Efficiency of amplitude suppression on sinusoidal interference

Polygonal approximation of nonlinear conversion is considered for amplitude suppression of non-Gaussian interference, specifically sinusoidal, with predominance of radio-frequency components in spectrum. Sinusoidal interference is defined as additive mixture of Gaussian white noise and harmonic fluctuations with constant amplitude and frequency but random initial phase uniformly distributed over the -pi, pi interval and arbitrary phase modulation. Statistical characteristics of this interference are functions of a parameter representing ratio of harmonic fluctuation power to Gaussian noise power. The amplitude suppression factor characterizing efficiency of the nonlinear converter, i.e., signal-to-interference ratio at output divided by signal-to-interference ratio at input, is calculated taking into account converter amplitude characteristic relative to fundamental frequency component and envelope distribution over interference. Results showed that amplitude suppression with a polygonal converter is somewhat less efficient than optimal amplitude suppression but simpler to implement than amplitude suppression with polynomial converter.