TY - GEN
T1 - Urban noise permanent monitoring and pattern recognition
AU - Fernandez, Luis Pastor Sanchez
AU - Ruiz, Arturo Rojo
AU - De Jesús Medel Juárez, José
PY - 2010
Y1 - 2010
N2 - The research is based on 10-node autonomous, distributed wireless monitoring system placed in various locations throughout Historical Center of Mexico City to take permanent measurements of the noise levels and stream the data back to the main monitoring station every five minutes and the measurements of signals of the noise produced during the takeoff in a location of the International Airport of Mexico City. As first stage, this paper presents a novel computational model that allows aircrafts identification based on generated noise. This model also helps to foresee potential effects to health caused by this kind of noise during the aircraft takeoff, which is when the greatest impact are generated due to the sonorous levels that are reached. The data acquisition is made at 25 KHz at 24 bits. The identification of the aircraft noise is done through two parallel neural networks combined with a weighted addition. In order to generate the inputs to the neural networks, parameters that were obtained from the auto-regressive model and the 1/12 octave analysis are used. This system has 13 categories of aircrafts and it has an identification level of 80% in real environments.
AB - The research is based on 10-node autonomous, distributed wireless monitoring system placed in various locations throughout Historical Center of Mexico City to take permanent measurements of the noise levels and stream the data back to the main monitoring station every five minutes and the measurements of signals of the noise produced during the takeoff in a location of the International Airport of Mexico City. As first stage, this paper presents a novel computational model that allows aircrafts identification based on generated noise. This model also helps to foresee potential effects to health caused by this kind of noise during the aircraft takeoff, which is when the greatest impact are generated due to the sonorous levels that are reached. The data acquisition is made at 25 KHz at 24 bits. The identification of the aircraft noise is done through two parallel neural networks combined with a weighted addition. In order to generate the inputs to the neural networks, parameters that were obtained from the auto-regressive model and the 1/12 octave analysis are used. This system has 13 categories of aircrafts and it has an identification level of 80% in real environments.
KW - Aircraft
KW - Monitoring
KW - Recognition
KW - Urban noise
UR - http://www.scopus.com/inward/record.url?scp=79959889481&partnerID=8YFLogxK
M3 - Contribución a la conferencia
AN - SCOPUS:79959889481
SN - 9789604742509
T3 - European Conference of Systems, ECS'10, European Conference of Circuits Technology and Devices, ECCTD'10, European Conference of Communications, ECCOM'10, ECCS'10
SP - 143
EP - 148
BT - European Conference of Systems, ECS'10, European Conference of Circuits Technology and Devices, ECCTD'10, European Conference of Communications, ECCOM'10, ECCS'10
T2 - European Conference of Systems, ECS'10, European Conference of Circuits Technology and Devices, ECCTD'10, European Conference of Communications, ECCOM'10, European Conference of Computer Science, ECCS'10
Y2 - 30 November 2010 through 2 December 2010
ER -