Localization of Internet of Things Network via Deep Neural Network Based Matrix Completion

In this paper, we propose a technique to acquire the sensor map of Internet of Things (IoT) network. Our approach consists of two main steps to reconstruct the Euclidean distance matrix. First, we recast Euclidean distance matrix completion problem into the alternating minimization problem. We next employ a cascade of multiple deep neural networks to recover the location map of sensors (and the original distance matrix) from the noisy observed matrix. From the numerical experiments, we demonstrate that the proposed method can achieve an accurate reconstruction performance of the distance matrix with much smaller measurement required by conventional approaches and also outperforms state-of-the-art matrix completion algorithms both in noisy and noiseless scenarios.