Citizen Science in Acoustics dataset
收藏资源简介:
## Overview This repository contains a curated collection of six datasets composed of smartphone-based acoustic measurements derived from the NoiseCapture and OpeNoise apps. These datasets support the analyses, experiments, and conclusions of a PhD thesis conducted at the University of São Paulo. The datasets are shared in alignment with the OPEN and FAIR principles (Findable, Accessible, Interoperable, Reusable). All participant data have been anonymized, and NoiseCapture measurements can be identified using UUIDs. ## Datasets Included Below is a summary of the six datasets included in this collection: **Exp2_urbanparks** Description: Relates to Exp 2 described in the Thesis. Two smartphones (iPhone 7 Plus and a Samsung Galaxy S7) were used to conduct 5 to 10-minute measurements in open areas of São Paulo, alongside a reference Class 1 sound level meter (SLM). **Exp3_noiseparty_usp** Description: Relates to Exp 3 described in the Thesis. Two smartphones (iPhone 7 Plus and a Samsung Galaxy S7) were used to conduct 2 to 5 minute walking measurements in University of São Paulo campus, alongside a reference Class 1 SLM. It was also linked to a NoiseParty code, which generated a noise map available at: https://noise-planet.org/map_noisecapture/noisecapture_party.html#14/-23.5630/-46.7299/SAOPAULO23 **Exp4_participative2023_lab_usp** Description: Relates to Exp 4 described in the Thesis. Approximately 170 Participants attended a studio and performed 30-second measurements at the same position as a Class 1 SLM (sequential measurements), using a pink noise as the acoustic source. **Exp5_participative2023_field_usp** Description: Relates to Exp 5 described in the Thesis. Approximately 170 Participants passed through four fixed monitoring stations (equipped with SLMs) across the University of São Paulo campus. It took place during a 2023 course activity at the Faculty of Architecture and Urbanism. **Exp6_participative2024_field_usp** Description: Relates to Exp 6 described in the Thesis. Approximately 122 Participants passed through four fixed monitoring stations (2 of them equipped with references sensors) across the University of São Paulo campus. It took place during a 2024 course activity at the Faculty of Architecture and Urbanism. **Exp7_citizen_science** Description: Relates to experiment described in Chapter 5 and 6 of the Thesis. After calibration procedures (Experiments 4, 5, and 6), the participants conducted 1min measurements at various locations across the city of São Paulo. ## File Structure The dataset is organized into three main components: mobile_sensors/ Data collected from smartphones fix_sensors/ Data from reference sensors (e.g., Sound Level Meters) colocated with some mobile sensors input.xlsx Main input file with structured information about mobile and fixed sensors outputs/ Processed outputs (available for some datasets)├── frequency_domain_by_sensor/ Frequency-domain graphs and numerical data per measurement├── time_series_by_sensor/ Time-domain graphs and numerical data per measurement└── reports/ PDF reports summarizing each measurement match_sensors_output.xlsx Table mapping spatial and temporal correspondence between mobile and fixed sensors ## File Naming Convention Files follow the structure: device-measurement_location.extension Rules: "_" separates device and measurement "-" separates measurement and location Device is the identifier of the sensor; Measurement it refers to all measurements collected by a given device at a specific location.Location is the place where the device performs the measurements. Examples: LARSON-m1_p1.xlsx In this example, the mobile sensors took only one measurement in each location, then the identifier device-measurement is the unique ID of a sensor. For instance, the identifier G1-Aluno1_p1.txt and G2-Aluno5_p3.geojson, p1 and p3 are the locations. ## Data processing The outputs included in this repository were derived using a dedicated processing pipeline available at: https://github.com/GabrielaNara/smartphone_sound_calibration Data processing steps include: * Read files from SLM, OpeNoise and NoiseCapture apps * Temporal and spatial matching between mobile and fix sensors measurements * Provide graphs and numerical information ## Caution Some files derived from Exp 5 and 7 were set as the calibration value obtained in Exp4. This value is specified in the input.xlsx file as "pre-calibration" and must be subtracted from the measurements. Data collected from iPhones using the OpeNoise app must be configured with a pre-calibration value of 24 dBA. Some of these devices did not apply this pre-calibration and/or failed to capture frequency content, which may result in distortion in certain frequency graphs.. ## Contact For questions or clarifications, please contact: Nara Perixoto / nara.gabriela.mesquita@gmail.com



