Deep Learning and Matrix Completion-aided IoT Network Localization in the Outlier Scenarios

Abstract--In this paper, we propose a deep learning and matrix completion aided approach for recovering an outlier contaminated Euclidean distance matrix D in IoT network loc al-ization. Unlike conventional localization techniques tha t search the solution over a whole set of matrices, the proposed techn ique restricts the search to the set of Euclidean distance matric es. Specifically, we express D as a function of the sensor coordin ate matrix X that inherently satisfies the unique properties of D, and then jointly recover D and X using a deep neural network. T o handle outliers effectively, we model them as a sparse matri x L and add a regularization term of L into the optimization prob lem. We then solve the problem by alternately updating X, D, and L. Numerical experiments demonstrate that the proposed techn ique can recover the location information of sensors accurately even in the presence of outliers.