Evaluation of stochastic-resonance-based detectors of weak harmonic signals in additive white Gaussian noise

A thorough evaluation of stochastic resonance in the framework of statistical detection theory is presented both as a nonlinear signal preprocessor and as a detector. The pertinent receiver operating characteristics are compared with those of the known statistically optimum detector using extensive Monte Carlo simulations. Parameter optimization and computational budget aspects are discussed.