FSB-System: A Detection System for Fire, Suffocation, and Burn Based on Fuzzy Decision Making, MCDM, and RGB Model in Wireless Sensor Networks

Wireless sensor networks (WSNs) are composed of low-power, large-scale, low-cost sensor nodes to sense environmental conditions (e.g., temperature). Fire is one of the most common hazards in the world so that detection of the fires can prevent a lot of damages to the lives. Fire detection process can be improved by using knowledge-based systems such as fuzzy decision making and multi-criteria decision making (MCDM). This paper proposes a detection system, called FSB-System, to predict the fire, suffocation, and burn probabilities over areas using fuzzy theory, MCDM, and an RGB model. The system uses sensing data of the temperature, smoke, and light sensors to determine appropriate, assorted decisions under different conditions. Three fuzzy controllers are suggested in FSB-System: fire fuzzy controller (namely FFC), suffocation fuzzy controller (namely SFC), and burn fuzzy controller (namely BFC). FFC determines the fire probability, SFC measures the suffocation probability, and BFC calculates the burn probability. Sensor nodes are randomly scattered over areas in a way that they form multiple clusters. Non-cluster heads (NCHs) transmit their sensing data to cluster heads (CHs). Furthermore, CHs transmit the gathered data to the native sink to report environmental conditions toward a base station (e.g., a fire department). The number of sinks is determined by a suggested MCDM controller based on network size and the number of clusters. Simulation results demonstrate that the proposed system surpasses the threshold methods in terms of remaining energy, the number of alive nodes, network lifetime, the number of wrong alerts, and financial losses. This system can be applied in various environments including forests, buildings, etc.