Spectral magnitude and phase estimation of nonuniformly sampled modulated signals

A technique of estimating the spectral amplitude and phase of nonuniformly sampled digitally modulated signals is described. The novelty of this method resides in the use of lower sampling rates that it would be necessary when classical DSP was used. The technique has many advantages such as digital alias-free spectrum estimation, amplitude and phase estimation, and accurate carrier frequencies estimation. The estimation is performed over a, wide range of frequencies while maintaining low sampling rate, which can not be achieved using uniform sampling under Nyquist bound. The method involves segmenting the received sampled signal and averaging the nonuniform power spectrum of these segments. Then, a phase estimation algorithm based on least squares estimation is exploited, utilising the intrinsic information obtained from the power spectrum estimation. Simulation results on the performance of the algorithm are reported.