Pathloss-based non-Line-of-Sight Identification in an Indoor Environment: An Experimental Study

We report the findings of an experimental study on the problem of line-of-sight (LOS)/nonline-of-sight (NLOS) classification in an indoor environment. Specifically, we deploy a pair of NI 2901 USRP software-defined radios (SDR) in a large hall, which communicate on a center frequency of 2.4 GHz, using three different signal-to-noise ratios (SNR). The receive SDR constructs a dataset of pathloss measurements from the received signal as it moves across 15 equi-spaced positions on a 1D grid (for both LOS and NLOS scenarios). This allows us to estimate the pathloss parameters using the least-squares method to construct a parameterized pathloss model for a binary hypothesis test (BUT) for NLOS identification. Since the pathloss measurements slightly deviate from Gaussian distribution, we also feed our custom dataset to a range of machine learning (ML) algorithms. It turns out that the performance of the ML algorithms is only slightly superior to the Neyman-Pearson-based BHT.