Use of Soft-Decision TOA for Location Estimation

A soft-decision range estimation has been proposed, which outputs a list of likely discrete distances with a list of weights similar to likelihood values. Its application to location estimation has a potential for improving the estimation accuracy, but we need to consider two fundamental problems such as how to furthermore improve the performance of the soft-decision range estimation and how to modify the discrete output to be suited for the continuous maximization in location estimation. In this paper, we tackle the above two problems; to solve the first problem, we introduce the K-means algorithm instead of a conventional threshold-based clustering, and to solve the second problem, we propose a continualization method using an asymmetric Gaussian function which makes it possible to apply gradient-based maximization algorithms.