Optimization and Application of Indoor Localization Algorithm Based on ELMAN Neural Network

It is a key problem in indoor localization accuracy that the received signal strength indicator (RSSI) is affected by changing environment such as diffraction, obstacles, and multi-path effect. To improve the accuracy, we propose an indoor localization algorithm based on ELMAN Neural Network. Firstly, we estimate the distance from point-to-point RSSI by using Kalman average filter. Then we reconstruct the RSSI according to the estimated distance. At last, we estimate the target coordinates by using ELMAN neural network. Experimental results show that our method can achieve the localization accuracy of 0.915 meter, and has good stability against RSSI variation in the localization.