Modülasyon Türlerinin Hiyerarşik Sınıflandırılmasının Performans Analizi

Loading...
Thumbnail Image

Date

2020

Journal Title

Journal ISSN

Volume Title

Publisher

Open Access Color

OpenAIRE Downloads

OpenAIRE Views

Research Projects

Organizational Units

Organizational Unit
Electrical-Electronics Engineering
The Department of Electrical and Electronics Engineering covers communications, signal processing, high voltage, electrical machines, power distribution systems, radar and electronic warfare, RF, electromagnetic and photonics topics. Most of the theoretical courses in our department are supported by qualified laboratory facilities. Our department has been accredited by MÜDEK since 2013. Within the scope of joint training (COOP), in-company training opportunities are offered to our students. 9 different companies train our students for one semester within the scope of joint education and provide them with work experience. The number of students participating in joint education (COOP) is increasing every year. Our students successfully completed the joint education program that started in the 2019-2020 academic year and started work after graduation. Our department, which provides pre-graduation opportunities to its students with Erasmus, joint education (COOP) and undergraduate research projects, has made an agreement with Upper Austria University of Applied Sciences (Austria) starting from this year and offers its students undergraduate (Atılım University) and master's (Upper Austria) degrees with 3+2 education program. Our department, which has the only European Remote Radio Laboratory in Foundation Universities, has a pioneering position in research (publication, project, patent).
Organizational Unit
Department of Electrical & Electronics Engineering
Department of Electrical and Electronics Engineering (EE) offers solid graduate education and research program. Our Department is known for its student-centered and practice-oriented education. We are devoted to provide an exceptional educational experience to our students and prepare them for the highest personal and professional accomplishments. The advanced teaching and research laboratories are designed to educate the future workforce and meet the challenges of current technologies. The faculty's research activities are high voltage, electrical machinery, power systems, signal and image processing and photonics. Our students have exciting opportunities to participate in our department's research projects as well as in various activities sponsored by TUBİTAK, and other professional societies. European Remote Radio Laboratory project, which provides internet-access to our laboratories, has been accomplished under the leadership of our department with contributions from several European institutions.

Journal Issue

Events

Abstract

Otomatik modülasyon sınıflandırması (AMC), bilinmeyen bir modülasyon tipine sahip gelen modüle edilmiş bir sinyalin modülasyon tipini belirlemek için sıklıkla ihtiyaç duyulan bir yapıdır. AMC uygulamaları literatürde olabilirlik tabanlı (LB) ve özellik tabanlı (FB) yöntemler olarak iki ana başlık altında bölünmüştür. Bu tezde, FB yaklaşımı ile bir AMC algoritması geliştirilmiştir. Sınıflandırıcı olarak lineer, kuadratik ve kübik çekirdek kullanan Destek Vektör Makinesi (SVM) seçilmiş ve performansları karşılaştırılmıştır. SNR değerleri 0 ila 30 dB arasında olan havadan toplanan modüle edilmiş sinyaller kullanılmıştır. Sinyaller, yüksek derecelere kadar M-ASK, M-PSK, M-APSK içeren 12 farklı dijital modülasyon tipiyle modüle edilmiştir. İstatistiksel özellikler, yani sinyalin anlık genliği, fazı ve frekansının ortalaması, varyansı, çarpıklığı ve basıklığı, 8. dereceye kadar olan daha yüksek dereceli momentlere ve kümülanlara ek olarak kullanılmıştır. Sınıflandırıcılar arasından ikinci dereceden çekirdek kullanan SVM daha yüksek performans göstermiştir. Ayrıca, özellikle tek bir sınıflandırıcı kullanılarak sınıflandırıldığında çok düşük performans gösteren yüksek dereceli modülasyon tiplerinde, performansı arttırmak için literatüre kıyasla daha az karmaşıklığa sahip bir hiyerarşik sınıflandırma yapısı önerilmiştir. Bu modülasyonların doğruluklarında geleneksel yönteme kıyasla önemli bir gelişme gözlenmektedir. Genel performans %80'den %90'a yükselmiştir.
Automatic modulation classification (AMC) is a frequently required framework to determine the modulation type of an incoming modulated signal with an unknown modulation type. AMC applications are divided under two main titles in the literature as likelihood-based (LB) and feature-based (FB) methods. In this thesis, an AMC algorithm is developed with a FB approach. As classifier, Support Vector Machine (SVM) using linear, quadratic and cubic kernel is chosen and their performances are compared. Over-the-air collected modulated signals with the SNR values between 0 and 30 dB are used. Signals are modulated with 12 different digital modulation types containing M-ASK, M-PSK, M-APSK up to higher orders. Statistical features i.e. mean, variance, skewness and kurtosis of the instantaneous amplitude, phase and frequency of the signal are used in addition to higher-order moments and cumulants up to 8th order. SVM using quadratic kernel showed slightly higher performance. In addition, a hierarchical classification structure with less complexity compared to the literature has been proposed in order to improve performance especially in high order modulation types which show very poor performance when classified with using a single classifier. A significant improvement is observed in the accuracies of these modulations comparing with the traditional method. The overall performance is increased from 80% to 90%.

Description

Keywords

Elektrik ve Elektronik Mühendisliği, Electrical and Electronics Engineering, Sayısal modülasyon sistemleri, Digital modulation systems

Turkish CoHE Thesis Center URL

Fields of Science

Citation

WoS Q

Scopus Q

Source

Volume

Issue

Start Page

0

End Page

74