Development of a Time-Sensitive Smoke Detection System with Dynamic Threshold Scheduling Using Real-Time Clock on Arduino

Abstract

This research paper presents the design, implementation, and experimental evaluation of a time-sensitive smoke detection system that addresses the persistent false-alarm problem in residential fixed-threshold detectors by introducing scheduled sensitivity adjustments driven by a DS3231 real-time clock module. Building upon the Arduino Uno-based prototype developed by Calderon et al. at EARIST, which employed an MQ-2 smoke sensor, a secondary gas sensor, and an infrared flame sensor operating against a single fixed threshold of 150, this study introduces one low-cost hardware addition and four layered software improvements. The DS3231 RTC module provides the Arduino with accurate time-of-day information to within ±2 minutes per year via its integrated temperature-compensated crystal oscillator. The four software improvements are: (1) averaged sensor sampling across five consecutive readings to suppress electrical noise; (2) a sustained-detection timer requiring continuous threshold exceedance for 3,000 milliseconds before alarm activation; (3) a rate-of-change filter that identifies and rejects sharp spike-and-drop signal trajectories characteristic of non-fire aerosol sources; and (4) a time-based threshold scheduling mechanism applying a higher-tolerance threshold of 280 during cooking hours, a strictest threshold of 160 during nighttime sleep hours, and a standard threshold of 220 during all other periods. The LCD continuously shows the current active mode and threshold value. Experimental results demonstrate that the improved system reduced the false alarm rate from 100 percent across three non-fire source types in the original system to zero percent, while maintaining reliable alarm activation for all genuine fire sources across all three threshold modes. The findings establish time-of-day sensitivity scheduling as a viable, low-cost strategy for improving smoke detector reliability in household environments.