Cintal logo

K2D - Knowledge and Data from the DeepSpace, D4.2 - Exploratory Acoustic Data Analysis


R. Duarte, rjduarte(at)ualg.pt
S.M. Jesus, sjesus(at)ualg.pt
and
A. Silva , asilva(at)ualg.pt
SiPLAB, University of Algarve, Campus de Gambelas,
PT-8005-139 Faro, Portugal

Comments: download file (cintal_rep).
Ref.: SiPLAB Report 01/23, SiPLAB, University of Algarve, April 2023.

Abstract
This report aims to cover the requirements for task 4.2 - exploratory acoustic data analysis, of the K2D program of work, regarding the passive acoustic data for soundscape monitoring. Exploratory data analysis encompasses various stages such as 1) on-the-sensor hardware calibration, 2) pre-processing for security sensitivity mitigation and data reduction, 3) visualization for event identification and tracking and finally 4) data archiving in a well documented and open format, encouraging data sharing. The various steps are detailed while insisting in the required system parameters: 1) for equipment calibration aiming at setting measurements that are hardware independent and comparable in international standards; 2) detailing the various options for on/off-the-sensor pre-processing, advantages and drawbacks; 3) the importance of data averaging and frequency domain representation for both sensitivity mitigation and data reduction purposes; and 4) the requirements for acoustic data archiving, re-usage and file sharing, with all its components and metadata descriptors. Tools, diagrams and examples are given to illustrate each step. The link to K2D is done through two illustrative simulated data examples. The first one is based on the time series recorded on an hypothetical cable repeater module in a deep water area off the coast of Portugal mainland at the level of Figueira da Foz. The receiver is located in between the up and down going tracks of the TSS (Traffic Separation System) so that simulated shipping using real AIS data from 2019, is intense and representative. One of the open source available tools for data analysis is run on the simulated data to close the example. The second example is set on the scenario of the Setúbal plateau, near the town of Sesimbra where a project wet demonstrator is planned to be deployed. The report is concluded on a number of requirements and recommendations to be taken into account for setting up a coherent data management system of the future CAM-ring infrastructure.