Answers are grounded in this paper and cite their sources.
This study presents an MQTT-based prototype for water-level monitoring, local flood warnings, and remote dashboard monitoring. The system integrates a JSN-SR04T ultrasonic sensor, ESP32, SHT31, LCD, buzzer, relay-controlled lamps, Wi-Fi, an MQTT broker, and a dashboard. Tests included 15 JSN-SR04T readings at each reference distance (20, 30, and 50 cm) and 15 paired SHT31 measurements. MAPE was 0.38% at 30 cm and 0.12% at 50 cm; the 20 cm test point was below the cited 25 cm minimum range. SHT31 MAE was 0.193 °C for temperature and 0.713 percentage points for relative humidity. In the QoS 1 test, 596 of 600 messages were uniquely received, four were not received, and two duplicates were recorded, yielding 99.33% PDR, 0.67% PLR, a reception rate of 0.993 messages/s, and a mean RTT of 137.8 ms. Ten trials per output gave mean response times of 14.0 ms for the buzzer and 26.5 ms for the relay-controlled lamp. Across five 60 s outages, mean MQTT recovery and dashboard update times were 2.14 s and 2.66 s, respectively; local warnings remained active. The recorded threshold states matched the programmed rule at all six tested points (100% agreement, 6/6). Full-range accuracy and field reliability remain unestablished.
No takes yet. Share an insight, caveat, or question.
Hasin et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: