An Efficient Feature Matrix for Urban Noise Annoyance Measurement

Page view(s)
Checked on Jul 17, 2023
An Efficient Feature Matrix for Urban Noise Annoyance Measurement
An Efficient Feature Matrix for Urban Noise Annoyance Measurement
Journal Title:
IEEE World Forum on Internet of Things 2018
Publication URL:
Publication Date:
08 February 2018
Traditionally the noise measurement and monitoring are based on the average A-weighting sound pressure level. An example that music with high loudness level does not disturb us shows using A-weighting sound pressure level only is insufficient. It is desirable to study the noise annoyance measuring method for urban environmental noise monitor and control. In this paper, we propose an efficient noise annoyance measurement scheme which uses less noise features with two dimensions. Supervised with more accurate annoyance score obtained through a novel subjective listening test, the proposed scheme efficiently reduces the dimension of noise annoyance matrix by using a forward iterative method. We also analyze the influence of different noise features to the annoyance level. The analysis shows that the $90$ percentile instantaneous loudness is the most significant feature and tonality is the second significant feature for annoyance measurement. The efficiency of proposed annoyance matrices finally is verified using several linear regression methods.
License type:
Funding Info:
This material is based on research/work supported by the Singapore Ministry of National Development and National Research Foundation under No.L2NICCFP1-2013-7
© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Files uploaded:

File Size Format Action
wfiot18-sy.pdf 128.37 KB PDF Open