Super-resolution frequency estimation of the sinusoidal signals with unknown lowpass envelopes

The problem of estimating the frequency of the sinusoidal signals with low pass envelopes in the presence additive white Gaussian noise is addressed. A super-resolution frequency estimator based on the eigenanalysis and quadratic programming is proposed. The frequency estimation is achieved from the single experiment data without a priori knowledge of the lowpass envelop, which is omnipresent in many applications. The performance studies, which include the theory analysis, the numerical simulations of the mean square errors (MSEs) of the frequency estimates compared with the Cramer-Rao bound (CRB), the comparisons between the proposed method and the other representative methods, are presented.