Fuzzy interpolation of the average signal steps

In this paper is proposed a fuzzy interpolation method of the average signal steps in each processing stage, for extraction of signal drowned in noise. The fuzzy interpolation method increases the signal-to-noise-ratio (SNR) gain for a periodic signal drowned in noise, and may gives good results for different other signal processing applications, such as: extraction of periodic signals combination drowned in noise, signal shape reconstruction etc. Recommended sampling frequency is up to N_s times of frequency given by the Shannon´s condition, where number of signal samples on one time stage, N_s, is usually the order of hundreds or thousands. The simulation and experimental results obtained with periodic signals drowned in noise are given using the Matlabcopy and a digital signal processing (DSP) platform, respectively. The proposed filtering method is compared with other similar methods by computing the SNR gain.