Multi-sensor Data Fusion based on Fuzzy Theory and FWA-BP Neural Network

With the development of science and technology, there are more and more types of sensors, and their performance is getting better and better. But a single sensor is less effective in the perception of complex systems. In order to solve the problem of data access difficulties caused by the diversification of sensors, a new multi-sensor data fusion model is designed based on fuzzy theory and combining fireworks algorithm and back propagation neural network. The experimental results show that the accuracy of the model is up to 92%, the training time of the algorithm is only 90s, and the computational resources are less consumed, and the algorithm has 30 iterations. The computational time consuming of the model is relatively stable, mainly concentrated in the time range from 10ms to 20ms. Therefore, the proposed multi-sensor data fusion model has the advantages of less resource consumption, high accuracy and short computational time consuming, which provides new technical support for the development of the Internet of Things.