The purpose of this repository is to provide access to the code used in my MSc thesis "Applicability of Fully Mobile Measurements for Noise Maps" (included as pdf). The code required to reproduce the methodology of the study is included in steps in the root folder. Note that it cannot be run without the relevant data (may be made available upon request), and that it creates several new files the folder locations of which need to be created/adjusted when running locally.
It is recommended to use ipynb provided md section headers for navigation in method notebooks.
- Written by: Floris Copraij
- Thesis in cooperation with: Max v.d. Wielen
- 1st supervisor: Jules Kerckhoffs
- 2nd supervisor: Marc Kreveld
Noise maps provide insight into the distribution of environmental noise throughout an area. This makes them an essential tool in communication regarding- and planning measures against- noise pollution. Currently, most noise maps that are produced are based on mathematical models which create detailed analysis but are unable to capture all of the complexity inherent to noise. This study evaluates the feasibility of a hybrid approach which allows modelled noise maps to be extended with simple to collect car-based mobile noise measurements.
Anoise measurement setup was mounted to the roof of a car and measurements were collected in three different areas. Recording the noise measurements from a moving vehicle included interfering noise from both the collection vehicle and from wind hitting the microphone. To address this noise reduction, regression adjustments and data selection approaches were tested. The resulting models suffered from local inaccuracies and smaller value ranges when compared to reference models. Despite this, the models created were able to capture both general trends in the data and particularities of the areas measured, showing the potential of this hybrid approach given a larger collection of noise samples and more detailed model input.