Version 1
: Received: 1 August 2022 / Approved: 3 August 2022 / Online: 3 August 2022 (03:15:26 CEST)
How to cite:
Razzaq, H.S.; Hussain, Z.M. Instantaneous Frequency Estimation of FM Signals under Gaussian and Symmetric α-Stable Noise: Deep Learning versus Time-Frequency Analysis. Preprints2022, 2022080068. https://doi.org/10.20944/preprints202208.0068.v1
Razzaq, H.S.; Hussain, Z.M. Instantaneous Frequency Estimation of FM Signals under Gaussian and Symmetric α-Stable Noise: Deep Learning versus Time-Frequency Analysis. Preprints 2022, 2022080068. https://doi.org/10.20944/preprints202208.0068.v1
Razzaq, H.S.; Hussain, Z.M. Instantaneous Frequency Estimation of FM Signals under Gaussian and Symmetric α-Stable Noise: Deep Learning versus Time-Frequency Analysis. Preprints2022, 2022080068. https://doi.org/10.20944/preprints202208.0068.v1
APA Style
Razzaq, H.S., & Hussain, Z.M. (2022). Instantaneous Frequency Estimation of FM Signals under Gaussian and Symmetric α-Stable Noise: Deep Learning versus Time-Frequency Analysis. Preprints. https://doi.org/10.20944/preprints202208.0068.v1
Chicago/Turabian Style
Razzaq, H.S. and Zahir M. Hussain. 2022 "Instantaneous Frequency Estimation of FM Signals under Gaussian and Symmetric α-Stable Noise: Deep Learning versus Time-Frequency Analysis" Preprints. https://doi.org/10.20944/preprints202208.0068.v1
Abstract
Deep Learning (DL) and Machine Learning (ML) are widely used in many fields, but rarely used in Frequency Estimation (FE) and Slope Estimation (SE) of signals. Frequency and slope estimation for Frequency-Modulated (FM) and single-tone sinusoidal signals are essential in various applications, such as wireless communications, sonar, and radar measurements. In this work, artificial neural network (ANN) and convolutional neural network (CNN) are used in frequency and slope estimation for FM signals under Additive White Gaussian Noise (AWGN) and Additive Symmetric alpha Stable Noise (SαSN). SαS distributions are impulsive noise disturbances found in many communication environments like marine systems; their distribution lacks a closed-form Probability Density Function (PDF), except for specific cases, and infinite second-order statistic, hence Geometric SNR (GSNR) is used in this work to determine the impulsiveness of noise in a mixture of Gaussian and SαS noise processes. ANN is a machine learning classifier, designed with few layers for reducing FE and SE complexity while getting higher accuracy as compared with classical techniques. CNN is a deep learning classifier, designed with many layers for FE and SE, and proved to be more accurate than ANN when dealing with big data and finding optimal features. Simulation results show that SαS noise can be much more harmful for FE and SE of FM signals than Gaussian noise. DL and ML can significantly reduce FE complexity, memory cost, and power consumption, which is important in many systems such as some Internet of Things (IoT) sensor applications. After training DCNN for frequency and slope estimation of LFM signals, the performance of DCNN (in terms of accuracy) can give acceptable results at very low signal-to-noise ratios where TFD fails, giving more than 20dB difference in the GSNR working range.
Keywords
Frequency estimation; FM; sensors; Internet of Things (IoT); software-defined radio (SDR); al-pha-stable noise; time-frequency distribution; deep learning
Subject
Engineering, Electrical and Electronic Engineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
