Frequency content and autocorrelation function of noisy periodic signals

TL;DR

This paper presents a method using Discrete Fourier Transform to analyze the frequency content of noisy periodic signals, effectively removing spurious autocorrelations and demonstrated on temperature data.

Contribution

It introduces a technique that overcomes incommensurate lattice limitations in frequency analysis and improves autocorrelation removal in noisy signals.

Findings

Effective frequency extraction from noisy signals

Reduction of spurious autocorrelations in residuals

Application to temperature time series

Abstract

We extract the frequency content of a noisy signal by use of Discrete Fourier Transform. Our analysis overcomes the limitations imposed by incommensurate lattices. After computing the deterministic component, we show the relevance of the method in removing spurious autocorrelations from the signal residuals. Results are presented for a temperature time series.