Estimation of spectrum occupancy in heterogeneous radio access environments using random spectral sampling

In this paper, we investigate the performance of randomized spectral sampling for temporal characterization of spectrum occupancy in wideband spectrum sensing. The theoretical framework for performance analysis is developed for scenarios involving heterogeneous radio access technologies (RAT). We propose a channel - agnostic random sampling approach for wideband spectrum occupancy modeling across low-to-high occupancy settings. The performance is studied for various randomized spectral sampling schemes using spectrum measurements in Worcester, MA spanning 800MHz-1GHz.