CINXE.COM

<html><head><title>Hybrid Noise-Resilient Deep Learning Architecture for Modulation Classification in Cognitive Radio Networks - EUDL</title><link rel="icon" href="/images/favicon.ico"><link rel="stylesheet" type="text/css" href="/css/screen.css"><link rel="stylesheet" href="/css/zenburn.css"><meta http-equiv="Content-Type" content="charset=utf-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><meta name="Description" content="The increasing maturity of the concepts which would allow for the operation of a practical Cognitive Radio (CR) Network require functionalities derived through different methodologies from other fields. One such approach is Deep Learning (DL) which can be applied to diverse problems in CR to enhance"><meta name="citation_title" content="Hybrid Noise-Resilient Deep Learning Architecture for Modulation Classification in Cognitive Radio Networks"><meta name="citation_publication_date" content="2019/09/17"><meta name="citation_author" content="Antoni Ivanov"><meta name="citation_author" content="Krasimir Tonchev"><meta name="citation_author" content="Vladimir Poulkov"><meta name="citation_author" content="Hussein Al-Shatri"><meta name="citation_author" content="Anja Klein"><meta name="citation_pdf_url" content="http://eudl.eu/pdf/10.1007/978-3-030-23976-3_20"><meta name="citation_conference_title" content="Future Access Enablers for Ubiquitous and Intelligent Infrastructures. 4th EAI International Conference, FABULOUS 2019, Sofia, Bulgaria, March 28-29, 2019, Proceedings"><meta name="citation_isbn" content="978-3-030-23976-3"><script type="text/javascript" src="https://services.eai.eu//load-signup-form/EAI"></script><script type="text/javascript" src="https://services.eai.eu//ujs/forms/signup/sso-client.js"></script><script type="text/javascript">if (!window.EUDL){ window.EUDL={} };EUDL.cas_url="https://account.eai.eu/cas";EUDL.profile_url="https://account.eai.eu";if(window.SSO){SSO.set_mode('eai')};</script><script type="text/javascript" src="/js/jquery.js"></script><script type="text/javascript" src="/js/jquery.cookie.js"></script><script type="text/javascript" src="/js/sso.js"></script><script type="text/javascript" src="/js/jscal2.js"></script><script type="text/javascript" src="/js/lang/en.js"></script><script type="text/javascript" src="/js/jquery.colorbox-min.js"></script><script type="text/javascript" src="/js/eudl.js"></script><script type="text/javascript" src="/js/article.js"></script><link rel="stylesheet" type="text/css" href="/css/jscal/jscal2.css"><link rel="stylesheet" type="text/css" href="/css/jscal/eudl/eudl.css"><link rel="stylesheet" type="text/css" href="/css/colorbox.css"></head><body><div id="eudl-page-head"><div id="eudl-page-header"><section id="user-area"><div><nav id="right-nav"><a href="/about">About</a> | <a href="/contact">Contact Us</a> | <a class="register" href="https://account.eai.eu/register?service=http%3A%2F%2Feudl.eu%2Fdoi%2F10.1007%2F978-3-030-23976-3_20">Register</a> | <a class="login" href="https://account.eai.eu/cas/login?service=http%3A%2F%2Feudl.eu%2Fdoi%2F10.1007%2F978-3-030-23976-3_20">Login</a></nav></div></section></div></div><div id="eudl-page"><header><section id="topbar-ads"><div><a href="https://eudl.eu/"><img class="eudl-logo-top" src="https://eudl.eu/images/eudl-logo.png"></a><img class="eudl-ads-top" src="https://eudl.eu/images/eai-eudl.jpg"></div></section><section id="menu"><nav><a href="/proceedings" class="current"><span>Proceedings</span><span class="icon"></span></a><a href="/series" class=""><span>Series</span><span class="icon"></span></a><a href="/journals" class=""><span>Journals</span><span class="icon"></span></a><a href="/content" class=""><span>Search</span><span class="icon"></span></a><a href="http://eai.eu/">EAI</a></nav></section></header><div id="eaientran"></div><section id="content"><section id="article"><section class="cover-and-filters"><a href="https://eai.eu/eai-sponsorship/?mtm_campaign=call%20for%20bids&mtm_kwd=bids&mtm_source=organize%20conference%20page&mtm_medium=eudl"><img src="https://eudl.eu/images/banner-outside.png"></a></section><section class="info-and-search"><span class="article-proceedings-ref"><a href="/proceedings/FABULOUS/2019">Future Access Enablers for Ubiquitous and Intelligent Infrastructures. 4th EAI International Conference, FABULOUS 2019, Sofia, Bulgaria, March 28-29, 2019, Proceedings</a></span><p class="article-type">Research Article</p><h1>Hybrid Noise-Resilient Deep Learning Architecture for Modulation Classification in Cognitive Radio Networks</h1><section id="download"><a class="download-pdf" href="https://account.eai.eu/cas/login?service=http%3A%2F%2Feudl.eu%2Fdoi%2F10.1007%2F978-3-030-23976-3_20">Download</a><a class="eai-login" href="https://account.eai.eu/cas/login?service=http%3A%2F%2Feudl.eu%2Fdoi%2F10.1007%2F978-3-030-23976-3_20">(Requires a free EAI acccount)</a><div><span class="download">165 downloads</span></div></section><section class="meta-section cite-area"><dl class="main-metadata"><dt class="title" id="cite-switchers">Cite</dt> <dd class="value" id="cite-switchers"><a href="" class="bibtex">BibTeX</a> <a href="" class="plain-text">Plain Text</a></dd></dl></section><ul id="cite-blocks"><li class="bibtex"><pre>@INPROCEEDINGS{10.1007/978-3-030-23976-3_20, author={Antoni Ivanov and Krasimir Tonchev and Vladimir Poulkov and Hussein Al-Shatri and Anja Klein}, title={Hybrid Noise-Resilient Deep Learning Architecture for Modulation Classification in Cognitive Radio Networks}, proceedings={Future Access Enablers for Ubiquitous and Intelligent Infrastructures. 4th EAI International Conference, FABULOUS 2019, Sofia, Bulgaria, March 28-29, 2019, Proceedings}, proceedings_a={FABULOUS}, year={2019}, month={9}, keywords={Cognitive Radio Deep Learning Modulation classification Spectrum sensing}, doi={10.1007/978-3-030-23976-3_20} } </pre></li><li class="plain-text"><div class="cite">Antoni Ivanov<br>Krasimir Tonchev<br>Vladimir Poulkov<br>Hussein Al-Shatri<br>Anja Klein<br>Year: 2019<br>Hybrid Noise-Resilient Deep Learning Architecture for Modulation Classification in Cognitive Radio Networks<br>FABULOUS<br>Springer<br>DOI: 10.1007/978-3-030-23976-3_20</div></li></ul><section id="authors"><span class="author-list">Antoni Ivanov<sup>1</sup><sup>,*</sup>, Krasimir Tonchev<sup>1</sup><sup>,*</sup>, Vladimir Poulkov<sup>1</sup><sup>,*</sup>, Hussein Al-Shatri<sup>2</sup><sup>,*</sup>, Anja Klein<sup>2</sup><sup>,*</sup></span><ul class="affiliation-list"><li>1: Technical University of Sofia</li><li>2: Technical University Darmstadt</li></ul><section class="corresponding-email">*Contact email: astivanov@tu-sofia.bg, k_tonchev@tu-sofia.bg, vkp@tu-sofia.bg, h.shatri@nt.tu-darmstadt.de, a.klein@nt.tu-darmstadt.de</section></section><section class="full-abstract"><h2>Abstract</h2><div><p>The increasing maturity of the concepts which would allow for the operation of a practical Cognitive Radio (CR) Network require functionalities derived through different methodologies from other fields. One such approach is Deep Learning (DL) which can be applied to diverse problems in CR to enhance its effectiveness by increasing the utilization of the unused radio spectrum. Using DL, the CR device can identify whether the signal comes from the Primary User (PU) transmitter or from an interferer. The method proposed in this paper is a hybrid DL architecture which aims at achieving high recognition rate at low signal-to-noise ratio (SNR) and various channel impairments including fading because such are the relevant conditions of operation of the CR. It consists of an autoencoder and a neural network structure due to the good denoising qualities of the former and the recognition accuracy of the latter. The autoencoder aims to restore the original signal from the corrupted samples which would increase the accuracy of the classifier. Afterwards its output is fed into the NN which learns the characteristics of each modulation type and classifies the restored signal correctly with certain probability. To determine the optimal classification DL model, several types of NN structures are examined and compared for input comprised of the IQ samples of the reconstructed signal. The performance of the proposed DL architecture in comparison to similar models for the relevant parameters in different channel impairments scenarios is also analyzed.</p></div></section><section class="metas-section"><section class="meta-section"><dl class="main-metadata"><dt class="title">Keywords</dt> <dd class="value">Cognitive Radio Deep Learning Modulation classification Spectrum sensing</dd></dl></section><section class="meta-section"><dl class="main-metadata"><dt class="title">Published</dt> <dd class="value">2019-09-17</dd><dt class="title">Appears in</dt> <dd class="value"><a href="https://rd.springer.com/book/10.1007/978-3-030-23976-3">SpringerLink</a></dd></dl></section><section class="meta-section"><dl class="main-metadata"><dt class="title"></dt> <dd class="value"><a href="http://dx.doi.org/10.1007/978-3-030-23976-3_20">http://dx.doi.org/10.1007/978-3-030-23976-3_20</a></dd></dl></section></section><div class="copyright-block">Copyright © 2019–2024 ICST</div></section></section></section><div class="clear"></div><footer><div class="links"><a href="https://www.ebsco.com/" target="_blank"><img class="logo ebsco-logo" src="/images/ebsco.png" alt="EBSCO"></a><a href="https://www.proquest.com/" target="_blank"><img class="logo proquest-logo" src="/images/proquest.png" alt="ProQuest"></a><a href="https://dblp.uni-trier.de/db/journals/publ/icst.html" target="_blank"><img class="logo dblp-logo" src="/images/dblp.png" alt="DBLP"></a><a href="https://doaj.org/search?source=%7B%22query%22%3A%7B%22filtered%22%3A%7B%22filter%22%3A%7B%22bool%22%3A%7B%22must%22%3A%5B%7B%22term%22%3A%7B%22index.publisher.exact%22%3A%22European%20Alliance%20for%20Innovation%20(EAI)%22%7D%7D%5D%7D%7D%2C%22query%22%3A%7B%22query_string%22%3A%7B%22query%22%3A%22european%20alliance%20for%20innovation%22%2C%22default_operator%22%3A%22AND%22%2C%22default_field%22%3A%22index.publisher%22%7D%7D%7D%7D%7Dj" target="_blank"><img class="logo doaj-logo" src="/images/doaj.jpg" alt="DOAJ"></a><a href="https://www.portico.org/publishers/eai/" target="_blank"><img class="logo portico-logo" src="/images/portico.png" alt="Portico"></a><a href="http://eai.eu/" target="_blank"><img class="logo eai-logo" src="/images/eai.png"></a></div></footer></div><div class="footer-container"><div class="footer-width"><div class="footer-column logo-column"><a href="https://eai.eu/"><img src="https://eudl.eu/images/logo_new-1-1.png" alt="EAI Logo"></a></div><div class="footer-column"><h4>About EAI</h4><ul><li><a href="https://eai.eu/who-we-are/">Who We Are</a></li><li><a href="https://eai.eu/leadership/">Leadership</a></li><li><a href="https://eai.eu/research-areas/">Research Areas</a></li><li><a href="https://eai.eu/partners/">Partners</a></li><li><a href="https://eai.eu/media-center/">Media Center</a></li></ul></div><div class="footer-column"><h4>Community</h4><ul><li><a href="https://eai.eu/eai-community/">Membership</a></li><li><a href="https://eai.eu/conferences/">Conference</a></li><li><a href="https://eai.eu/recognition/">Recognition</a></li><li><a href="https://eai.eu/corporate-sponsorship">Sponsor Us</a></li></ul></div><div class="footer-column"><h4>Publish with EAI</h4><ul><li><a href="https://eai.eu/publishing">Publishing</a></li><li><a href="https://eai.eu/journals/">Journals</a></li><li><a href="https://eai.eu/proceedings/">Proceedings</a></li><li><a href="https://eai.eu/books/">Books</a></li><li><a href="https://eudl.eu/">EUDL</a></li></ul></div></div></div><script type="text/javascript" src="https://eudl.eu/js/gacode.js"></script><script src="/js/highlight.pack.js"></script><script>hljs.initHighlightingOnLoad();</script><script type="application/ld+json">{"@context":"http://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"item":{"@id":"http://eudl.eu","name":"Home","image":null}},{"@type":"ListItem","position":2,"item":{"@id":"http://eudl.eu/proceedings","name":"Proceedings","image":null}},{"@type":"ListItem","position":3,"item":{"@id":"http://eudl.eu/proceedings?by_acronym=\"FABULOUS\"","name":"FABULOUS","image":null}},{"@type":"ListItem","position":4,"item":{"@id":"http://eudl.eu/proceedings?by_acronym=\"FABULOUS\"/2019","name":"2019","image":null}},{"@type":"ListItem","position":1,"item":{"@id":"http://eudl.eu/doi/10.1007/978-3-030-23976-3_20","name":"Hybrid Noise-Resilient Deep Learning Architecture for Modulation Classification in Cognitive Radio Networks","image":null}}]}</script></body></html>