Automatic Selection of Optimal Sample Rate for Band-Pass Signals

Data acquisition systems (DAS) play a fundamental role in a lot of modern measurement solutions. One of the parameters characterizing a DAS is its maximum sample rate, which poses constraints on signals that can be alias-free digitized. Band-pass sampling theory singles out separated ranges of admissible sample rates which can be significantly lower than carrier frequency. But, how to choose the most convenient sample rate according to the purpose at hand? The paper proposes a method mandated to the automatic selection of optimal sample rate for band-pass signals, which takes also into account the effects of sample clock instability and limited resolution. The proposed method allows the user to choose the location of spectral replicas of the sampled signal in terms of normalized frequency, and the minimum guard-band between replicas, thus introducing a feature that no DAS currently available on the market seems to offer. Experimental tests on band-pass digitally modulated signals for communications are carried out to verify the concurrence of actual central frequency with the expected one