Multi-sensor information fusion based on BP network

The output of pressure sensor is affected by non-objection parameters in its application, such as temperature, power fluctuation and so on. Multi-sensor information fusion based on BP network can eliminate the side effect of non-objection parameters and improve the accuracy and reliability for the pressure sensor. This method based on LMBP algorithm of BP network has not only a simple network structure, but also a quick learning rate, showing a better prospect. The paper describes the LMBP algorithm, deduces the matrix formula of the two-layer network, and makes use of LMBP algorithm to fuse the temperature, power fluctuation and pressure information of the pressure sensor. Finally the paper fuses the example data of the pressure sensor with the help of the MATLAB software and draws a conclusion that the LMBP algorithm eliminates easily the affection of the non-objection parameters and improves the accuracy of the pressure sensor.