UoM–IITKgp PhD projects
To apply for one of these projects, please contact the lead academics from both institutions to discuss the opportunity.
Project Title: Guaranteeing quality of crowdsourcing public transportation run-time information
Crowdsourcing public transportation run-time information is a promising approach in the absence of authoritatively provided run-time information. However, crowdsourcing suffers from data quality issues (inhomogeneous coverage). This project will investigate the impact of inhomogeneous coverage, and search for countermeasures such as learning from history or incentivizing in real-time, in order to guarantee certain levels of information quality.
Project Title: Fingerprinting public transportation services
Soumya K. Gosh, Niloy Ganguly, Bhargab Maitra (email@example.com)
Fingerprinting of individuals can be translated to fingerprinting of bus lines (reliability) or fingerprinting of individual bus drivers (performance). The derived information should be applied in visual analytics for various uses, such as the citizens searching for PT performance, or the operators searching for performance at particular times or by particular drivers
Project Title: Extracting the semantics of places from NL descriptions
Sudeshna Sarkar (firstname.lastname@example.org)
NL place descriptions contain contextualised human knowledge of places, where the context is critical for interpreting the extractable terms (place names as well as qualitative spatial prepositions). The project will investigate which elements of context can be extracted, how these elements can be represented in a database, and how these representations can be applied in querying the database. The project will also implement and test the methods required.
Project Title: Coupled socio-economic, biophysical scheme for climate induced water resources planning and management.
Dr Renji Remesan
River basins in both developed and developing world are facing a number of demand and sustainability challenges in the context of climate change. Potential impacts of global climate change on river basins would be severe unless proper proactive adaptation plans are put in place. In water resources planning and management, agent-based modelling can be applied to explore, simulate, and predict the performance of infrastructure design and policy decisions as they are influenced by human decision-making, behaviors, and adaptations. All next generation river basin modelling process should treat river basins as coupled human and natural systems; thus, should incorporate behavioral models (eg: agent based models) in conjunction for simulation of the dynamics of human systems (eg: irrigation, pumping activates, farmers and industry dynamics, economic and other policies etc) and how peoples behaviors respond to environmental and hydrological changes. This PhD proposal aims to couple an agent-based model and biophysical hydrological model (eg: SWAT or something similar). Such an integrated modelling framework will be applied in two urban river basins of both India and Australia to investigate water resources planning problems encompassed by both climate change and anthropogenic behaviors.
Project Title: Software Defined Networks for Internet of Things
Prof Marimuthu Palaniswamy, Dept of Electrical and Electronic Engineering
Prof Sudip Misra
SDN and IoT are presently two growing verticals. However, there are lots of use cases, which can benefit from the application of SDN to IoT. There are patches of work worldwide on this issue. The nature of challenges in the integration of these two technologies brings in lots of research opportunities.
Labs/softwares. We have some very basic infrastructure on IoT. There is lot of infrastructure requirement. Details need to be worked out.
Project Title: Real-time Internet of Things (IoT) for Smart Transportation Applications
IIT Kharagpur Staff
The real-time IoT applications have stringent delay requirements when implemented over distributed sensing and communication networks. Their finite-time performance matters much more than asymptotic results in the literature. A good illustrative example is smart traffic control, where sensor-laden vehicles pass through intersections by communicating and remaining at a safe distance from each other, rather than grinding to a halt at traffic lights. Thanks to increased investment in smart city infrastructure and projected penetration of autonomous vehicles, smart intersections are expected to replace traditional traffic lights and become prevalent in the immediate future. Safe and efficient operation of such systems requires control actions to be taken by each system agent, i.e., automated vehicles and road side units, within fraction of a second under local communication and computing limitations. The IoT agents in this scenario clearly must attain global objectives of safety and efficiency in real time.
The overarching aim of the project is to develop new practical algorithms for real-time IoT applications and to investigate novel methods for describing their performance in a finite time. This project aims to investigate fundamental performance guarantees of real-time IoT systems using methodology from networking, optimisation, game theory, and information theory. The project addresses the above mentioned two challenges through (i) the derivation of performance bounds of finite-time algorithms under communication and computing limitations; (ii) the development and analysis of novel real-time algorithms within these bounds for efficient allocation of limited resources considering incentives and minimum performance constraints.
Project Title: Hydrological processes in a changing climate
IIT Kharagpur Supervisor
Impacts of climate change are manifest through increases in temperature, sea level and extreme events. One of the most important impacts, from a human perspective, is the future redistribution and availability of water resources across the globe. How different hydrologic processes will respond is crucial for an improved understanding of climate change impacts on the hydrologic cycle and improving future water resources management. This research proposal is aimed to consider the key hydrologic variables—precipitation, evapotranspiration, runoff and watershed characteristics and to explore the factors behind their likely variation in a changing climate. The key research question is to identify likely changes in rainfall-runoff processes due to a change in climate. Once identified, a model of these processes will be either developed, or modified, to represent those hydrologic processes under changing climatic conditions. With this improved understanding, an assessment of hydrologic phenomena—mean, variability, floods and droughts, will be carried out for future climatic conditions. This assessment will facilitate identification of vulnerable regions that may require attention in terms of strategic planning for a better water resources availability future.