Abstract—This article introduces to the best of our knowledge a novel approach for simultaneous estimation of time delay and frequencies in noisy complex sinusoidal signals received at two spatially separated sensors. The proposed method comprises two main components. Firstly, a Convolutional Neural Network (CNN) regression model is employed to estimate frequencies using data from the first sensor. The model is trained on a synthetic dataset specifically designed for this task. Secondly, a deep learning model is developed, incorporating densely connected layers and dropout layers for regularization, to effectively estimate the time delay between the received signal copies at the two sensors. Extensive computer simulations demonstrate the effectiveness of the proposed method, showcasing its accuracy in joint time delay and frequency estimation. This deep learning-based technique offers a promising alternative to classical signal processing approaches, enabling advanced signal analysis in diverse engineering domains.


Keywords—delay and frequency estimation, deep learning, convolutional neural networks, temporal patterns


Cite: Mahmoud M. Qasaymeh and Ahmad Falah Aljaafreh, “Joint Time Delay and Frequency Estimation Based on Deep Learning," Journal of Communications, vol. 19, no. 1, pp. 1-6, 2024. 

Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.