Sensor networks for monitoring the environment in Smart Cities
Mentor: Manuel Rosa Zurera
Phone: (+34) 918856695
University: Universidad de Alcalá
Partner Host Institution:
Keywords: Machine learning, array processing, acoustic scene classification, non-destructive testing, distributed sensor networks, data fusion

Sensor networks for monitoring the environment in Smart Cities

Smart Cities are living environments where sensors are used with different purposes: security, traffic control, efficient energy consumption, smart provision of services, etc. In the near future, these sensors will be connected to the Internet, and interconnected among them, creating heterogeneous Smart Sensor Networks. The aim of these networks is to monitor the environment, to collect information, and to use this information to plan the provisión of services, such as security, cleaning services, traffic arrangement, and to guarantee quality of living, for example, warning for sound pollution or environmental noise, air pollution, etc, There are many problems to be solved in heterogenous sensor networks, such as heterogeneous information fusion, automatic knowledge of the sensor network geometry to study the coverage areas taking into consideration geographical data, communication links, etc. The implementation of array processing algorithms to determine the source of information will be also studied. This research line aims at: 1) Studying different alternatives to create heterogeneous smart sensor networks; 2) Developing signal processing algorithms for data fusion, coverage estimation, information source localisation and tracking; 3) Using the obtained information to improve services provision. One of the main objectives is to improve the quality of life in our cities, and to optimise energy consumption when providing these services.

Departament: Signal Theory and Communications
Research Group: Sound Signal Processing / Applied Signal Processing
More Information:
Relevants projects on the area: Smart sound processing for the digital living: cognitive listening (SSPRESING-COLIST). The sub-project "Cognitive Listening", jointly carried out by researchers of the Applied Signal Processing Group at the University of Alcala (UAH) and professors from the University of Las Palmas de Gran Canaria (ULPGC), is focussed on the analysis of acoustic scenes, both in close and open scenarios. This project tries to put a set of localization, separation and classification techniques together.
Relevants publications on the area: 1.- Sánchez-Hevia, H. A., Ayllón, D., Gil-Pita, R., & Rosa-Zurera, M. “Maximum likelihood decision fusion for weapon classification in wireless acoustic sensor networks”. IEEE/ACM transactions on audio, speech, and language processing, Vol. 25, No. 6, pp. 1172-1182, 2017.
2.- Ayllón Alvarea, David; Rosa-Zurera, Manuel; Utrilla Manso, Manuel; Gil-Pita, Roberto; "An evolutionary algorithm to optimize the microphone array configuration for speech acquisition in vehicles" (ISSN:0952-1976). Engineering Applications of Artificial Intelligence. 2014, vol 34, p. 37-44