Research on Indoor Environment Positioning System considering Multisensor in the Multi-Information Data Fusion

Abstract

Outdoor positioning can often achieve accurate positioning according to GPS and mobile phone signaling, while indoor positioning is difficult to meet the needs of practical application due to the limitations of satellite reception. In order to effectively solve the problem of large error in the individual positioning strategy in the indoor environment, this paper applies multisensor in the multisource information fusion indoor positioning system. By using the positioning results of multiple sensors to limit the range of geomagnetic matching for combined matching, the matching error can be effectively reduced. Then, the global optimal value of indoor network is calculated by using the multi-information data fusion algorithm, which can optimize the initial value and threshold of the multi-information data fusion algorithm, improve the network accuracy as much as possible, and accelerate the convergence speed at the same time. After completing the optimization processing, the indoor network can obtain the combined positioning and predicted positioning results, so as to facilitate the fusion training to the actual position coordinates, and finally obtain the optimal positioning results. The simulation results show that the mean square error predicted by the multi-information data fusion algorithm calculated by the multi-information data fusion algorithm can be effectively reduced by 76%, and the fusion positioning accuracy can be improved by 48% compared with the accuracy of a single positioning strategy. The method proposed in this paper effectively improves the positioning accuracy, indicating that the positioning performance is better.