data, (b) study and analysis of data storage and archiving in multimedia digital libraries, (c) the development of innovative methodologies for harvesting of such data sets in digital libraries, taking into account object’s semantic signatures, and finally, (d) the development of innovative methodologies for reuse of such complex structures from digital libraries.

Research Advisor: Marinos Ioannides, Senior Lecturer, marinos.ioannides@cut.ac.cy

 

  • One (1) position in the following field:Holistic Heritage Management”

Description: Heritage Management is a multiparametric field facing nowadays a variety of challenges. The progressive expansion of the term of Cultural Heritage (CH) has led to a type of management of it (CH), which goes beyond the conservation and restoration of cultural assets. A wide spectrum of values, a variety of involved stakeholders, multiple, even conflicting, objectives, are only some of the challenges CH is facing. Even nowadays involved authorities and stakeholders act within their own narrow spectrum without taking into consideration a number of other interrelated parameters; an attitude which not rarely results to fragmented and not so beneficial interventions. The proposed project aims to approach Heritage Management in a holistic way; As a “procedure” of management, starting from the phase of data acquisition, but also as a “result”, leading to concrete actions; As an embracement not only of the lifecycle of the cultural asset, but also of the lifecycle of the human, starting at his early schooling age, since human is the provider but also the user of CH. For the achievement of this goal a continuous shift between different scientific domains, the skilful management of differentiated input and its transformation into new information and knowledge, exploitable by various sectors, becomes crucial. For this reason it is needed: a broad educational background on Arts and Culture, the tools and the methodological thinking of engineering as well as the pedagogical techniques, in order for CH to become an actual “public asset”. Required Qualifications:  A BSc and MSc degree in Architecture, an MSc in the field of Cultural Heritage as well as pedagogical education. Prior research experience or specialization in Cultural Heritage and Education will be considered an advantage.

Research Advisor: Marinos Ioannides, Senior Lecturer, marinos.ioannides@cut.ac.cy

 

  • One (1) position in the following field: “Applying Machine Learning methods in processing Cultural Heritage assets”

Description: Cultural Heritage is the legacy of a nation from previous generations, for which efforts are made maintain their present status but also to safeguard its future existence. Nowadays, the technological outbreak has led to the development of intelligent systems, which can actively contribute in areas like the documentation, preservation and promotion of Cultural Heritage. Machine Learning constitutes an integral part of intelligent systems as it is a category of artificial intelligence, which enables modern computer systems to "learn" to develop and adapt their function upon exposure to new data.

The proposed research will be focused on the development of machine learning methods for their use in cultural applications. As part of the research activities will be the study of existing machine learning methods (supervised, non-supervised, reinforcement) which are currently used for the classification of cultural assets over time.

Required Qualifications: Applicants should have a BSc and an MSc degree in Computer Science, Science of Electrical Engineering or other related field. Previous research experience in the study and the application of machine learning in Cultural Heritage sector will be considered an asset.

Research Advisor: Marinos Ioannides, Senior Lecturer, marinos.ioannides@cut.ac.cy

 

Information:

Department Secretary Tel: 25002533, Fax: 25002635

Electrical Engineering, Computer Engineering and Informatics

PhD positions, starting September 2018

The last date to apply for postgraduate studies is Friday, 13th of April 2018.

For applications, click  here

  • One (1) post in the following topic: “Neural attention mechanisms for learning to detect subtlefinancial crime patterns in big data”

Description: Financial crime, including tax evasion and financial fraud, is a major problem that plagues European economies. Typically, local European authorities employ various heuristic criteria to detect it, which require extensive human labor.

This thesis envisages the examination and development of novel machine learning techniques to facilitate automatic detection and characterization of patterns of such illicit activity. Specifically, we will develop statistical machine learning models that learn to extract salient temporal dynamics in vast amounts of data. We will focus on neural attention mechanisms that entail differentiable memory modules; these offer an unprecedented capacity to learn over long temporal horizons, cope with abrupt as well as slower pattern shifts, and learn to distinguish between salient and less important dynamics. Their success in hard problems, such as speech recognition and natural language understanding, provide strong promise   for their success in this context, whereby it is needed that they correlate across millions of individuals or corporates over long time horizons, in a way that humans cannot easily do.

Research Advisors:  Sotiris Chatzis, Assistant Professor, sotirios.chatzis@cut.ac.cy

 

  • One (1) post in the following topic: “Novel deep learning models for neural machine summarization”

Description: This thesis will develop novel neural statistical models for abstractive document summarization. Our research will focus on: 

(i) Policy learning algorithms for model training, which may enable us to ameliorate the “exposure bias” problem of the current state-of-the-art paradigm;

(ii) Novel statistical formulations of attention mechanisms and differentiable memory modules that may give rise to more potent mechanisms for capturing and modelling sequential dynamics.

Required Qualifications: We expect that the candidates will have some rudimentary affinity with linear algebra concepts, and the Python programming language. We will definitely consider a plus some experience with fundamental machine learning techniques, especially maximum-likelihood training and regularization techniques.  

Research Advisors:  Sotiris Chatzis, Assistant Professor, sotirios.chatzis@cut.ac.cy

 

  • One (1) post in the following topic: “Complex Networks: Geometry, Dynamics and Prediction”

Description: Real-world complex networked systems (technological, biological, social, financial, etc.) can be mapped to geometric spaces that lie hidden beneath their observable topologies. These geometric spaces are called “hidden”, as they play the role of an underlying coordinate system, not readily observable by examining the network topology. Nodes closer in the underlying space are connected in the observable network topology with higher probability.

The PhD candidate will focus in studying the properties of these underlying geometric spaces and the spatial dynamics of network nodes in these spaces. It is anticipated that important fundamental and practical questions will be addressed through this PhD thesis such as: (i) what are the laws governing the “motion” of network nodes in these spaces? (ii) Can this motion be modeled using classical mechanics laws, e.g., Newton’s laws of motion or stochastic versions of it? (iii) Is this motion chaotic or can be predicted? (iv) Given that we can predict this motion, can we predict the future structure and evolution of real complex networks?

This position falls under the general scientific areas of Network Science, Data Science, and Predictive Analytics. The ideal candidate should like networks. He/she should also like mathematics and probability, and should have excellent computer programming skills.

Research Advisor: Fragkiskos Papadopoulos, Assistant Professor, f.papadopoulos@eecei.cut.ac.cy

 

  • One (1) post in the following topic:  “Geometric Analysis and Dynamics of Brain Networks”

Description: Mapping the structural and functional connections of the human brain is one of the great scientific challenges of the 21st century, and real data with unprecedented resolution in space and time are being made publicly available for the first time (http://www.humanconnectome.org/). In this context, a great deal of recent research studies brain dynamics; the dynamics of the functional brain connectivity, i.e., the functional connections and disconnections taking place in the brain, at rest, during various tasks, or during abnormal behaviors, such as epileptic seizures. Furthermore, it has been recently recognized that the brain’s structural and functional systems have features common to other complex networks found in nature and society.

The PhD candidate will focus on: (i) data extraction and graph-theoretic analysis of brain network data from the human connectome project; (ii) mapping of these network data into different geometric spaces; (iii) studying the spatial dynamics of network nodes in these spaces; (iv) identifying laws/processes that can potentially describe these spatial dynamics; and (v) use the discovered laws to predict brain network dynamics.

This position falls under the general scientific areas of Network Science, Data Science, Brain Science, and Predictive Analytics. The ideal candidate should like networks. He/she should also like mathematics (especially statistics), and aspects of neuroscience. He should also have excellent computer programming skills. The research will take place in collaboration with researchers from the Department of Bioengineering at McGill University, Canada.

Research Advisor: Fragkiskos Papadopoulos, Assistant Professor, f.papadopoulos@eecei.cut.ac.cy

 

  • One (1) post in the following topic:  “Evaluation of a Magnetic Resonance Imaging (MRI) Guided Focused Ultrasound System for Ablation in the Abdominal Area”

Description: Focused ultrasound is a modality that can be used to treat various diseases in the area of oncology using thermal protocols. The thermal effects of Focused ultrasound can be monitored with excellent contrast using Magnetic resonance imaging (MRI).

The offered position will concentrate on the evaluation of an existing 4-D MR compatible robotic system. A major task is to design an agar-based prostate phantom.  Simulations will be performed in order to optimize the focused ultrasound therapeutic protocols.  A transducer design dedicated for ablation in the abdominal area will be performed.  The successful applicant is expected to extensively evaluate the system in the developed phantom in the laboratory setting and inside an MRI scanner.  MRI sequences will be optimized in order to monitor the thermal effects of ultrasound.

Required qualifications: MSc in Electrical Engineering, or Mechanical Engineering, or Physics.

Research Advisor: Christakis Damianou, Professor, christakis.damianou@cut.ac.cy

 

  • One (1) position in the following field: “Distributed Data-Driven Multiprocessing”

Description: High-Performance Computing (HPC) is seen as the only way so solve mankind’s pending big problems that require computational capability measured in terms of at least a million trillion of computations per second (ten to the power of 18), i.e., exascale. Such problems involve reverse engineering the human brain, creating medicine to eradicate diseases such as cancer, and simulating weather phenomena to predict climate change. This Doctoral Thesis concerns the research and development of a novel distributed multiprocessor architecture to address the power and concurrency challenges of future HPC/exascale systems. The system will be based on a Hybrid Data-Flow model, the Data-Driven Multithreading (DDM) model of execution. The multiprocessor architecture will be implemented and evaluated on a large capacity Field-Programmable Gate Array (FPGA) and will consist of low-power and low-complexity non-coherent processing elements and hardware support for the DDM model.  Additionally, it will incorporate a lightweight, mainly cache-based memory hierarchy, augmented with automated deterministic prefetching into scratchpad memories. Last, an Application Programming Interface (API) in C++ will be implemented to allow programmers to develop DDM applications for rapid architectural prototyping and evaluation. This Doctoral Thesis will build upon existing infrastructure in terms of both hardware and software, and extensive know-how that the Research Team has built over the years.

Required Qualifications: Candidates should possess a Bachelor’s Degree and a Master’s-level postgraduate degree from accredited universities in the field of Computer Science, or Electrical Engineering, or Computer Engineering with a preferred specialization in either computer architecture, distributed systems and networks, embedded systems, or related.  The candidate should have 2+ years of experience in object-oriented programming and be fluent in C++ programming and/or Python, and also possess experience in parallel and distributed computing (i.e., PThreads, OpenMP and MPI). Next, the candidate should have 2+ years of working experience with hardware description languages such as VHDL or Verilog. Excellent command of the English language is a must. Any research experience with data-flow/data-driven models (e.g., TBB, OmpSs, etc.) and command in Xilinx HDL tools (ISE or Vivado Design Suites) will be considered as an advantage.

Advisor:  Vassos Soteriou, Associate Professor, vassos.soteriou@cut.ac.cy

 

  • One (1) post in the following field: “Resilient Wear-Aware Computer Architectures”

Description: Moore’s Law scaling continues to yield higher transistor density with each succeeding process generation, leading to today’s many-core chip multiprocessors (CMPs) with hundreds of interconnected cores or tiles. Unfortunately, deep submicron CMOS process technology is marred by increasing susceptibility to wear. Prolonged operational stress gives rise to accelerated wear out and failure due to several physical failure mechanisms, including hot-carrier injection (HCI), electro-migration (EM), and negative-bias temperature instability (NBTI). Unfortunately such wear can prove catastrophic to the reliable operation of CMPs, as various chip components may introduce errors and/or timing violations during computation and data transportation across the chip, deeming it inoperable. To avoid such detrimental effects this Doctoral Thesis will deal with the development of wear out-decelerating techniques so as to slow-down wear in CMP components and improve their resilience, including processors, memory and on-chip interconnect. Such techniques will be incorporated seamlessly into the existing CMP architecture to work online during chip operation without any intervention from the programming stack or the user. Since wear in CMOS transistors is usage-based, and correlates heavily on how workloads utilize them over time, a key drive of this Thesis will be to understand the usage patterns of applications so as to adopt appropriate wear-aware policies to them for maximum positive lifetime extending effect. As such, wear-aware policies may be based on artificial neutral network techniques or algorithms which are very good in recognizing patterns and adapting to them. Other pattern recognition schemes will also be examined to further explore the design space of wear-reducing architectural-level policies. Wear-aware hardware augmentations to the base CMP architecture will be implemented using hardware description languages (e.g., VHDL) to prove their feasibility. This Doctoral Thesis will build upon existing and extensive know-how that the Research Team has developed over the years, and will utilize and substantially extent existing methods from the field of wear-aware multiprocessor architectures.

Required Qualifications: Candidates should possess a Bachelor’s Degree and a Master’s-level postgraduate degree from accredited universities in the field of Computer Science, or Computer Engineering, or Electrical Engineering with a preferred specialization in either computer architecture, distributed systems, interconnection/computer networks, embedded systems, artificial neural network architectures and algorithms, or related.  The candidate should have 2+ years of experience in object-oriented programming and be fluent in C++ programming and/or Python, and also have a good command in calculus. Last, the candidate should have 2+ years of working experience with hardware description languages such as VHDL or Verilog. Excellent command of the English language is a must. Good knowledge in using Xilinx HDL tools (ISE or Vivado Design Suites) will be considered as an advantage.

Research Advisor:  Vassos Soteriou, Associate Professor, vassos.soteriou@cut.ac.cy

 

  • Two (2) Posts  in any of the following subjects:

“Software Reliability”

Description: Methods, techniques, models and algorithms for studying software reliability. Software Reliability Growth Models (SRGM) based on mathematical and statistical approaches. Use of empirical data measured from real world software systems. Application of non-linear dynamics and time-series analysis for revealing the nature of software reliability in various application types (classic, web-based, mobile, etc.) Use of Computational Intelligence or/and of other sub-areas of Computer Science and Engineering for improving SRGM.

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc in Software Engineering) will be considered as advantage.

Funding: The candidates with the appropriate qualifications can be funded as participants in research projects or as teaching assistants.

 

“Automated Software Testing”

Description: Methods, techniques, models and algorithms for performing software testing in an automated way, with little or no human intervention. Use of Computational Intelligence or/and of other sub-areas of Computer Science for performing black-box (specifications-based) and glass-box (source code-based) testing for classic software systems, web applications and mobile software.

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc in Software Engineering) will be considered as advantage

 

“Software Engineering for the Cloud”

Description: The research to be conducted will revolve around issues of software development for the Cloud environment. This new environment poses several restrictions to the way we usually follow to develop classic software and necessitates the study of parameters to help raising the quality of software systems. In addition, the Cloud requires elasticity and automation of the development process to speed up release times and satisfy clients’ requirements for fast change. Finally, this thesis will investigate DevOps as there is strong need to bridge the two teams involved, the one that develops the software (Dev) and the one that operates (manages) it after its delivery (Ops). In this context, new life cycle models tailored to the need of the Cloud will be proposed, along with software development methodologies and techniques that will address issues like automatic detection of Service Level Agreements (SLA) violations, automatic software testing, reduction of cycle time and release time, etc. The research will utilize Computational Intelligence notions which will be combined with core software engineering subjects like Agile Processes, software testing, project management, team organization etc. This research will be supported by collaboration activities with Politecnico di Milano and University of Tilburg under a Horizon 2020 Twinning project that was recently awarded to our group.

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc in Software Engineering) or any involvement with research in the past will be considered as advantage.

 

“Automatic Resource Management for the Cloud”

Description: This research topic will concentrate on algorithms, methods and techniques for automating certain process in the Cloud environment dealing with how resources are managed. More specifically, Computational Intelligence – CI approaches will be utilized to tackle issues and solve problems related to optimizing the way resources are managed (e.g. physical servers, virtual machines, etc.) in such a way so that clients are serviced according to their Service Level Agreements – SLA, with high quality and performance, but at the same time energy and cost preservation is taken into consideration. Fog computing will also be investigated as the paradigm that pushes processing intelligence and data down to the local area network level of network architecture and a fog node to avoid latencies.

In this context different CI models will be investigated and apply in single- and multi-objective optimization of Cloud resources. This research will be supported by collaboration activities with Politecnico di Milano and University of Tilburg under a Horizon 2020 Twinning project that was recently awarded to our group.

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc in Software Engineering) or any involvement with research in the past will be considered as advantage.

 

“Software Engineering and Intelligent Data Processing in Distributed Blockchain Processing Systems”

Research will focus on methodologies and techniques for the development of software systems for distributed processing in the new computational model of Blockchain. Particular emphasis will be placed on the automation of smart data processes in this environment, using models and algorithms of Computational and Artificial Intelligence, as well as single and multi-objective optimization methods. This subject will apply research results to real-world systems that are already running or are going to be developed in the Blockchain environment, mainly from the Financial Sector, and may be supported by local private organizations by means of data provision and/or funding (currently in discussions with positive reactions).

Required qualifications: BSc and/or MSc in Computer Science or Computer Engineering or Informatics or any other related field. Prior experience or specialization (i.e. during BSc or MSc) in Software Engineering and/or Computational or Artificial Intelligence, or any involvement with research in the past will be considered as advantage.

Research Advisor: Andreas S. Andreou, Professor, andreas.andreou@cut.ac.cy

 

  • One (1) post in the following field: “Event Detection, Localization and Tracking using Wireless Sensor Networks”

Description: Wireless Sensor Networks (WSNs) are a fairly new technology that can potentially provide an interface between the physical world and computers allowing the latter to vanish into the background. They have a wide variety of applications including military sensing, infrastructure security, environment and habitat monitoring, industrial sensing, building and structure monitoring, and traffic control.  The proposed research is expected to be based on ideas and techniques from a variety of different fields including Wireless Communication Systems, Computer Networks, Collaborative Signal and Information Processing and Computational Intelligence. The offered positions will concentrate on the development of new algorithms and techniques for detecting, localizing and tracking an event. The developed algorithms should feature low computational complexity, distributed implementation and fault tolerance in order to address the limitations of WSNs in terms of energy and bandwidth and the harsh conditions of operation. The successful applicants are expected to perform real–time experiments in order to verify the performance of their algorithms using the WSN platform at the Cyprus University of Technology.    

Required qualifications: BSc (required) and MSc (preferably) in Electrical Engineering and/or Computer Science. Prior research experience or specialization in related topics will be considered an advantage.

Research Advisor: Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

 

  • One (1) position in the following field: “Contaminant Event Monitoring in Intelligent Buildings”

Description: An Intelligent Building is a system that incorporates computer technology to autonomously govern and adapt the building environment in order to enhance operational and energy efficiency, cost effectiveness, improve user’s comfort, productivity and safety, and increase system robustness and reliability. The dispersion of contaminants from sources (events) inside a building can compromise the indoor air quality and influence the occupants' comfort, health, productivity and safety. These events could be the result of an accident, faulty equipment or a planned attack. Under these safety-critical conditions, immediate event detection should be guaranteed and the proper actions should be taken to ensure the safety of the people. The proposed research will investigate and produce solutions for the problem of monitoring the indoor building environment against the presence of contaminant events.  Distributed sensor networks have been widely used in buildings to monitor indoor environmental conditions such as air temperature, humidity and contaminant concentrations (e.g. CO, CO2). The goal of this research will be the development of methods for interpreting the real-time-collected data coming from the sensors in order to ensure the accurate and prompt identification of contaminant sources. The results can help determine appropriate control solutions such as: (i) indicating safe rescue pathways and/or refugee spaces, (ii) isolating contaminated spaces and (iii) cleaning contaminant spaces by removing sources, ventilating and filtering air.  

Required qualifications: BSc (required) and MSc (preferably) in Electrical Engineering and/or Computer Science. Prior research experience or specialization in related topics will be considered an advantage.

Research Advisor: Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

 

  • One (1) position in the following field: “Air Quality Monitoring in Smart Cities using Wireless Sensor Networks”

Description: Currently, there is a lack of sufficient infrastructure for environmental monitoring, both spatially (in multiple points) and temporally (in regular time intervals). The proposed wireless sensor network can constitute an economical and reliable solution to the problem of sufficient monitoring and control of the city air quality. The proposed research will focus on the development of innovative algorithms and techniques for detecting, identifying and tracking the release of pollutants in an urban environment using a wireless sensor network. More specifically, the successful candidate is expected to use signal processing and machine learning methods to analyze the collected data from the sensors in order to: (i) construct a fine-grained pollution map of the city, (ii) identify the main sources of pollution and estimate their locations, (iii) develop models for predicting the pollution levels in the near future. These results are expected to provide the necessary information for reducing the pollution levels through appropriate actions and policies, leading to a cleaner and safer city environment.  The successful applicants are expected to work with real data in order to verify the performance of their algorithms using the established WSN platform at the Cyprus University of Technology.

Required qualifications: BSc (required) and MSc (preferably) in Electrical Engineering and/or Computer Science. Prior research experience or specialization in related topics will be considered an advantage.

Research Advisor: Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

 

  • One (1) position in the following field: “Wireless Sensor Networks for Smart Ports”

Description: Ports play a crucial role in connecting the European markets with island areas, such as Cyprus, as 74% of goods imported or exported from Europe travel through sea transport. Wireless Sensor Networks (WSNs) are a fairly new technology that can potentially improve the efficiency of the various port operations and services via enabling real-time situation awareness to all participants involved in maritime activities in the ports of Cyprus. The proposed research will investigate the implementation of such a WSN at the Port of Limassol for collecting real-time information related to ship movements, the environment, and the tracking of cargo by exploiting innovative technological solutions such as buoys, UAVs, and RFIDs. The collected information, will be further processed using advanced data analytics for ensuring high quality data, calculating KPIs, and creating new decision-support tools and services for maritime stakeholders. The successful applicants are expected to perform real-time experiments in order to verify the performance of the developed algorithms and solutions using the WSN platform at the Port of Limassol.

Required Qualifications:  Undergraduate (BSc) and postgraduate (MSc) degrees in Electrical Engineering and/or Computer Science or related field. Prior research experience or specialization in related topics will be considered an advantage.

Funding: There is possibility of funding through involvement in a Research Program for full-time qualified applicants.

Research Advisors:

Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

Herodotos Herodotou Assistant Professor, herodotos.herodotou@cut.ac.cy

 

  • One (1) post in the following field: “Big Data Infrastructure and Analytics for the Maritime Industry”

Description: The maritime industry has been essential towards European economic growth and prosperity, as 74% of goods imported or exported from Europe travel through sea transport and ports. The digitization of port information and the introduction of new technologies such as AIS, sensors on buoys and UAVs, and RFIDs, has led to an explosion of maritime-related data that hold the key to improving the efficiency of the various port operations and services. The proposed research involves designing and developing the necessary big data infrastructure, which will form a comprehensive solution for (i) collecting and integrating data efficiently from various sources; (ii) enabling stream processing in real time; (iii) storing the data in a fault-tolerant way; and (iv) supporting machine learning and advanced analytical processing of data with the goal of extracting new meaningful insights and supporting decision-making activities for the maritime stakeholders. The proposed platform will be validated via processing and storing real-time data related to ship movements, the environment, and the tracking of cargo at the Port of Limassol.

Required Qualifications: Undergraduate (BSc) and postgraduate (MSc) degrees in Computer Science or related field. The ideal candidate should enjoy working on cutting-edge systems research problems and have good software development skills. Prior research experience or specialization in related topics will be considered an advantage.

Funding: There is possibility of funding through involvement in a Research Program for full-time qualified applicants.

Research Advisors:

Michalis Michaelides, Assistant Professor,   michalis.michaelides@cut.ac.cy

Herodotos Herodotou Assistant Professor, herodotos.herodotou@cut.ac.cy

 

  • One (1) position in the following field: “Smart Resource Management in Distributed Stream Data Processing Systems”

Description: The need to reduce the gap between the generation of data and extraction of insights from these data has led to significant innovations for distributed stream data processing engines (DSPEs). Such systems are driven by a data-centric model that allows for near real-time consumption and analysis of data. However, ensuring good and robust system performance at large scale for streaming applications poses several new challenges, including the distributed management of system resources, the heterogeneity of computing clusters, and the increased complexity of streaming applications. The goal is to develop automated resource management techniques for addressing the aforementioned challenges in order to make DSPEs more robust in their operating characteristics. Specifically, the PhD candidate will design and implement new algorithms and tools for optimizing the allocation of compute resources (e.g., CPU cores, memory) to both the underlying streaming engines and the applications processing the data streams. In addition, the tools will be responsible for monitoring the cluster utilization, finding and eliminating rogue usage, and defining automatic actions in response to anomalies and inefficiencies with the dual goal of maximizing resource utilization and minimizing application delays.

Required Qualifications: Undergraduate (BSc) and postgraduate (MSc) degrees in Computer Science or related field. The ideal candidate should enjoy working on cutting-edge systems research problems and have good software development skills. Prior research experience or specialization in related topics will be considered an advantage.

Funding: There is possibility of funding through involvement in a Research Program for full-time qualified applicants or as teaching assistants.

Research Advisor: Herodotos Herodotou Assistant Professor, herodotos.herodotou@cut.ac.cy

 

  • One (1) post in the following topic: “Development of Optical Fiber Plasmonic Sensors and Nanoantennas Using Femtosecond Laser Pulses”

Required Qualifications:  BSc and/or MSc in Electrical Engineering or Physics, or any other related subject. Strong mathematical background will be considered an advantage.

The PhD will focus on the development of photonic (bio)chemical sensing platforms, using custom sensors developed in-house with a femtosecond laser system. The PhD will focus on tilted fibre Bragg gratings surrounded by nanoscale coatings of metal layers and nanoparticles that will be studied and optimized to exploit the plasmonic enhancement of the sensing transduction mechanisms.

Research Advisors:   Kyriacos Kalli, Professor, kyriacos.kalli@cut.ac.cy

 

  • One (1) post in the following topic:  “Optical Fibre Sensors for Biomedical Applications”

Required Qualifications: BSc and/or MSc in Electrical Engineering or Physics, or any other related subject. Strong mathematical background will be considered an advantage.

Research Advisors:   Kyriacos Kalli, Professor, kyriacos.kalli@cut.ac.cy

 

  • One (1) post in the following topic: “Predictive Learning Algorithms for Distributed Acoustic Sensor (DAS) Networks for Oil and Gas Pipelines”

Description: The goal of this thesis is to develop novel machine learning algorithms, suitable for performing data-driven predictive tasks in the context of Distributed Acoustic Sensor (DAS) networks. DAS environments generate data that entail great deals of epistemic uncertainty, due to several hard to model artifacts, such as skewness, heavy tails, non-stationarity, and measurement noise. These data properties call for the development of deep generative models with novel statistical assumptions, that are not yet reported in the related literature. In addition, the very nature of DAS networks necessitates the development of novel distributed inference algorithms, as well as sensor hardware that effectively facilitates the operation of such algorithms. This thesis will address these challenges in a comprehensive way. We will develop in-house sensor networks for our experimentations, and will leverage state-of-the-art machine learning software, such as TensorFlow. There is also the strong prospect of real-world deployment and validation of our novel solutions, in the context of our existing collaboration with a world leader in Fiber-optic sensing technology. This thesis requires some basic affinity with DAS and statistical modelling.

Research Advisors:

Kyriacos Kalli, Associate Professor, kyriacos.kalli@cut.ac.cy

Sotirios Chatzis, Assistant Professor, sotirios.chatzis@eecei.cut.ac.cy .

 

  • One (1) post in the following topic:  “Night Cooling Systems: Modeling and monitoring systems”

Required Qualifications: BSc and/or MSc in Electrical Engineering or Physics, or any other related subject. Strong mathematical background will be considered an advantage.

Research Advisor:  Paul Christodoulides, Assistant Professor, paul.christodoulides@cut.ac.cy

 

  • One (1) post in the following topic:  Application of Structural Equation Models and Satellite orbits

Required Qualifications: BSc and/or MSc in Electrical Engineering or Physics, or any other related subject. Strong mathematical background will be considered an advantage.

Research Advisor:  Paul Christodoulides, Assistant Professor, paul.christodoulides@cut.ac.cy

  • One (1) position in the following field: “Mathematical modeling and performance analysis of micro-network router architectures and traffic flows”

Description: This research topic concentrates on elaborate detailed mathematical models to capture in detail the behavior of the architecture of pipelined micro-routers utilized in micro-interconnect networks found in today’s multi-core processors and embedded systems. Various state-of-the-art architectures will be considered. The interaction of the underlying router organization will be considered in tandem with various flow-control protocols and routing algorithms, along with numerous traffic flow spatio-temporal behaviors in order to determine throughput and network latency levels that will act as indicators of architectural router performance. Results obtained from software simulations of equivalent architectures will be carried out to confirm the validity and the accuracy of the mathematically-modeled micro-router architectures.

Required Qualifications: Candidates should possess a Bachelor’s Degree and a Master’s-level postgraduate degree from accredited universities in the field of Computer Science, or Electrical Engineering, or Computer Engineering or Mathematics with a preferred specialization in either computer architecture, computer networks, discrete mathematics, statistics, or related.  The candidate should have 2+ years of experience in object-oriented programming and be fluent in C++ programming and/or Python. A strong mathematical background is desired. Excellent command of the English language is a must.

Research Advisors:  

Vassos Soteriou, Associate Professor, vassos.soteriou@cut.ac.cy

Paul Christodoulides, Assistant Professor, paul.christodoulides@cut.ac.cy

 

  • One (1) position in the following field: “New techniques for data storage and archiving of massive and complex amounts of 2D/3D/4D Cultural assets”

Description: Cultural Heritage (CH) is an integral element of Europe and vital for the creation of a common European identity. The rapid growth of technology has led to mass digitization of cultural assets, requiring for their cost–effective preservation, documentation, protection and presentation in online digital libraries. The aim is to shed light, through technological innovation and digital media, on all aspects of cultural heritage, both tangible (books, newspapers, photographs, drawings, manuscripts, costumes, maps, objects, archaeological sites, monuments) and intangible (eg, music, performing arts, folklore, theater), as well as their semantic interrelations, and finally enhancing their added value by reusing them in the fields of education, tourism industry, advertising and art

The proposed research will focus on (a) the study and analysis of massive and complex amounts of multimedia 3D/4D data, (b) study and analysis of data storage and archiving in multimedia digital libraries, (c) the development of innovative methodologies for harvesting of such data sets in digital libraries, taking into account object’s semantic signatures, and finally, (d) the development of innovative methodologies for reuse of such complex structures from digital libraries.

Research Advisor: Marinos Ioannides, Senior Lecturer, marinos.ioannides@cut.ac.cy

 

  • One (1) position in the following field:Holistic Heritage Management”

Description: Heritage Management is a multiparametric field facing nowadays a variety of challenges. The progressive expansion of the term of Cultural Heritage (CH) has led to a type of management of it (CH), which goes beyond the conservation and restoration of cultural assets. A wide spectrum of values, a variety of involved stakeholders, multiple, even conflicting, objectives, are only some of the challenges CH is facing. Even nowadays involved authorities and stakeholders act within their own narrow spectrum without taking into consideration a number of other interrelated parameters; an attitude which not rarely results to fragmented and not so beneficial interventions. The proposed project aims to approach Heritage Management in a holistic way; As a “procedure” of management, starting from the phase of data acquisition, but also as a “result”, leading to concrete actions; As an embracement not only of the lifecycle of the cultural asset, but also of the lifecycle of the human, starting at his early schooling age, since human is the provider but also the user of CH. For the achievement of this goal a continuous shift between different scientific domains, the skilful management of differentiated input and its transformation into new information and knowledge, exploitable by various sectors, becomes crucial. For this reason it is needed: a broad educational background on Arts and Culture, the tools and the methodological thinking of engineering as well as the pedagogical techniques, in order for CH to become an actual “public asset”. Required Qualifications:  A BSc and MSc degree in Architecture, an MSc in the field of Cultural Heritage as well as pedagogical education. Prior research experience or specialization in Cultural Heritage and Education will be considered an advantage.

Research Advisor: Marinos Ioannides, Senior Lecturer, marinos.ioannides@cut.ac.cy

 

  • One (1) position in the following field: “Applying Machine Learning methods in processing Cultural Heritage assets”

Description: Cultural Heritage is the legacy of a nation from previous generations, for which efforts are made maintain their present status but also to safeguard its future existence. Nowadays, the technological outbreak has led to the development of intelligent systems, which can actively contribute in areas like the documentation, preservation and promotion of Cultural Heritage. Machine Learning constitutes an integral part of intelligent systems as it is a category of artificial intelligence, which enables modern computer systems to "learn" to develop and adapt their function upon exposure to new data.

The proposed research will be focused on the development of machine learning methods for their use in cultural applications. As part of the research activities will be the study of existing machine learning methods (supervised, non-supervised, reinforcement) which are currently used for the classification of cultural assets over time.

Required Qualifications: Applicants should have a BSc and an MSc degree in Computer Science, Science of Electrical Engineering or other related field. Previous research experience in the study and the application of machine learning in Cultural Heritage sector will be considered an asset.

Research Advisor: Marinos Ioannides, Senior Lecturer, marinos.ioannides@cut.ac.cy

 

Information:

Department Secretary Tel: 25002533, Fax: 25002635