Smart Home Security System and Algorithm Improvement

Starting from the actual functional requirements of families, this paper mainly studies the inspection and realization of environmental indicators such as household temperature and humidity and combustible gas concentration in smart homes. The traditional smoke and temperature alarm system may have false positives in the face of factors such as smoke and temperature rise in daily life, and its stability is poor. In order to improve the accuracy of early warning of fire events, this paper designs an algorithm based on fuzzy reasoning and neural network reasoning, and uses multi-source information fusion technology to select CO, smoke concentration and temperature signals as the input signals of the system according to the characteristic phenomenon of fire. The controller uses an improved algorithm to process the collected data and output the fire probability to predict the fire.