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

• Published in: Mathematics (Basel) Vol. 11; no. 6; p. 1400
• Main Authors: Kim, O-Jong, Kee, Changdon
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.03.2023
• Subjects: Analysis; Artificial intelligence; Artificial neural networks; Building materials; Construction materials; Data compression; Decomposition; Error analysis; Error reduction; Global navigation satellite system; Global Positioning System; Global positioning systems; GPS; Indoor environments; indoor navigation; Machine learning; Methods; multipath mitigation; Navigation systems; Neural networks; Phase measurement; Radio navigation; Receivers & amplifiers; Signal detection (Electronics); Signal processing; single transmitter-based positioning system; Support vector machines; transfer learning; Wavelet analysis; Wavelet transforms; South Korea
• ISSN: 2227-7390; 2227-7390
• DOI: 10.3390/math11061400

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