Design and Implementation of a Low-Power Edge WSN Architecture for Mangrove Health Monitoring in Restoration Sites

Abstract

This research presents an AI-ready, scalable, low-power wireless sensor network (WSN) framework for smart environmental monitoring in remote and ecologically sensitive coastal areas, with a focus on mangrove replantation sites impacted by plastic pollution and climate change. The system integrates custom-designed ESP32 microcontroller-based master and slave sensor nodes powered by solar energy, capable of measuring key environmental parameters including sea water level, total dissolved solids (TDS), pH, and temperature, while supporting AI-driven analytics for predictive ecosystem health assessment. Communication between nodes is achieved using LoRa on the free ISM band, employing time-slot scheduling, multi-path routing, and acknowledgment-based protocols to ensure reliable long-range, low-power data delivery. Edge computing capabilities allow on-node data preprocessing with statistical filtering, event-based reporting, and adaptive transmission to optimize bandwidth and energy consumption. A cloud-based Firebase dashboard enables real-time visualization, historical trend analysis, and integration with AI models for anomaly detection and early-warning alerts. By linking water quality monitoring with plastic pollution impact assessment and providing a robust, low-maintenance, and intelligent monitoring infrastructure, this system offers a powerful tool for advancing ecosystem restoration, sustainability management, and informed, data-driven decision-making.