Method of moments estimation for energy spectrum sensing

Abstract

Energy detection is a well-known detection method for spectrum sensing in cognitive radio. Its low complexity and the fact that it does not require any prior knowledge of the primary signals, has made it a popular method. Despite its simplicity, energy detectors require knowing some parameters to set the decision threshold (according to a predefined criterion) and also to estimate its detection performance. Those parameters are the noise power, the primary signal power, and the duty cycle of the primary network. In this work, we propose a new sequential estimation method for jointly estimating those parameters from the energy measurements exclusively. Applying the Method of Moments, we derive the estimators as closed-form functions of the energy values. Their estimation performance is experimentally evaluated by means of over-the-air experiments with a testbed based on software-define-radio devices. The experiments also show the performance of the energy detectors with the estimated parameters.