Development and performance evaluation of Correntropy Kalman Filter for improved accuracy of GPS position estimation

It is well known that a Global Positioning System (GPS) receiver needs to ‘see’ at least four satellites to provide a three-dimensional fix solution. However, if any GPS receiver is operated in urban canyons, the visibility further reduces. To improve the position estimation accuracy, a novel kinematic positioning algorithm designated as Correntropy Kalman Filter (CKF) is proposed in this study. Instead of minimum mean square error (MMSE), the correntropy criterion (CC) is used as the optimality criterion of CKF. Like the traditional Kalman Filter (KF), the prior estimate of the state and covariance matrix are computed in CKF, and a novel fixed-point algorithm is then used to update the posterior estimates. The data of a dual-frequency GPS receiver located at the Indian Institute of Science (IISc), Bangalore (13.021°N/77.5°E) is collected from Scripps Orbit and Permanent Array Centre (SOPAC) to implement the proposed algorithm. The results of the proposed CKF algorithm are promising and exhibit significant improvement in position estimation compared to the conventional methods.