<b>Architecting a Low-Cost and Low-Latency, AI-Integrated Edge Node for Real-Time Aquaculture Monitoring in Resource-Constrained Environments</b>

This manuscript details the design, deployment, and socio-technical validation of a low-cost, low-latency Internet of Things (IoT) architecture for precision aquaculture in resource-constrained environments. The proposed system features an autonomous, solar-powered ESP32 edge node capable of in-situ calibration of analog pH and temperature sensors.

To circumvent the high latency and unreliability of traditional database models over fluctuating rural cellular networks, the architecture employs a Local-to-Wide Area Network (LAN-WAN) MiFi bridge, synchronizing lightweight telemetry via Google Firebase's NoSQL infrastructure. Real-time data is pushed to a custom React Native mobile application with an end-to-end latency of ~350ms. Field surveys utilizing the UTAUT framework confirm high user acceptance among Nigerian catfish farmers, proving the system's economic feasibility and its potential to drastically reduce mass mortality events in earthen ponds.