Title :
Artificial neural network estimation of vehicle audio signal in high way
Author_Institution :
Coll. of Eng., Bohai Univ., Jinzhou, China
Abstract :
In this paper, during traffic flow information signal extraction and analysis of the sound of cars on the freeway, the linear transfer function of three-layer artificial neural network was used for the hidden layer neurons. Appropriate AR model order selection can distinguish different models in order to find the feature vector. Experimental results show that the estimation AR model parameters and power spectrum of freeway vehicle audio signal using the neural network method is effective.
Keywords :
audio signal processing; automobiles; feature extraction; neural nets; parameter estimation; traffic engineering computing; AR model order selection; AR model parameter estimation; car sound analysis; feature vector; freeway vehicle audio signal; hidden layer neurons; high way; linear transfer function; power spectrum; three-layer artificial neural network estimation; traffic flow information signal extraction; Artificial neural networks; Equations; Mathematical model; Road transportation; Spectral analysis; Transfer functions; Vehicles; AR model; Artificial Neural Network; feature vector; high way; vehicle audio signal;
Conference_Titel :
Chinese Automation Congress (CAC), 2013
Conference_Location :
Changsha
Print_ISBN :
978-1-4799-0332-0
DOI :
10.1109/CAC.2013.6775857