Research at Lancaster University
Master of Philosophy in Computer Science - 2012/16
The project seeks to enact Anglian Water's vision of a smart wastewater network by developing an integrative communication platform based on recent advances in distributed computing techniques. The envisaged platform will combine sensor input (wet well water levels, pump status, inflows), actuator control (pumps, vanes), user interfaces, and artificial intelligence algorithms into a cohesive software system that is easy to deploy, maintain and reconfigure without the need for specialist expertise in low-level WSAN and networking technologies.
Publication
Absract
This thesis will provide an insight on how weather factors such as temperature, humidity and air pressure affects radio links and use this body of knowledge to better understand how to mitigate unnecessary radio switching to take place and then use this knowledge to suggest ways of developing a Link Quality Estimator that utilize online weather data to be able to conduct smart link switching. In the context of this thesis, we focus on the case study of a water utility company as these entities are under increasing economic and environmental pressures to optimise their infrastructure, in order to save energy, mitigate extreme weather events, and prevent water pollution. One promising approach consists in using smart systems. However, a smart infrastructure requires reliable communication links which are difficult to provide. In particular, communication links that are distributed and geographically located in rural areas are highly affected by changing weather conditions, hence efficient control of these distributed hosts requires robust communications. Multiple communication transceivers are used to mitigate this issue and to enable nodes to switch to reliable links. Short-term link quality estimators are used to decide which link to use which often leads to the situation where a link switch is initiated which does not prove helpful in the long term. It is not beneficial to switch a link and associated routing for only a brief duration hence we conduct test bed experiments to better understand the relationships between the radio links and weather factors and use this body of data to devise a LQE that can use this data and then make a smart choice based on this data which reduces switching costs.
Cite this
Resilient control in long-range sensor and actuator networks. Linares, Jose. Lancaster University, 2017. 208 p.
Poster Presentation
Linares J, Taiani F, 2012 How would you distribute intelligence to wastewater pumps?, 36th intelligent sensing conference, Knowledge Transfer Network, London.
Unpublished and works in progress
Linares J, Taiani F, Singerton D, 2012, Building a distributed smart system for waste water networks, workshop paper -Download
Linares J,Taiani F, Singerton D, 2012, Fuzzy Waters - Towards a smart system middleware for wastewater networks, conference paper -Download
Linares J, Roedig U, Singerton D, 2014, An Integration Platform for Smart Waste-water infrastructures,conference paper -Download
Linares J, Roedig U, 2015, Sweeping Unnecessary Switching; Using Weather Data to improve Resilience in Large Scale Wireless Sensor Networks, conference paper -Download