Improved fourier polynomial based phase modeling for estimating instantaneous frequency from a noisy FM signal

Abstract

Instantaneous frequency (IF) from a frequency modulated (FM) signal has number of applications in engineering disciplines such as telecommunication, analysis of radar signal, music signal, electrocardiogram signal, speed estimation from encoder signal, etc. There are many IF estimation techniques and these are divided into various groups, namely, zero crossing detection technique, time domain technique, frequency domain technique, time frequency domain technique. Among the various IF estimation techniques, Phase modeling by Fourier polynomial can estimate IF from a FM signal by fitting the Fourier polynomial to the zero crossing points. However, in a noisy FM signal, pseudo zero crossing points appear in place of actual zero crossing point. Therefore, a simple Fourier polynomial based phase modeling technique encounters some difficulties while it is applied in a signal of lower signal-to-noise ratio (SNR). Hence, improved Fourier polynomial based phase modeling technique has been developed for estimating IF from FM signal with low SNR. This improved technique is superior to simple Fourier polynomial based phase modeling technique and estimates better IF.