News and current affairs from the network of La Salle Universities in Europe. Exchange of best practices in higher education.

15 July 2015 | Posted by Editorial Team La Salle Universities

La Salle-URL develops an algorithm that automatically detects if noises come from traffic

The Research Group in Media Technology of La Salle Campus Barcelona-URL (GTM) participates in the European project Dynamap, which focuses on the development of networks of low-cost sensors to create noise maps in real time. The objectives of the project, funded by the European Commission under the LIFE + 2013 Program, are to facilitate and reduce the cost of regularly updating the noise maps, as required by the European Directive 2002/49 /EC on environmental noise. To do this, it develops a dynamic system generating noise maps able to detect and display in real time the noise impact of road infrastructure. In addition, the project also aims to automate the process of generating the noise maps using the information obtained through a network of monitoring along with the development of low-cost sensors to gather necessary information to update the noise maps in real time. This automation process will allow a significant reduction in the resources needed to update the noise maps and provide this information to a general public, assessing the possibility of improving the system for providing information on various types of additional environmental data. The GTM La Salle-URL leads the development of the Anomalous Event Detection Algorithm (ANED) software. This part of the project focuses on developing algorithms for detecting abnormal noises, in order to avoid noises that may come from other sound sources, which are not caused by traffic, alter the levels detected by the sensors and cause an erroneous representation of the Dynamic map, thus increasing the reliability of the process of real-time monitoring. If it is possible to detect the source of the noise, it will be easier to detect areas or cities with heavy traffic flow and provide important information for the Smart City. This project has been led by senior researchers of the group, Dr. Xavier Sevillano, Dr. Joan Claudi Socoró and Dr. Francesc Alias, and also counted with the participation of students such as Gerard Ribera and Xavier Albiol. This is the fourth European research project related to acoustics and signal processing in this group of La Salle-URL, which has positioned itself as one of the most active and a leader in acoustics and sound and voice processing in Catalonia. The project consortium is led by  Anas S.p.A, a company of the Italian government that is responsible for the construction and maintenance of roads and Italian motorways, and includes companies, universities and research centres in Germany, Italy and Spain. The project, lasting five years (2014-2018), in a first phase will develop low-cost sensors and tools to manage, process and generate noise maps in real time. In addition, the systems will also be monitored for at least a year to check their operation in critical situations, problems and failures that occur during the trial will be analyzed and a guide for the design and implementation of noise maps in real time will be generated. Furthermore, the progress of this project will be tested on two different monitoring systems noise, installed in the cities of Milan and Rome. The first will cover a significant portion of the urban centre of the capital and the second will be located along a main street in Rome. The initial phase of the project has been completed, in which a review of the state of the art on the recognition of sources of noise and elimination of anomalous events has been conducted. In the 22nd International Congress on Sound and Vibration (ICSV), to be held next July in Florence, the first work results obtained by researchers at the GTM will be presented.

Share