IEEE Project Abstract

Identification of line-of-sight (LOS) and non-line of-sight (NLOS) propagation conditions is very useful in ultra wide band (UWB) localization systems. In the identification, supervised machine learning is often used, but it requires exorbitant efforts to maintain and label the LOS and NLOS database. In this paper, we apply unsupervised machine learning approach called expectation maximization (EM) for Gaussian mixture models (GMM) to classify LOS and NLOS components. The key advantage of applying unsupervised machine learning is that it does not require any rigorous and explicit labeling of database at a certain location. Simulation results demonstrate that by using the proposed algorithm LOS and NLOS signals can be classified with 86.50 percent correct rate, 12.70 percent false negative, and 0.8 percent false positive rate. We also compare the proposed algorithm with the existing cutting-edge supervised machine learning algorithms in terms of computational complexity and signals classification performance.