Skip to main navigation Skip to search Skip to main content

Yapay Sinir Aglari ile D s k Karma siklikli Y n Bulma

Translated title of the contribution: Low-Complexity Direction Finding with Artificial Neural Networks
  • ASELSAN Inc.
  • Istanbul Technical University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this study, an artificial neural network-based direction finding method is proposed as an alternative to traditional correlation-based direction finding algorithms. Through simulations, it has been verified that the proposed method achieves a lower root mean square error (RMSE) compared to correlative interferometer. Additionally, it has been demonstrated that the computational complexity and memory requirements of the proposed method are lower than correlation-based direction finding method. The proposed artificial neural network-based method has been validated to perform direction finding with high accuracy for frequency values not encountered in the training set during the testing phase.

Translated title of the contributionLow-Complexity Direction Finding with Artificial Neural Networks
Original languageTurkish
Title of host publication33rd IEEE Conference on Signal Processing and Communications Applications, SIU 2025 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331566555
DOIs
Publication statusPublished - 2025
Event33rd IEEE Conference on Signal Processing and Communications Applications, SIU 2025 - Istanbul, Turkey
Duration: 25 Jun 202528 Jun 2025

Publication series

Name33rd IEEE Conference on Signal Processing and Communications Applications, SIU 2025 - Proceedings

Conference

Conference33rd IEEE Conference on Signal Processing and Communications Applications, SIU 2025
Country/TerritoryTurkey
CityIstanbul
Period25/06/2528/06/25

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Fingerprint

Dive into the research topics of 'Low-Complexity Direction Finding with Artificial Neural Networks'. Together they form a unique fingerprint.

Cite this