Documenti full-text disponibili:
![[img]](http://amsdottorato.unibo.it/style/images/fileicons/application_pdf.png) |
Documento PDF (English)
- Richiede un lettore di PDF come Xpdf o Adobe Acrobat Reader
Disponibile con Licenza: Salvo eventuali più ampie autorizzazioni dell'autore, la tesi può essere liberamente consultata e può essere effettuato il salvataggio e la stampa di una copia per fini strettamente personali di studio, di ricerca e di insegnamento, con espresso divieto di qualunque utilizzo direttamente o indirettamente commerciale. Ogni altro diritto sul materiale è riservato.
Download (37MB)
|
Abstract
The objective of the thesis is to analyze the possibility of exploiting the additional degrees of freedom in multi-input (MI) power amplifier (PA) topologies to improve their trade-offs among the main figures of merit: linearity, efficiency, and power. The proposed algorithms were tested on three different MI PA topologies, each presenting different optimization challenges and objectives.
On a PA array, a beam-dependent digital predistortion (BD-DPD) algorithm was tested to enhance the linearity of the system under varying PA load conditions, by adjusting the beam direction. The method is based on machine learning and allows a low-complexity real-time update of the DPD coefficients by exploiting feature-based model reduction. The validation is performed through over-the-air measurements of a 1x4 array operating at 28 GHz with 100-MHz modulated BW.
On a supply-modulated PA (SM-PA), a gradient-based multi-objective optimization (MOO) algorithm was applied to optimize the trade-off between linearity and efficiency of the PA. To avoid dealing with an unbearably high number of experimental acquisitions, MOO is made feasible by fast simulation of an empirical surrogate model of the PA, which is progressively refined from a reduced set of iterative acquisitions. The method outperforms classical SM approaches. This is demonstrated by the experimental results on a SM-PA operating at 3.5 GHz in the presence of OFDM-like high-PAPR modulated signals with 10-MHz and 20-MHz BWs.
Finally, a derivative-free algorithm, the Bayesian optimization, was tested on a multi-input load-modulated amplifier, specifically on a dual-input Doherty PA (DIDPA) targeting 100-MHz modulated BW at 24 GHz. The optimization focuses on improving power added efficiency (PAE) while maintaining linearity with DPD linearization. A joint optimization considers the mutual effect of DPD linearization and PAE-maximization. Validation is done with a multiport on-wafer measurement system using a vector network analyzer for wideband characterization of a millimeter-wave DIDPA.
Abstract
The objective of the thesis is to analyze the possibility of exploiting the additional degrees of freedom in multi-input (MI) power amplifier (PA) topologies to improve their trade-offs among the main figures of merit: linearity, efficiency, and power. The proposed algorithms were tested on three different MI PA topologies, each presenting different optimization challenges and objectives.
On a PA array, a beam-dependent digital predistortion (BD-DPD) algorithm was tested to enhance the linearity of the system under varying PA load conditions, by adjusting the beam direction. The method is based on machine learning and allows a low-complexity real-time update of the DPD coefficients by exploiting feature-based model reduction. The validation is performed through over-the-air measurements of a 1x4 array operating at 28 GHz with 100-MHz modulated BW.
On a supply-modulated PA (SM-PA), a gradient-based multi-objective optimization (MOO) algorithm was applied to optimize the trade-off between linearity and efficiency of the PA. To avoid dealing with an unbearably high number of experimental acquisitions, MOO is made feasible by fast simulation of an empirical surrogate model of the PA, which is progressively refined from a reduced set of iterative acquisitions. The method outperforms classical SM approaches. This is demonstrated by the experimental results on a SM-PA operating at 3.5 GHz in the presence of OFDM-like high-PAPR modulated signals with 10-MHz and 20-MHz BWs.
Finally, a derivative-free algorithm, the Bayesian optimization, was tested on a multi-input load-modulated amplifier, specifically on a dual-input Doherty PA (DIDPA) targeting 100-MHz modulated BW at 24 GHz. The optimization focuses on improving power added efficiency (PAE) while maintaining linearity with DPD linearization. A joint optimization considers the mutual effect of DPD linearization and PAE-maximization. Validation is done with a multiport on-wafer measurement system using a vector network analyzer for wideband characterization of a millimeter-wave DIDPA.
Tipologia del documento
Tesi di dottorato
Autore
Mengozzi, Mattia
Supervisore
Co-supervisore
Dottorato di ricerca
Ciclo
36
Coordinatore
Settore disciplinare
Settore concorsuale
Parole chiave
Digital Predistortion, Radio Frequency Power Amplifiers, Optimization, Beamforming, Doherty, Supply Modulation, Gallium Nitride, Microwave Instrumentation and Measurements.
URN:NBN
DOI
10.48676/unibo/amsdottorato/11527
Data di discussione
9 Luglio 2024
URI
Altri metadati
Tipologia del documento
Tesi di dottorato
Autore
Mengozzi, Mattia
Supervisore
Co-supervisore
Dottorato di ricerca
Ciclo
36
Coordinatore
Settore disciplinare
Settore concorsuale
Parole chiave
Digital Predistortion, Radio Frequency Power Amplifiers, Optimization, Beamforming, Doherty, Supply Modulation, Gallium Nitride, Microwave Instrumentation and Measurements.
URN:NBN
DOI
10.48676/unibo/amsdottorato/11527
Data di discussione
9 Luglio 2024
URI
Statistica sui download
Gestione del documento: