國立高雄大學圖資館 |

Design of Machine Learning Based Optimization Algorithms for Electrical Balance Duplexers /

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Design of Machine Learning Based Optimization Algorithms for Electrical Balance Duplexers /Björn Lenhart.
作者:	Lenhart, Björn,
面頁冊數:	1 electronic resource (130 pages)
附註:	Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
附註:	Advisors: Weigel, Robert; Fischerauer, Gerhard.
Contained By:	Dissertations Abstracts International86-12B.
標題:	Wireless communications.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31978660
ISBN:	9798311964739

Design of Machine Learning Based Optimization Algorithms for Electrical Balance Duplexers /
Lenhart, Björn,

Design of Machine Learning Based Optimization Algorithms for Electrical Balance Duplexers /Björn Lenhart. - 1 electronic resource (130 pages)

Source: Dissertations Abstracts International, Volume: 86-12, Section: B.

Electrical Balance Duplexers(EBD) have demonstrated remarkable capabilities in achieving high TX-to-RX isolation within Frequency-Division Duplex(FDD) applications, rivaling conventional Surface Acoustic Wave(SAW) filters commonly used in cell phones. Unlike SAW filters, EBDs offer the advantage of achieving high TX-to-RX isolation over a wider frequency range, making them more adaptable for modern communication standards.However, the inherent challenge lies in maintaining optimal isolation levels, as EBD performance is sensitive to the impedance mismatch between the antenna and the balance network. The need for rapid readjustment of the balance network to counteract impedance variations at the antenna is crucial for sustaining high TX-to-RX isolation. Traditional optimization algorithms employed for tuning EBDs often exhibit extended convergence times and susceptibility to local minima, hindering their effectiveness in dynamic environments. This work addresses these challenges by proposing innovative tuning strategies that leverage the capabilities of machine learning techniques.The thesis begins with an explanation of the fundamental principles of EBDs and a novel EBD concept, the Phase Gradient Supported Electrical Balance Duplexer(PBD), which overcomes the physical limitations of the EBD. The balancing mechanisms are derived and the effects on the tuning of the balance network are discussed. Different system modeling techniques are explained and a machine learning based model, which is the basis of the proposed tuning concept, is presented. The dissertation examines tuning concepts in detail and focuses on machine-learning based approaches, offering a thorough grasp of their capabilities and possible use cases.To validate the theoretical groundwork, a balance network is assembled and analyzed extensively to evaluate its Smith chart coverage and the effectiveness of antenna mismatch compensation. The measurement data obtained from the balance network serves as the training data for the development of a machine learning model. Utilizing neural networks, the inverse function of the balance network is learned to serve as an inverse model of the balance network.The presented methods achieve significant initial isolation (average isolation >40 dB) and more than 55 dB in less than 5 iterations of retuning for LTE band 8 (10 MHz bandwidth, 45 MHz duplex spacing). The performance of the algorithms is limited only by Smith chart coverage of the balance network and training data accuracy, both of which can be optimized offline. In realistic simulations with a dynamic antenna impedance, the proposed concepts maintain high isolation by adjusting the balance network to the antenna impedance. The results were compared with conventional balanced network optimization algorithms and demonstrate superior performance with less overshooting, higher average and worst-case isolation.

English

ISBN: 9798311964739Subjects--Topical Terms:

1010303
Wireless communications.

Design of Machine Learning Based Optimization Algorithms for Electrical Balance Duplexers /
LDR:07286nmm a22003853i 4500 001 693689
005 20260511063012.5
006 m o d
007 cr|nu||||||||
008 260819s2025 miu||||||m |||||||eng d
020 $a 9798311964739
035 $a (MiAaPQD)AAI31978660
035 $a (MiAaPQD)FAUErlangenFAU_35358_
035 $a AAI31978660
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Lenhart, Björn, $e author. $3 1010301
245 1 0 $a Design of Machine Learning Based Optimization Algorithms for Electrical Balance Duplexers / $c Björn Lenhart.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2025
300 $a 1 electronic resource (130 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
500 $a Advisors: Weigel, Robert; Fischerauer, Gerhard.
502 $b Ph.D. $c Friedrich-Alexander-Universitaet Erlangen-Nuernberg (Germany) $d 2025.
520 $a Electrical Balance Duplexers(EBD) have demonstrated remarkable capabilities in achieving high TX-to-RX isolation within Frequency-Division Duplex(FDD) applications, rivaling conventional Surface Acoustic Wave(SAW) filters commonly used in cell phones. Unlike SAW filters, EBDs offer the advantage of achieving high TX-to-RX isolation over a wider frequency range, making them more adaptable for modern communication standards.However, the inherent challenge lies in maintaining optimal isolation levels, as EBD performance is sensitive to the impedance mismatch between the antenna and the balance network. The need for rapid readjustment of the balance network to counteract impedance variations at the antenna is crucial for sustaining high TX-to-RX isolation. Traditional optimization algorithms employed for tuning EBDs often exhibit extended convergence times and susceptibility to local minima, hindering their effectiveness in dynamic environments. This work addresses these challenges by proposing innovative tuning strategies that leverage the capabilities of machine learning techniques.The thesis begins with an explanation of the fundamental principles of EBDs and a novel EBD concept, the Phase Gradient Supported Electrical Balance Duplexer(PBD), which overcomes the physical limitations of the EBD. The balancing mechanisms are derived and the effects on the tuning of the balance network are discussed. Different system modeling techniques are explained and a machine learning based model, which is the basis of the proposed tuning concept, is presented. The dissertation examines tuning concepts in detail and focuses on machine-learning based approaches, offering a thorough grasp of their capabilities and possible use cases.To validate the theoretical groundwork, a balance network is assembled and analyzed extensively to evaluate its Smith chart coverage and the effectiveness of antenna mismatch compensation. The measurement data obtained from the balance network serves as the training data for the development of a machine learning model. Utilizing neural networks, the inverse function of the balance network is learned to serve as an inverse model of the balance network.The presented methods achieve significant initial isolation (average isolation >40 dB) and more than 55 dB in less than 5 iterations of retuning for LTE band 8 (10 MHz bandwidth, 45 MHz duplex spacing). The performance of the algorithms is limited only by Smith chart coverage of the balance network and training data accuracy, both of which can be optimized offline. In realistic simulations with a dynamic antenna impedance, the proposed concepts maintain high isolation by adjusting the balance network to the antenna impedance. The results were compared with conventional balanced network optimization algorithms and demonstrate superior performance with less overshooting, higher average and worst-case isolation.
520 3 $a Electrical Balance-Duplexer (EBDs) haben sich in Frequenzduplex-Anwendungen als wirksame Schaltung zum Erreichen hoher Isolationen zwischen Sende- und Empfangspfad erwiesen und konkurrieren dabei mit herkömmlichen akustischen Oberflächenwellenfiltern (SAW), die üblicherweise in Mobiltelefonen verwendet werden. Im Gegensatz zu SAW-Filtern bieten EBDs den Vorteil, eine hohe Isolation über einen breiteren Frequenzbereich zu erreichen, was sie geeigneter für moderne Kommunikationsnetze macht.Die durchschnittliche Isolation ist jedoch stark von der Antennenimpedanz abhängig und erfordert eine schnelle Nachjustierung des Balance-Netzwerks, um den EBD im balancierten Zustand mit hoher Isolation zu halten. Herkömmliche Optimierungsalgorithmen, die zur Abstimmung von EBDs eingesetzt werden, zeigen oft lange Konvergenzzeiten und können zu lokalen Minima führen, was ihren Einsatz in dynamischen Umgebungen beeinträchtigt. Diese Arbeit begegnet diesen Herausforderungen, indem sie innovative Tuning-Konzepte vorschlägt, die auf maschinellem Lernen basieren.Die Arbeit beginnt mit den Grundlagen von EBDs und einem neuartigen Konzept für EBDs, dem Phase Gradient Supported Electrical Balance Duplexer(PBD), der die physikalischen Grenzen der EBDs überwindet. Die Balancing-Mechanismen werden abgeleitet und die Auswirkungen auf die Abstimmung des Balance-Netzwerks werden diskutiert. Verschiedene Systemmodellierungstechniken werden erläutert und ein auf maschinellem Lernen basierendes Modell wird vorgestellt. Diese Dissertation untersucht darauf aufbauende Tuning-Konzepte im Detail und zeigt mögliche Anwendungsfälle.Zur Validierung der theoretischen Grundlagen wird ein Balance-Netzwerk aufgebaut und umfassend analysiert, um die Abdeckung im Smith-Diagramm und die Wirksamkeit der Kompensation von Antennenfehlanpassungen zu bewerten. Die aus dem Balance-Netzwerk gewonnenen Messdaten dienen als Trainingsdaten für die Entwicklung eines maschinellen Lernmodells. Mithilfe neuronaler Netze wird die Umkehrfunktion des Balance-Netzwerks erlernt, um als inverses Modell des Balance-Netzwerks zu dienen.Die vorgestellten Methoden erreichen eine signifikante anfängliche Isolation (mittlere Isolation >40 dB) und mehr als 55 dB in weniger als 5 Iterationen für das LTE-Band 8 (10 MHz Bandbreite, 45 MHz Duplex-Abstand). Die Leistung der Algorithmen ist nur durch die Abdeckung des Balance-Netzwerks im Smith-Diagramm und die Genauigkeit der Trainingsdaten begrenzt, die beide offline optimiert werden können. In realistischen Simulationen mit einer dynamischen Antennenimpedanz halten die vorgeschlagenen Konzepte eine hohe Isolation aufrecht, indem das Balance-Netzwerk kontinuierlich an die Antennenimpedanz angepasst wird. Die Ergebnisse werden mit konventionellen Optimierungsalgorithmen für Balance-Netzwerke verglichen und zeigen eine wesentlich bessere Leistung mit weniger Überschwingen und einer höheren mittleren und Worst-Case-Isolation.
546 $a English
590 $a School code: 0575
650 4 $a Wireless communications. $3 1010303
650 4 $a Receivers & amplifiers. $3 1010304
650 4 $a Network management systems. $3 1010305
650 4 $a Signal to noise ratio. $3 1010306
650 4 $a Radio frequency. $3 182476
650 4 $a Bandwidths. $3 966959
650 4 $a Neural networks. $3 915594
650 4 $a Signal processing. $3 182382
650 4 $a Antennas. $3 1010307
650 4 $a Communications networks. $3 915632
650 4 $a Transmitters. $3 1010308
650 4 $a Transceivers. $3 1010309
650 4 $a Spectrum allocation. $3 1010310
650 4 $a Communications systems. $3 1010311
650 4 $a Optimization algorithms. $3 1010312
650 4 $a Surface acoustic waves. $3 1010313
650 4 $a Electrical engineering. $3 454503
690 $a 0544
710 2 $a Friedrich-Alexander-Universitaet Erlangen-Nuernberg (Germany). $e degree granting institution. $3 1010302
720 1 $a Weigel, Robert $e degree supervisor.
720 1 $a Fischerauer, Gerhard $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-12B.
790 $a 0575
791 $a Ph.D.
792 $a 2025
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31978660