Wavelet and Neural Network-Based Multipath Detection for Precise Positioning Systems

Multipath errors are significantly challenging in radio navigation systems. In particular, multipath indoor environments cause significant the position domain because not only building materials that surround environment but also all objects inside can reflect signals. outdoor environments, such as global satellite system (GNSS) signal applications, have been widely studied for precise positioning. However, studies applications rarely conducted of complicated and many made various small areas. this study, mitigation methods using a shallow neural network transfer learning-based deep were respectively considered to overcome complexity caused by reflected signals environments. These classify each measurement according whether exhibits severe error. Carrier-phase measurements broadcasted from transmitter used wavelet transform, magnitude values after transform network-based learning. Shallow networks attain approximately 87.1% 85.6% detection accuracies, respectively, positioning error be reduced 10.4% 9.4%, mitigation.