DNN-Based Approach to Mitigate Multipath Errors of Differential GNSS Reference Stations
Dongchan Min, Minchan Kim, Jinsil Lee, M.-S. Circiu, Michael Meurer, Jiyun Lee
Abstract
One of the major error components of differential global navigation satellite systems is a multipath error in a reference station. This paper introduces a deep neural network based multipath modeling method. A signal to noise ratio, as well as satellite geometry, is used as a feature parameter to capture the variation of the multipath error caused by unavoidable changes in the vicinity of the reference station. The performance of the proposed method is demonstrated for both normal and varying multipath cases using experimental data. The remaining multipath error after mitigation is well bounded by the standardized error model.
Topics & Concepts
GNSS applicationsMultipath propagationComputer scienceMultipath mitigationSatelliteSatellite systemArtificial neural networkDifferential (mechanical device)AlgorithmRemote sensingGlobal Positioning SystemReal-time computingTelecommunicationsArtificial intelligenceEngineeringGeographyChannel (broadcasting)Aerospace engineeringGNSS positioning and interferenceInertial Sensor and NavigationTarget Tracking and Data Fusion in Sensor Networks