Η διαμόρφωση και η πολύπλεξη είναι από τα πιο σημαντικά τμήματα των συστημάτων ψηφιακής μετάδοσης και στόχος τους είναι να επιτύχουν την αποτελεσματική χρήση του καναλιού. Η τεχνική OFDM είναι μια μέθοδος διαμόρφωσης και πολύπλεξης για τη μετάδοση με πολλαπλές φέρουσες σε χρονικώς και συχνοτικώς επιλεκτικά κανάλια. Προσφέρει μεγαλύτερη ανοσία στη Διασυμβολική Παρεμβολή (ISI) και μπορεί να υλοποιηθεί εύκολα με χρήση του γρήγορου μετασχηματισμού Fourier. Η παρούσα διπλωματική εργασία στοχεύει στην καλή κατανόηση και παρουσίαση του ασύρματου καναλιού, του ισοδύναμου μοντέλου βασικής ζώνης του, καθώς και των φυσικών παραμέτρων, βάσει των οποίων κατηγοριοποιούμε τα ασύρματα συστήματα. Εξαιτίας των διαλείψεων πολυδιόδευσης, που δεν μπορούν να αποφευχθούν στα ασύρματα συστήματα, η τεχνική OFDM είναι περισσότερο κατάλληλη για αυτά τα συστήματα, παρά για τα ενσύρματα. Στη συνέχεια, προσομοιώνεται ένα σύστημα OFDM για διάφορα είδη καναλιών. Συγκεκριμένα, αρχικά θεωρείται ως περιβάλλον μετάδοσης το κανάλι AWGN και στη συνέχεια, το συχνοτικώς επιλεκτικό, σταθερό κανάλι. Στην επόμενη προσομοίωση χρησιμοποιήθηκε (συχνοτικώς επιλεκτικό) σταθερό κανάλι με εκθετική κρουστική απόκριση. Στις δύο τελευταίες προσομοιώσεις θεωρήσαμε κανάλι Rayleigh επίπεδης διάλειψης και ένα είδος συχνοτικώς επιλεκτικού καναλιού με διαλείψεις Rayleigh, αντίστοιχα. / Modulation and multiplexing are between the most important parts of a digital transmission system and their goal is to achieve an efficient use of the channel. Orthogonal Frequency Division Multiplexing (OFDM) is both a modulation and multiplexing method for multicarrier transmission through time and frequency selective channels. It offers a greater immunity to Inter-Symbol Interference (ISI) and can be easily implemented using the fast Fourier transform (FFT). This Diploma thesis aims at the interpretation and the presentation of wireless channel and OFDM technique, in detail. Initially, we described the wireless channel, its baseband equivalent, and the physical parameters that are used to classify the different types of it. Because of the multipath fading which is unavoidable in wireless systems, OFDM is more appropriate for these ones than for wire systems. Then, we simulate an OFDM system. The simulations take place in several types of (wireless) channel. Especially, we firstly considered an AWGN channel and then a frequency-selective, non-fading channel. We also used an exponential frequency-selective, non-fading channel. For the two last simulations we considered the one-tap (flat), Rayleigh fading channel, and a type of frequency-selective, fading channel.

<p> Conventional method for testing the total power loss and internal temperature rise of induction motors under full load can often be a complex and costly process. The new phase modulation method for synthetic full-load testing of induction machines has been proven viable, provided the power oscillation in this method can be minimized. This thesis explores two techniques for stabilizing power fluctuation in the new method, and determines the test system's power sensitivities to parameter changes in the equivalent circuit of the induction motors under test.</p> <p> A computer simulation representing the test system used in the new phase modulation method was developed, and an experimental testing facility was built to test the technique devised for suppressing power oscillation in the test system.</p> <p> The results from simulations are analyzed and compared to those obtained from the actual experiments in order to identify the feasible power-suppressing technique, and determine the induction machine parameters responsible for causing power unbalance in the test system.</p> / Thesis / Master of Applied Science (MASc)

Due to the increased consumption of energy in the modern era, many new installations of different energy solutions are required. One of the more preferred is the renewable category, which governs vast technologies ranging from wind, solar, hydro, and nuclear power. One major issue is that they are usually located up to thousands of kilometers away from the load source. Thus, a reliable transmission system is a must. For such applications, a high-voltage direct current power transmission system (HVDC) is the favorable option, most of the modern types use a modular multilevel voltage source converter (MMC) to execute the AC/DC conversion and vice versa. The control of such devices is ultimately dependent on the modulation method, it serves as a key component of the stability of the whole system. Four different variants of description for such a control method are found in the published literature, as more methods can be found, but the focus is on the four introduced. Therefore, the aim of this master's thesis is to investigate how the highlighted standard modulation methods affect the total system's behavior. Note that the implementation by the company, referred to in this context as the benchmark method, is being compared to assess its efficacy against established standard modulation methods. Detailed comparison involving this method is intentionally omitted. First, the underlying system components are described using circuit topology, which serves as the first course before the analytical derivation using dynamic modeling. By obtaining the final dynamic expression and equation for the insertion index, a detailed control strategy is displayed. The total control system is separated into two parts, one governed by the higher-level controllers and one governed by the modulation methods. As the former is predefined and given, therefore, it will be referred to as the "black box." Then a comprehensive method regarding the implementation of modulation methods is provided, serving as the most crucial part of the work. Several different studies are conducted, including harmonic analysis, fault studies, impedance scanning, and screening studies of torsional interaction. The result indicated that some degree of differences can be concluded between the modulation methods, several cases showed that controlling done in a closed loop manner is the preferred option due to its accuracy and robustness. Some of the investigated modulation methods showed severe oscillations in circumstances where the connected grids were weak. Due to the complexity of the overall system, including higher-level controllers, no direct conclusion can be drawn to definitively say whether any of the modulation methods have a clear advantage in every tested aspect. But at least, based on the results obtained in this work, in this specific simulation set-up, the closed loop method is indeed the preferred option.

Isolated inverters feature the freedom in voltage step-up/down, electrical safety, and modularity. Among them, pseudo-dc-link inverters have the advantage of high efficiency due to their single-stage structure. Traditionally, pseudo-dc-link inverters are based on pulse-width-modulated converters, which suffer from hard switching, the need for auxiliary components, and/or high current stresses. Meanwhile, the series resonant converter has been prevalent in past decades due to its simplicity and high efficiency. Therefore, it is intriguing to design a single-stage inverter based on a series resonant converter. However, there are limited papers regarding such an inverter topology. To figure out the reason, basic modulation methods proposed or implied in the literature are summarized and evaluated through circuit simulation software. It turns out each basic modulation method has at least one critical drawback in modulation range, hard switching, and/or high current stresses. Given the deficiencies in the basic modulation methods, a hybrid modulation method is proposed here. The proposed method combines variable-frequency modulation in the high-output region and short pulse-density modulation in the low-output region. In this way, all the aforementioned critical drawbacks can be greatly alleviated. The hybrid modulation method is compared to the basic modulation methods based on three design metrics: the rms value of the resonant current, the magnetic flux of the transformer, and the turn-off current. By these design metrics that directly related to power losses, the benefit of the proposed method in terms of efficiency can be explained. Moreover, a power loss model is also established to provide more insights into the inverter's efficiency performance. It helps demonstrate how the selection of resonant tank and other factors affects the power loss distribution. Also, an inverter design procedure is introduced and a prototype is built to verify the proposed modulation method. The results show that the switching losses, especially the turn-on loss, can be well suppressed, and the losses in other passive components are well restrained. This implies the proposed method is suitable for high-frequency applications. Other than efficiency, output waveform quality is also important for an inverter. However, the changing plant model makes the controller design difficult. Therefore, a third-order model established by other researchers has been adopted to identify the pole locations. In addition, a gain-varying method is proposed for the compensator to reduce the gain variance caused by different operating conditions. The experimental results show that without the gain-varying method, the inverter may have issues in slow tracking and/or instability. Finally, in some scenarios, the inverter based on a series resonant converter can be regarded as a module. A multi-modular inverter can be formed by connecting the modules in an input-parallel-output-series configuration. In this case, a technique termed sequential waveform synthesis can be applied. The proposed technique can extend the region of variable-frequency modulation and shorten the region of short pulse-density modulation. This is beneficial to efficiency based on an analysis. With more than a certain amount of modules connected, the short pulse-density modulation can even be waived, which means the multi-modular inverter can be free from turn-on loss. In summary, this dissertation focuses on developing modulation methods for inverters based on the series resonant converter. Soft-switching feature and high efficiency are the two top priorities. The analytic and experimental results are provided based on standalone applications. / Doctor of Philosophy / Inverters are an important part of a modern electric power system, as they convert dc electric power into ac electric power. In some applications, inverters with electrical insulation (isolated inverters) are preferred due to the need for engineering freedom, safety, and other reasons. However, each conventional isolated inverter has some of the following drawbacks: hard-switching in semiconductor devices, high circulating current, poor transformer utilization, and high complexity. These drawbacks limit the efficiency and compactness of an inverter system, making the system less attractive to practical applications. An inverter based on a series resonant converter seems to be a solution because the series resonant converter is known for being simple and highly-efficient. However, there has yet to be a proper modulation method for it. Therefore, the main contribution of this dissertation is to propose a hybrid modulation method. With the proposed method, the inverter can operate with high efficiency. Furthermore, the hard-switching can be well suppressed, which means a high-frequency, compact design is possible. Besides the theory of the proposed method, this dissertation also includes a power loss model, a hardware design procedure, and analytic comparisons with other methods. In addition, a digital approach to control the inverter is proposed. Without it, the output voltage waveform may be highly distorted. Finally, another sequential control strategy is proposed in this dissertation for an integrated system. The integrated system is composed of multiple inverters based on a series resonant converter. With the sequential control strategy, the overall output waveform quality of the integrated system can be improved.

Abstract This thesis deals with the fascinating subject of the design of digital communication systems - or more precisely one topic therein, i.e., modulation. Modulation and its counterpart demodulation are the means of making the information ride the beast of a wireless radio channel. The introduction of mMFSK and amMFSK generated ideas of applying the approach to other modulation methods. The straightforward extension was to apply the method to CSK modulation to introduce mMCSK modulation. The analysis shows that due to the orthogonality of the signaling waveforms of mMCSK (with Walsh codes) and mMFSK, the same performance is achieved for modulation methods with the same modulation alphabet. But with CSK it is rather easy to construct non-orthogonal signaling waveforms. Hence, the thesis also gives analytical results for non-orthogonal mMCSK and especially considers (as an example) the use of Gold and Kasami codes. The results indicate that the best choice of codes for a non-coherent mMCSK system is the orthogonal code family, which is a rather intuitive result. However, for a coherent mMCSK system, better performance could be achieved with non-orthogonal codes. Given that we can construct a code set where the cross-correlation between the codes in the family is negative (&#961;i,j&#60;0), performance improvement compared to the orthogonal code family is achieved. The results show that, for instance, a 0.5 dB performance improvement in the AWGN channel can be achieved by using a specifically constructed set of Kasami codes as the basis functions in the modulation. The thesis also presents the performance analysis results of mMCSK in a flat Rayleigh fading channel. To further enlarge the modulation alphabet sizes (with the price of larger spectrum usage) it was realized that combining the mMFSK and mMCSK would be an interesting choice. The mMCSK-mMFSK modulation was hence introduced, which can be viewed as the main contribution of this thesis. A method to analyze this two-component modulation was developed and the performance analyses give results for mMCSK-mMFSK modulation in AWGN and flat Rayleigh fading channels for both coherent and non-coherent receivers. The performance was also assessed with orthogonal and non-orthogonal code constructs. Moreover, an antipodal extension of the mMCSK-mMFSK modulation was introduced with the analysis of the performance. A third contribution of the thesis was to introduce the mPPM modulation method following the mMFSK principles. The performance results obtained in the mMCSK can also be applied to these modulation formats in certain scenarios. PPM modulation has been widely suggested to be used in UWB systems. In UWB systems, the means to adapt the data rate is the use of pulse repetition. Hence, the performance analysis of the mPPM modulation methods with pulse repetition and a non-coherent receiver was computed. The performance of pulse repetition in impulsive interference was also computed. / Tiivistelmä Tässä väitöskirjatyössä tarkastellaan tietoliikennejärjestelmän suunnitteluun liittyvää osa-aluetta modulaatiomenetelmä. Modulaatio ja sen vastinpari demodulaatio ovat mekanismit, joiden avulla siirrettävä tieto saatetaan muotoon, jolla se voidaan siirtää radiokanavassa. mMFSK ja amMFSK modulaatiomenetelmien kehittäminen loi pohjan, jota lähdettiin soveltamaan myös muihin modulaatiotekniikoihin. Suoritetut matemaattiset analyysit osoittavat, että samalle modulaatioaakkoston koolle mMCSK (käyttäen Walsh-koodeja) ja mMFSK saavuttavat saman suorituskyvyn olettaen modulaatioiden olevan ortogonaalisia. CSK-tekniikalla on kuitenkin helppo rakentaa koodeja, jotka ovat epäortogonaalisia. Väitöskirjassa analysoidaan myös muutamia tapauksia epaortogonaaliselle mMCSK:lle, jossa esimerkinomaisesti tarkastellaan Gold- ja Kasami-koodien käyttöä. Tulokset osoittavat, että epäkoherentille mMCSK:lle paras valinta koodiperheeksi on ortogonaalinen koodiperhe, joka tuloksena on myös intuitiivinen. Koherentille mMCSK:lle voidaan kuitenkin rakentaa epäortogonaalisia koodikonstruktioita, joissa koodien välinen ristikorrelaatio on negatiivinen ja tällöin saavutetaan suorituskyvyn parantumista ortogonaaliseen koodiperheeseen verattuna. Tulokset Gold- ja Kasami-koodeille AWGN-kanavassa osoittavat, että n. 0.5 dB suorituskyvyn paraneminen on saavutettavissa. Työssä johdetaan myös analyyttiset tulokset vastaaville tapauksille Rayleigh-häipyvässä kanavassa. Tutkittaessa mMCSK- ja mMFSK-modulaatioita, huomattiin, että yhdistämällä nämä kaksi voidaan saavuttaa erittäin suuria modulaatioaakkostoja. Hintana on suurempi spektrin käyttö. mMCSK-mMFSK-modulaation kehittämistä voidaan pitää tämän työn päätuloksena. Työssä kehitetään likimääräinen menetelmä tämän kaksi-komponenttisen modulaation suorituskyvyn analysoimiseksi. Tuloksina esitetään modulaatiomenetelmän suorituskyky sekä AWGN- että Rayleigh-häipyvässä kanavassa. Suorituskykytuloksia esitetään myös epäortogonaalisille koodikonstruktioille. Lisäksi kehitettiin antipodaalinen laajennus mMCSK-mMFSK-modulaatiosta ja suoritettiin suorituskykyanalyysi AWGN-kanavassa. Työn kolmantena tuloksena esitetään mPPM-modulaatiomenetelmä hyödyntäen mMCSK-modulaation periaatteita. mMCSK-suorituskykyanalyysia voidaan tietyin ehdoin soveltaa myös näihin modulaatiomenetelmiin, joiden käyttöä esitetään usein ultralaajakaistajärjestelmissä (UWB). UWB-järjestelmissä datanopeutta säädetään usein pulssin toistoa käyttämällä. Työssä johdetaan suorityskykytulokset epäkoherentille mPPM-vastaanottimelle myös pulssintoiston yhteydessä. Lisäksi johdettiin suorituskykylausekkeet ko. modulaatioille impulsiivisen häiriön läsnäollessa.