Empirical Characterization of Man-Made Impulsive Noise Using Open-Source Software Defined Radio Platforms

With the widespread use of wireless systems in the modern society, every household, office, and even automobiles, now contain radio-frequency (RF) sources capable of producing interference. Unlike thermal noise, whose statistics can usually be assumed stationary and Gaussian, man-made noise includes impulsive components that depend on the environment and is described by non-Gaussian statistical models that are in general non-stationary. This paper presents the use of affordable software-defined radio (SDR) platforms for measuring and characterizing impulsive noise for various locations using a Middleton Class A impulsive noise model.