Research on Multi-Sensor Fusion Firefighting Manpower Status Monitoring Technology

Firefighting soldiers are in danger during firefighting operations, and it is necessary to monitor the data of their physical indicators to ensure the safety of firefighting soldiers, improvement of the safety of firefighting personnel. A single sensor will appear the information blind spot, and has the defects of measurement limitation, low robustness and low precision, which can’t satisfy the need of accurately judging the current state of the firefighting soldier. Therefore, this paper investigates the distribution characteristics of multi-dimensional data of firefighting soldiers in the process of working movement, The goal of multi-sensor information fusion is based on the separation of observations from each sensor, and its ultimate goal is to utilize the advantages of multiple sensors operating together or in conjunction to improve the effectiveness of the entire sensor system, the distribution characteristics of multi-dimensional data effectively improve the shortcomings of a single sensor. By introducing the idea of improving the local optimum of back propagation of BP neural network algorithm and the defects of network weights, this paper proposes a relevant fusion algorithm based on firefighting soldiers, establishes a kinematic physiological monitoring model based on firefighting soldiers, and realizes the physiological state monitoring of firefighting soldiers in the complex working environment. Experimental results show that this method can improve the accuracy with which data from several sensors determine the condition of firefighters, decision making for firefighting singletons’ safety and security can be effectively improved by multi-sensor data fusion, and the accuracy of the improved FWA-BP data fusion algorithm model is increased by 3.64%, which improves the stability of the system.