Power spectral density estimation from random interleaved samples

Abstract

Random interleaved sampling (RIS) is a widely used non-uniform sampling technique in signal sampling system. RIS composites a waveform with high equivalent rate from low speed samples acquired using single analog-to-digital converter (ADC). However, in many applications, we are only interested in power spectral density (PSD) of signal, and the whole waveform reconstruction is inefficient and unnecessary. In this paper, an algorithm is proposed to estimate PSD of signal, which is captured using RIS technique. Duo to incoherence between trigger pulse and sampling clock, the trigger time interval may be randomly distributed within the range of a period of sampling clock. The quantization error always exists. The fractional delay filter is incorporated into RIS to mitigate the quantization error that would degrade estimate performance. Experiment is reported to evaluate the proposed PSD estimation algorithm, and the results indicate that the proposed approach is feasible and efficient.