Issue |
MATEC Web Conf.
Volume 125, 2017
21st International Conference on Circuits, Systems, Communications and Computers (CSCC 2017)
|
|
---|---|---|
Article Number | 05013 | |
Number of page(s) | 8 | |
Section | Signal Processing | |
DOI | https://doi.org/10.1051/matecconf/201712505013 | |
Published online | 04 October 2017 |
Development of Seasonal ARIMA Models for Traffic Noise Forecasting
1 Department of Civil Engineering, University of Salerno, via Giovanni Paolo II 132, Fisciano, Italy
2 Department of Electrical Engineering and Computer Science, Hellenic Naval Academy, Piraeus, Greece
3 Department of Electrical Engineering, Piraeus University of Applied Sciences, Athens, Greece
* Corresponding author: cguarnaccia@unisa.it
In this paper, a time series analysis approach is adopted to monitor and predict a traffic noise levels dataset, measured in a site of Messina, Italy. In general, acoustical noise shows a high prediction complexity, since its slope is strongly related to the variability of the sources and to intrinsic randomness. In the analysed site the predominant source is road traffic, that has a periodic and non-stationary behaviour. The study of the time evolution of this hazardous agent is very useful to assess the impact to human health and activities. The time series models adopted in this paper are of the stochastic seasonal ARIMA class; these types of model are based on the strong periodicity registered in the acoustical equivalent levels. The observed periodicity is related to the highly variability of urban traffic in the different days of the week. Three different seasonal ARIMA models are proposed and calibrated on a rich dataset of 800 sound level measurements. The predictive capabilities of these techniques are encouraging. The implemented models show a good forecasting performances in terms of low residuals, i.e. difference between observed and estimated noise values. The residuals are analysed by means of statistical indexes, plots and tests.
© The Authors, published by EDP Sciences, 2017
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.