RSSI-Based Indoor Localization Using Sparrow Search Algorithm and Backward Propagation Neural Network

Traditional localization algorithms based on received signal strength indication (RSSI) construct ranging model through the log-normal shadowing model, in which the parameters are usually chosen empirically and therefore sensitive to the influence of environments. In this paper, an indoor localization method based on sparrow search algorithm (SSA) and backward propagation (BP) neural network is proposed to construct the ranging model. In the proposed SSA-BP method, the collected RSSI values are firstly processed by Gaussian filtering and then input into the SSA-BP neural network to construct the ranging model, which can output distance values between the target nodes (TN) and anchor nodes. Finally, the TN's position can be estimated by using the maximum likelihood estimation (MLE) method. Simulation results show that compared with the existing BP algorithm and genetic algorithm (GA), the proposed SSA-BP algorithm has faster convergence speed and higher positioning accuracy.