A Novel Carrier Loop Based on Adaptive LM-QN Method in GNSS Receivers

A well-designed carrier tracking loop in a receiver of the Global Navigation Satellite System (GNSS) is premise accurate positioning and navigation an aircraft-based surveying mapping system. To deal with problems Doppler estimation high-dynamic maneuvers, interest on maximum-likelihood (MLE) increasing among academic community. Levenberg-Marquardt (LM) method usually regarded as effective promising approach to obtain solution MLE, but computation Hessian matrix loads great burden algorithm. Besides, poor performance convergency final iterations common failing LM implementations. solve these problems, based Gauss-Newton Quasi-Newton (QN) approximation are derived, making cost decline from O(N) xmlns:xlink="http://www.w3.org/1999/xlink">O( 1 xmlns:xlink="http://www.w3.org/1999/xlink">) . Then, basis two methods, closed adaptive LM-QN algorithm further proposed which can switch between QN adaptively according damping parameter. ideal super-linear convergence (SLM) constructed proved reference analysis. Finally, through analyses experiments using aircraft data, improvements verified. Compared scalar vector tracking, results indicate magnitude increase precision loop, even though more counts needed by LM-QN.