Research on a Wi-Fi RSSI Calibration Algorithm Based on WOA-BPNN for Indoor Positioning

Owing to the heterogeneity of software and hardware in different types mobile terminals, received signal strength indication (RSSI) from same Wi-Fi access point (AP) varies indoor environments, which can affect positioning accuracy fingerprint methods. To solve this problem consider nonlinear characteristics propagation attenuation, we propose a whale optimisation algorithm-back-propagation neural network (WOA-BPNN) model for RSSI calibration. Firstly, as selection initial parameters BPNN has considerable impact on calibration algorithm, use WOA avoid blindly selecting model. Then, an improved convergence factor balance searchability WOA, also help optimise algorithm. Moreover, change structure compare its influence effect WOA-BPNN Secondly, view low algorithms, region-adaptive weighted K-nearest neighbour algorithm based hierarchical clustering. Finally, effectively combine two proposed algorithms results with those other such linear regression (LR), support vector (SVR), BPNN, genetic algorithm-BPNN (GA-BPNN) algorithms. The test show that among increase (one sigma error) by 41%, 42%, 44% 36%, average. field tests suggest methods reduce error caused heterogeneous differences terminals.