Monitoring of Migratory Birds in the Wetlands of Taref and Annaba using Wireless Multimedia Sensor Networks

Résumé

Algeria boasts over 250 wetlands, 50 recognized internationally for their ecological importance. These habitats support tens of thousands of waterbirds that winter or stop during migrations, including threatened species like the Common Pochard, White-headed Duck, and Ferruginous Duck, listed by the IUCN. This highlights the critical need for conservation of Algerian wetlands. With rising environmental degradation impacting biodiversity, effective conservation requires advanced, systematic monitoring. This thesis presents a cost-efficient ecological surveillance system for wetlands using Wireless Multimedia Sensor Networks (WMSNs) to monitor waterbird populations. The research introduces a three-tier Internet of Multimedia Things (IoMT) architecture employing an Edge-Fog-Cloud model for continuous, automated wetland surveillance. WMSNs at the system's edge periodically capture and transmit images to a fog-level central station. These images are crucial for estimating populations, gathering habitat data, and conducting ecological analyses. A tailored surveillance topology is designed based on environmental criteria and sound attenuation in the study area. Low-Power and Lossy Networks (LLNs) in this architecture face challenges due to bandwidth and power limitations. To address these, the system uses endangered birds' unique calls to selectively trigger cameras, ensuring only relevant images are captured and transmitted. A noise reduction technique is integrated to enhance automatic bird call recognition in noisy wetland environments. Additionally, captured images undergo low-complexity compression before transmission to boost energy efficiency. Real-world comparisons evaluate the system's accuracy and performance. Although WMSNs offer potential for monitoring, harsh environmental conditions can challenge their operation. To enhance efficiency, Digital Twin (DT) technology is integrated, providing real-time virtual replicas of the network and application. This enables dynamic adjustments and informed decision-making based on real-time data. The interaction between application and network DTs ensures cooperative decisions, meeting Quality of Service (QoS) and Quality of Experience (QoE) requirements. A fuzzy logic-based transmission rate control dynamically adapts to current network states. DTs allow "what-if" scenario testing, optimizing settings for controllers. The integration of DTs in WMSN surveillance systems is validated through extensive real-world experiments, demonstrating improvements in energy efficiency, real-time monitoring, and handling of multiple image sources compared to conventional setups. This thesis addresses LLN challenges in IoMT frameworks, proposing robust solutions for sustainable bird population monitoring and data collection in natural habitats.

Citer ce document

Accès au document

Ce document est hébergé sur son dépôt institutionnel d'origine.

Auteur(s)

Statistiques

Consultations : 2

Téléchargements : 0