An FPGA-Based WASN for Remote Real-Time Monitoring of Endangered Species: A Case Study on the Birdsong Recognition of Botaurus stellaris

Abstract

Fast environmental variations due to climate change can cause mass decline or even extinctions of species, having a dramatic impact on the future of biodiversity. During the last decade, different approaches have been proposed to track and monitor endangered species, generally based on costly semi-automatic systems that require human supervision adding limitations in coverage and time. However, the recent emergence of Wireless Acoustic Sensor Networks (WASN) has allowed non-intrusive remote monitoring of endangered species in real time through the automatic identification of the sound they emit. In this work, an FPGA-based WASN centralized architecture is proposed and validated on a simulated operation environment. The feasibility of the architecture is evaluated in a case study designed to detect the threatened Botaurus stellaris among other 19 cohabiting birds species in The Parc Natural dels Aiguamolls de l’Empordà, showing an averaged recognition accuracy of 91% over 2h 55’ of representative data. The FPGA-based feature extraction implementation allows the system to process data from 30 acoustic sensors in real time with an affordable cost. Finally, several open questions derived from this research are discussed to be considered for future works.

Deploying Acoustic Detection Algorithms on Low-Cost, Open-Source Acoustic Sensors for Environmental Monitoring

“…Real-time analysis of data from acoustic sensors is increasingly becoming commonplace, with many research projects using low-cost micro-controllers and field-programmable gate arrays (FPGAs) to broadcast acoustic data to a central hub for analysis. This technique has found a wide range of applications, from monitoring noise pollution in urban environments [1] to non-intrusively monitoring threatened bird species [2]. To reduce the cost and energy requirements of environmental sensors, processing is now often carried out on the micro-controllers themselves, utilising their limited processing power to implement detection algorithms which recognise the unique vocalisations of many species [3,4].…”

Mallard Detection Using Microphone Arrays Combined with Delay-and-Sum Beamforming for Smart and Remote Rice–Duck Farming

“…Benchmark datasets and challenges for the verification and comparison of detection performance have been expanded in both vision [50][51][52][53] and acoustic [54][55][56][57][58][59][60][61][62][63][64][65] modalities. During the first half of the 2010s, probability models and conventional machine-learning (ML) models combined with part-based features obtained using local descriptors were used widely and popularly.…”

Bioacoustic IoT Sensors as Next-Generation Tools for Monitoring: Counting Flying Insects through Buzz

“…Image processing and neural network analysis were employed to identify orchard insects [37] and Lepidoptera [38] successfully. On the other hand, Passive Acoustic Monitoring (PAM), which encompasses real-time analysis of data collected by acoustic IoT (Internet of Things) sensors, is becoming an increasingly important tool of ecological research [39] used in various types of surveys, from autonomous bird species recording [40] to urban traffic noise monitoring [41]. Due to the low cost, ease of installation, efficiency, and energy independence of these acoustic devices, it is possible to facilitate real-time and large-scale monitoring [42], acquiring important information on population diversity and dynamics, over a finer daily scale or more extended periods to complement traditional methods.…”