SIGNAL-DEPENDENT ANALYSIS OF SIGNALS SAMPLED BY SEND ON DELTA SAMPLING SCHEME

摘要

Interest in the application of signal driven sampling schemes is increasing as they offer various advantages over traditional sampling. The paper describes the principles and discusses the properties of sampling, which is based on the send-on-delta concept. In such a way, it is possible to decrease the sampling density, and since the samples are placed non-equidistantly it is possible to suppress the distortion due to frequency aliasing. The non-uniform location of samples requires an advanced processing method. The paper discusses the spectral estimation, which is based on the use of a bank of minimum variance filters. To improve the resolution and accuracy, iterative updating of autocorrelation matrix is used. The results of simulations are presented. The use of an iterative algorithm allows correcting spectral estimation even if the mean sampling density is several times less than the Nyquist criterion. The proposed approach can be of interest for distributed wireless data acquisition in remote sensing systems, because it allows the amount of transmitted data to be decreased considerably.