Influence of A/D quantization in a renewable energy system using Generalized Maximum Sidelobe Decay windows for the frequency estimation

Abstract

This paper presents the influence of the number of A/D converter bits b on frequency estimation errors in a new method based on GMSD time windows, the FFT procedure and three points of the spectrum. Estimation errors are inversely proportional to 2^b and to the square root of the number of samples N in the measurement window. Simulation results are presented for practical applications for b in the range from 8 to 24 and N in the range from 32 to 2048. Proposed equations for rough upper limits of estimation errors allow to determine the accuracy and choose the A/D converter in the real time application.