Joint Doctoral Programme in Interactive and Cognitive Environments

The Joint Doctoral Programme in Interactive and Cognitive Environments (JD ICE) is a PhD programme organised by the Centre for Intelligent Sensing at Queen Mary University of London and the Department of Electrical, Electronic, Telecommunications Engineering and Naval Architecture at University of Genoa.

JD ICE provides PhD training in the fields of multimedia signal processing, data fusion, computer vision, cognitive dynamic systems and machine learning, and offers advanced research projects designed for ambitious students willing to become word-leading researchers. PhD students will benefit from a joint-supervision model and will spend part of their PhD programme in the London, UK and part in Genoa, Italy.



PhD projects

Machine learning techniques for cognitive dynamic systems: application of the Bayesian approach to radio and video domains
The proposed PhD involves the study of machine learning techniques with a Bayesian approach: the goal of the PhD is to develop novel learning techniques and apply them to possibly two different domains. On one side, bio-inspired learning approaches can be applied to cognitive radio networks thus allowing adaptive methods for analysis and decision to evolve under specific environmental operating conditions. Focus will be on Bayesian methods and game theory approach for modelling dynamic strategies for opportunistic spectrum access. Moreover, a Bayesian approach can be applied for learning from video data: signals acquired from both static surveillance cameras and egocentric vision sensors can be analysed for characterizing behaviours of both single users and complex active entities (e.g. crowds).
Activities will be oriented to address the use of developed methods within bio-inspired cognitive artificial systems for physical security and telecommunications.

PhD Student
Supervisors
Muhammad Irfan Muhammad Irfan

Muhammad did his Bachelor in Information Technology in 2006 from N.W.F.P Agriculture University Peshawar, Pakistan, and Masters of Sciences in Computer Engineering in 2009 from COMSATS Institute of Information Technology, Pakistan. He worked as lecturer in Afghanistan for two years and he worked as a lecturer at the state University of Swat, Pakistan between 2011 and 2015. His current research interests are machine learning, computer vision and crowd analysis.
Lucio Marcenaro
Lucio Marcenaro
Laurissa Tokarchuk
Laurissa Tokarchuk
Development of an m-learning system based on advanced Human-Computer Interaction
The goal of the PhD project is to design and develop an e-learning application for mobile devices. The system will be able to sense the environment/user (in particular emotions), assess the performance, adapt to the user and provide feedback and coaching. This requires studying and developing artificial intelligence algorithms such as SVM, k-NN, NNs, etc. Human-Computer Interaction will involve 3D augmented reality and advanced video gaming (in particular serious games, for instructional purposes) features.

PhD Student
Supervisors
Bajaj Nikesh Nikesh Bajaj

Nikesh did his Masters in Communication and Information Systems (Signal Processing stream) from Aligarh Muslim University, India. With five years of teaching experience, he worked in Signal Processing, Image and Audio Processing and Machine Leaning. His current research interests focus on Brain Signal Processing with Machine Learning for Serious Games.
Alessandro De Gloria
Alessandro De Gloria
Jesús R. Carrión
Jesús R. Carrión
Personal data, thinking inside the box
The research is in the general area of Personal Data & Human-Data Interaction. There is a critical need to provide technologies that enable alternative practices, so that individuals can participate in the collection, management and consumption of their personal data. In this studentship, we will investigate the feasibility of having a Databox, a personal networked device (and associated apps/services) that collates and mediates access to personal data, allowing us to recover control of our online lives. We hope the Databox is a first step to re-balancing power between us, the data subjects, and the corporations that collect and use our data. The Databox can act as a hub for individuals' mobility and physical activity data, social media data, third party tracking, and ambient data from the environment.

PhD Student
Supervisors
Yousef Amar Yousef Amar

Yousef has a background in Electronic Engineering and Systems Engineering. He gained experience in computer vision, machine learning and the real-time processing of large amounts of data through building drone sensor systems. His current research interests focus on personal data, networks and privacy.
Hamed Haddadi
Hamed Haddadi
Fabio Lavagetto
Fabio Lavagetto
Eco-friendly naturalistic vehicular sensing and driving behavior profiling
This project aims to analyse naturalistic vehicular driving behaviour from data extracted from mobile sensors, fused with in-car sensor data accessed using an On-Board Diagnostics-II (OBD-II) system and a Bluetooth/Wi-Fi/wired adapter. This will be used to profile and evaluate the driving behaviour in real time, taking into account eco-driving concerns and an awareness of the physical environment context. An additional focus is in applying the concept of serious games to help incentivise more eco driving through sharing driving metrics and comparing oneself with friends and other peers.

PhD Student
Supervisors
Rana Massoud Rana Massoud

Rana received her BSC in Computer Sciences in 2013 and MS of Computer Sciences in Software Eng in 2015 from Lebanese University-Faculty of Sciences, Lebanon. She has experience on VOIP, Windows Phone and the .Net Framework having worked for more than one year at Telepaty Holding, Tripoli, Lebanon. Her current research interests are naturalistic vehicular data analysis, data science, On-Board Diagnostics protocol, automotive sensing data, eco-driving, driving behavior profiling, and serious games.
Francesco Bellotti
Francesco Bellotti
Stefan Poslad
Stefan Poslad
Cognitive radio for PHY-layer security against jamming attacks
This PhD project focuses on algorithms and methods based on compressed-sensing theory as enabler for cognitive radio terminals for opportunistic spectrum usage. Since the proposed research addresses the problem of "non-stationary spectrum", time-frequency analysis in a compressed-sensing framework is investigated through the S-transform in order to perform signal processing in both the domains. The subsequent objective is to exploit recent advances in Machine Learning such as Generative Adversarial Networks to detect potential malicious signals and predict the strategy used by the jammer to disrupt legitimate communications.
The applications of the proposed research are the "nSHIELD project" based on a hardware platform with software-defined devices to optimise external attack robustness and "TV White Space (TVWS)" composed of unused channels in the digital terrestrial TV band which can be occupied by secondary users.

PhD Student
Supervisors
Andrea Toma Andrea Toma

Andrea obtained his Masters in Communication Engineering in 2014 from University of Salento (Italy). His main research focused on the localisation problem in Wireless Sensor Networks where he gained experience at Austrian Institute of Technology (Vienna) and University of Salento. He is currently a PhD candidate and involved in the JD ICE programme.
Carlo Regazzoni
Carlo Regazzoni
Frank Gao
Frank Gao
Ego things: networks of self-aware intelligent objects
This PhD project focuses on probabilistic signal-processing architectures and techniques to represent and capture generality, self-awareness, adaptability, flexibility and reconfigurability of an Internet of Things system through a learning experience in order to make the “Things” adapting their behaviors in the non-stationary environments where they operate. Specific focus will be given to the communication and cooperation issues between the entities (ego things) in the network. An in-depth analysis of the Cognitive Dynamic System framework associated with the 'Ego-Things' will be performed to understand how Dynamic Bayesian models and machine learning techniques can be employed.

PhD Student
Supervisors
Divya Thekke Kanapram Divya Thekke Kanapram

Divya did her BSC in Electronics and Communication Engineering at University College of Engineering, Trivandrum, India. After a period in Industry, she obtained her MS in Multimedia Signal Processing and Telecommunication Networks in 2014 from University of Genoa, Italy. Her research interests include machine learning and Internet of Things for interactive and cognitive environments, and networks of 'Ego things'.
Mario Marchese
Mario Marchese
Eliane L. Bodanese
Eliane L. Bodanese
Incremental learning techniques for proactive anti-jamming in Cognitive and Software Radio
This PhD project develops the understanding of jamming and anti-jamming attacks in Cognitive Radio Networks. The study involves the investigation of jamming and anti-jamming techniques in the dynamic spectrum of cognitive radio with the objective of detecting and tracking jammer present in a cognitive radio spectrum along with legitimate users, using Dynamic Bayesian Networks. The dynamic spectrum modeling is developed using Bayesian filtering, Switching models, Particle filter, Kalman filter and Machine learning techniques as a tool, can switch a user transmission in the case of jammer attacks, and attempts to minimize jamming attacks in cognitive radio.

PhD Student
Supervisors
Muhammad Farrukh Shahid Muhammad Farrukh Shahid

Muhammad did his BSC in Telecommunication Eng. (2010) and MEng in Communication Systems and Networks (2013) at Mehran Uni. of Eng. and Technology, Pakistan. He worked as a lecturer in the Dep. of Electrical Eng. at COMSATS Inst. of Inf. Technology (2013-2015) and in the College of Eng. at the Karachi Inst. of Economics and Technology (2015-2017), Pakistan. His research interests focus on dynamic spectrum modeling for cognitive radio, machine learning techniques for cognitive environment and adaptive modulation techniques for wireless communication.
Carlo Regazzoni
Carlo Regazzoni
Frank Gao
Frank Gao
Cognitive environments for telerobotics
Remotely controlled robots heavily rely on human supervisory control approaches. The quality of performed telerobotic tasks depends on the human-operator's experience, cognitive state and availability of required feedback information to support the decisions. Colour, depth cameras and range sensors are commonly used to create the visual and spatial representation of the remote environment for the human-operator. Additionally, in some applications force and tactile sensing data is required to provide the human-operator with haptic feedback from the robot's end-effector. We propose to integrate visual, spatial and haptic feedback with the help of interactive virtual reality environment to improve the efficiency of remote robot control. However, a direct merging of the sensory modalities may increase cognitive loads of the human-operator and therefore affecting the performance. We propose an intelligent feedback adaptation system which will automatically adjust the representation of the feedback information to provide better situation awareness (better field of view for visual feedback, relevant tactile information, and vision-based spatial maps). Such system will need to learn from previous experience (recorded sensing and actuation data from the robot and the operator) to form an internal dynamic task-model which will also be updated on-fly based on the actual task-specific operation parameters. The developed system will be composed of a mobile robotic manipulation system equipped with RGBD-cameras, a VR interface and an application for merging visual, spatial and haptic feedback from the robot and human body movement tracker to control the robotic system and algorithmic component implementing cognitive adaptation of the VR-interface with respect to actual user and task state. This research will use methods from robotics, computer vision, haptic and visual rendering for VR and ergonomics.

PhD Student
Supervisors
Bukeikhan Omarali Bukeikhan Omarali

Bukeikhan received his BSc in Aerospace Engineering from TU Delft, Netherlands and MSc in Mechanical Engineering from Nazarbayev University, Kazakhstan. He has worked for several years at the Nazarbayev University as a lab/research assistant at the Robotics department as well as an instructor for an elective CAD/CAM course. His research interests lies in robot teleoperation, predictive control and virtual environments.
Ildar Farkhatdinov
Ildar Farkhatdinov
Maurizio Valle
Maurizio Valle
Learning self-awareness models for physical layer security in Cognitive Radios
This PhD project focuses on learning dynamic environmental models in Cognitive Radios. Different machine learning algorithms and techniques will be exploited to analyse the dynamic behaviour of user and jammer within a cognitive radio network, and to optimise the cognitive radio performance against jammer attacks. The user and jammer activities will be characterised by their statistical properties through Bayesian models, and the current state of the observed entities will be estimated and their actions in the near future will be predicted through probabilistic graphical models. The project will demonstrate how the interaction between user and jammer in cognitive radios can be modelled by coupling multiple Dynamic Bayesian networks in a way that users can successfully avoid the jammer.

PhD Student
Supervisors
Ali Krayani Ali Krayani

Ali received his Bachelor’s and Master’s degree in Telecommunication Engineering from Politecnico di Torino (Italy) in 2014 and from University of Florence (Italy) in 2017, respectively. He worked as a Software Engineer for several companies and he is currently a JD ICE PhD candidate. His current research interests include Cognitive Radios, self-awareness, Dynamic Bayesian Networks and Machine Learning.
Carlo Regazzoni
Carlo Regazzoni
Frank Gao
Frank Gao
Project 1 confirmed and to be assigned
PhD Student
Supervisors




Project 2 confirmed and to be assigned
PhD Student
Supervisors




Project 3 confirmed and to be assigned
PhD Student
Supervisors




Project 4 confirmed and to be assigned
PhD Student
Supervisors