A NEW ALGORITHM FOR IMPROVING THE ACCURACY OF PERIODIC SIGNAL ANALYSIS

Digital periodic signal analysis often requires synchronized sampling with the signal being analyzed. In certain practical situations, howev er, this condition is difficult to satisfy. As a result, a number of u ndesirable effects such as the spectral leakage associated with the di screte Fourier transform (DFT), and the truncation errors in digital w attmeters arise and degrade system performance. This paper presents a new approach which attempts to remedy the underlying problem. The basi c idea of the proposed method is to modify the actual sampled sequence such that it becomes an ideal sample sequence which is synchronized w ith the signal subjected to sampling. A simple algorithm for modifying the sampled sequence on-line is derived based on interpolation. The p roposed approach requires quite modest additional computational burden which makes it suitable for real-time signal processing. To illustrat e the practical applicability of the proposed algorithm, the paper con siders two distinct but common cases, First, it shows how the proposed method can be used in the case of DFT analysis of harmonic signals, a nd secondly, it considers the digital wattmeter application area in el ectrical power-system measurement. Results show that the proposed algo rithm is capable of reducing both the leakage effect in DFT analysis a nd truncation errors in digital wattmeters.