Research on indoor positioning of power grid equipment based on deep learning

Abstract

In view of the increasing number of power grid equipment and a large number of indoor deployments, the existing positioning technology cannot meet the requirements. In this paper, a DOA estimation method based on deep learning is proposed to improve the indoor positioning accuracy. Firstly, by studying the propagation characteristics of wireless signals and the transmission characteristics of indoor wireless channels, a statistical channel model with multi subcarrier characteristics is established to generate channel impulse response information. Furthermore, the transformation matrix required for virtual antenna expansion is designed by using the idea of interpolation transformation. This matrix is combined with the channel impulse response to obtain the virtual antenna array data for DOA estimation. The DOA estimation model based on CNN is optimized to obtain high-precision DOA. Finally, the simulation results show that the positioning accuracy can be doubled by using the virtual antenna array technology. The performance of the two-stage DOA estimation algorithm proposed in this paper is slightly better than the accuracy of a large difference step. Even when there are fewer APS, the positioning accuracy is better than the traditional MUSIC algorithm, which is more suitable for indoor positioning of a large number of power grid equipment in the future.