On the Design of KF-Based Localization Based on Side Information

In this paper, we propose a 1-bit algorithm using spatial information to improve the accuracy of Kalman filter (KF)-based location estimation. The proposed algorithm aims to improve position estimation accuracy by re-estimating values outside a feasible region as being at the boundary of that region, based on the information that the user is present within that feasible region. This approach enhances position estimation accuracy without significantly increasing complexity. This paper discusses two methods for applying the 1-bit algorithm and verifies their performance by comparing Time of Arrival (ToA), the ToA-based KF, and the ToA-based KF with the 1-bit algorithm through simulations under three scenarios. Performance analysis was conducted from two perspectives: cumulative distribution function (CDF) and average position error (APE). The ToA-based KF with a 1-bit algorithm demonstrated the best performance. The proposed approach improved performance without high computational complexity and is suitable for real-time applications, making it applicable to indoor positioning, robot navigation, and wireless sensor networks that require high positioning accuracy.