For the purpose of response surface methodology and surrogate modeling, sampling the input space is an important problem in order to determine the accuracy of the models. In this report, we develop a stochastic technique of data sampling and we define a novel framework for optimizing experimental designs according to arbitrary performance criteria. Thus it is unifying solution which enhances features of several different standard sampling techniques. The approach is based around the tuning of the Fourier coefficients of the probability wave function, for our stochastic sampler.