This thesis presents an overview of the simulation, design, and measurement of state-of-the-art Silicon-Germanium Heterojunction Bipolar Transistor (SiGe HBT) variable gain amplifier (VGA). The VGA design trade-off space is presented and methods for achieving an optimized design are discussed. We demonstrate in this thesis that SiGe HBT VGA has the capability to meet the demanding needs for the next generation wireless systems.

Today s supercomputers require interconnection networks with high bandwidth and low latency to exploit parallelism. The data vortex is an all optical path interconnection network defined and then proven to achieve high level of message acceptance and low levels of message latency. In this thesis research, three enhancements to the data vortex are defined and tested for performance. They are compared to an unmodified data vortex using the average latency and offered traffic acceptance rates as metrics. Minimal angle counts are established where express lane enhancements are established. An express lane enhancement allows exploitation of locality yielding an 8% to 12 % reduction in average latency and a 4% to 6% increase in message acceptance. Semi-Express lanes cannot effectively exploit locality but still yield a 20% increase in message acceptance and a 4% decrease in average latency. Express outputs can exploit locality for a 28% to 32% increase in message acceptance and 12% to 15% decrease in average latency.

This work introduces fundamental knowledge of EMI, and presents three basic features correlated to EMI susceptibility: nonlinear distortion, asymmetric slew rate (SR) and parasitic capacitance. Different existing EMI-resisting techniques are analyzed and compared to each other in terms of EMI-Induced input offset voltage and other important specifications such as current consumption. In this work, EMI-robust analog circuits are proposed, of which the architecture is based on source-buffered differential pair in the previous publications. The EMI performance of the proposed topologies has been verified within a test IC which was fabricated in NCSU 0.5um CMOS technology. Experimental results are presented when an EMI disturbance signal of 400mV and 800mV amplitude was injected at the input terminals, and compared with a conventional and an existing topology. The tested maximal EMI-induced input offset voltage corresponds to -222mV for the new structure, which is compared to -712mV for the conventional one and -368mV for the one using existing source-buffered technique in literature. Furthermore the overall performances of the circuits such as current consumption or input referred noise are also provided with the corresponding simulation results.

With the rapid and ongoing developments in telecommunication and electronic warfare technology, faster and more flexible systems are in demand. Wideband signal processing is thus needed to implement such systems. Microwave photonics has been introduced as a tool for achieving such ultra broadband signal processing. Instantaneous Frequency Measurement (IFM) receivers play an important role in electronic warfare. They have been developed as a means of obtaining a rapid indication of the presence of a threat and to roughly identify the frequency of the threat signals. They also have the advantages of low-cost, compactness and moderate to good sorting capability in an interference-free environment. The main limitation of the traditional RF IFM receivers is constrained bandwidth. Microwave Photonic IFMs have been considered, but the main disadvantages of photonic realization of the recent IFM receiver is cost. This work aims to propose and demonstrate low-cost photonic IFM receivers with a broad frequency measurement range. The proposed methods are based on the use of photonic mixing to down-convert the RF modulated optical signals to DC. In a RADAR warning receiver, usually a bank of IFMs is required. Increasing the numbers of IFMs requires an increase in the number of photo-detectors. Thus if low-frequency, low-cost detectors can be used, then the net system cost will be reduced significantly. The concept is proven and the issues arising are analyzed. In the proof of concept system, measurement of the RF frequency required advance knowledge of the RF power. Secondly, the use of co-axial RF cables as delay elements limited the bandwidth and increased bulk. Using a photonic hybrid approach to achieve orthogonal measurements was demonstrated as a means of dentifying both RF frequency and power simultaneously and independently. Employing all optical mixing removed the need for co-axial RF cables delays using non-linear optical devices such as Semiconductor Optical Amplifier (SOA) and Highly Non-Linear Fiber (HLNF). The last investigation is to improve the sensitivity of the implemented IFM system. The sensitivity of the implemented system is characterized first and a lock-in technique is employed to improve the sensitivity of the system. The final system achieves a sensitivity of -41 dBm which is comparable with the traditional RF IFM receivers.

There has been a rapid growth of terrestrial and satellite communications in the last few decades of the 20th century. This has resulted in a heavy congestion of low microwave bands and has been a major driving force for exploring the upper microwave and millimeter-wave frequencies. One of the main requirements for a successful shift to the new frequency spectrum is the availability of high power solid-state transmitters. Solid-state devices such as diodes or transistors have been able to meet such demands when their output signals are combined using space-level power combining methods that avoid conduction losses, which become pronounced at millimeter wave frequencies. In this thesis, theoretical and experimental investigations are carried out into the spatial power combiners (SPCs) which employ active planar arrays formed by transistor amplifiers whose input and output ports are equipped with planar radiating elements. The SPC structures include the reflection-type combiner using the tile configuration of planar array and the transmission-type combiner using tile or tray configurations of planar arrays. The frequency bands chosen for the designing and testing of prototypes are X- and Ku-band. The first stage of the investigation concerns the 10 GHz reflection-type power combiner structure formed by a phased planar microstrip reflectarray (MRA) of 37-element dual-feed aperture coupled microstrip patch antennas equipped with open-circuit stubs as phasing components. The experimental tests reveal poor radiation performance and hence poor power combining efficiency of this structure. These results indicated the need for theoretical investigations into the operation of this type of SPC. The study of the unit cell of this power combiner reveals that the phase of an open-circuit stub does not increase linearly as a function of the stub length and its range is limited to less than (about is required for proper functioning). This finding, forms the basis for extending the investigations into alternative phasing mechanisms of a MRA which would offer a phasing range exceeding . A phasing mechanism exploiting variable size stacked patches is chosen. In order to accurately determine the phasing of the reflected wave, a theory based on an equivalent unit cell waveguide approach (WGA) is proposed and developed. The proposed theory is computationally efficient and is proven to be accurate compared with benchmark results published by other researchers. Following the verification, an offset feed 161-element two-layer printed MRA prototype with patches of variable size is designed and developed for operation in Ku-band. The test results aim at verifying the validity of applying a unit cell WGA to designing passive and active MRAs. The next investigations, which are presented in the thesis concern increasing operational bandwidth of the transmission-type SPC in tile configuration. The designs presented so far in the open literatures were based on edge-feed microstrip patch antennas as radiating elements of individual active stages and featured a narrow-band performance. In order to overcome this shortcoming, stacked patch (SP) microstrip antennas as receiving and transmitting elements in an active transmitarray (TXA) are proposed. For the aim of testing the proposed concept, a 16-element SP TXA is designed for operation in X-band. Two identical hard horn antennas with an approximately uniform field across the aperture for signal launching and collecting complete the design and development of this space-level power combiner. The performance of the developed device is assessed experimentally and an increased operational bandwidth is demonstrated. The final structure being investigated in the thesis project is the transmission-type SPC in tray configuration. This power combining structure employs a travelling wave antenna of uniplanar quasi-Yagi type as a radiating element to achieve broad-band operation. The investigated SPC is formed by seven trays of uniplanar quasi-Yagi antenna. In order to achieve uniform and in-phase excitation of individual trays, which is required to obtain high power combining efficiency, hard horn antennas and Schiffman phase shifters are employed in the design of this space-level combiner. The proposed device is developed and its performance is assessed through experiments. The work performed as part of this Ph.D. thesis project has resulted in 5 journal papers and 11 refereed conference papers. This acceptance rate supports the claim of the originality and significance of the research undertaken as part of the thesis project.

Thesis (M.S.)--Auburn University, 2008. / Abstract. Vita. Includes bibliographical references (p. 45-46).

Το αντικείμενο της παρούσης διπλωματικής επικεντρώνεται στη σχεδίαση ανάπτυξη και μέτρηση κυκλώματος κάτω μεταλλάκτη (down converter) συχνότητας στην περιοχή 1 έως 6 GHz. Η αρχιτεκτονική περιλαμβάνει ενισχυτή χαμηλού θορύβου (LNA) κατανεμημένης τοπολογίας (distributed amplifier), μίκτη καθώς και ενισχυτές και φίλτρα στην ενδιάμεση συχνότητα. Ο σχεδιασμός συνοδεύεται από μετρήσεις όπου και διαπιστώνεται η σύγκλιση με τα αποτελέσματα εντατικών εξομοιώσεων. Κύρια εργαλεία του σχεδιασμού απετέλεσαν κυκλωματικοί και ηλεκτρομαγνητικοί εξομοιωτές. / The object of this thesis focuses on design development and measurement down converter circuit in the frequency range 1 to 6 GHz. The architecture includes low noise amplifier (LNA) Distributed topology (distributed amplifier), mixer and amplifiers and filters in intermediate frequency. The design is accompanied by measurements and found where the convergence of the results of intensive simulations. Main tools of design were kyklomatikoi and electromagnetic simulators.

Αντικείμενο αυτής της διπλωματικής υπήρξε η μελέτη μιας νέας μεθόδου ενίσχυσης ακουστικών συχνοτήτων, που μπορεί να αναπαράγει τον ήχο με την υψηλή ποιότητα των ενισχυτών τάξης Α και ταυτόχρονα με υψηλή απόδοση ισχύος, που χαρακτηρίζει τους ενισχυτές τάξης D. / This Diploma Thesis studies a new method of audio amplifying that can reproduce sound with both high fidelity as found in a class A amplifier and high efficiency which is characteristic of class D amplifier.

Thesis (MEng)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: This thesis presents the design, construction and measurement of two wide-band LNA’s for the SKA-Mid range (350-1200 MHz). The first wide-band LNA involves the investigation of classic low noise amplifier techniques, which includes basic noise theory, stability analysis, feedback design and the development of sophisticated matching techniques for ultra wide-band performance. Final measurements show a flat gain response equal to 19 dB, with a noise figure of 1.5 dB and an output return loss of 10 dB across the entire bandwidth. A multi-path cascading concept is introduced for the second low noise amplifier design, which aims to connect two single frequency amplifiers in parallel to operate from 500 to 700 MHz. The design process involves several optimization schemes to realise the matching networks for the cascaded topology and the noise performance of the device was confirmed by using multi-port noise theory. The prototype presents significant bandwidth improvements compared to a single frequency LNA design. Excellent agreement between the simulation and measurement were obtained with a flat gain response of 20 dB across a 2:1 bandwidth, with a low noise figure of 0.95 dB and an output return loss of 13 dB across the operation bandwidth of 400 to 800 MHz. / AFRIKAANSE OPSOMMING: Hierdie tesis behandel die ontwerp, konstruksie en meting van twee wyeband laeruis versterkers vir die SKA - Mid reeks (350–1200 MHz). Die eerste wyeband laeruis versterker, ondersoek klassieke laeruis versterker tegnieke wat insluit basiese ruisteorie, stabiliteit analise, terugvoerontwerp en die ontwikkeling van gevorderde aanpassingstegnieke vir ultra wyeband werkverrigting. Finale metings het ’n plat aanwins van 19 dB, met ’n ruisfiguur van 1.5 dB en ’n uittree-refleksie koëffisiënt van -10 dB oor die hele bandwydte vertoon. ’n Multi-pad konsep word bekend gestel vir die tweede laeruis versterker. Die ontwerp het twee enkel frekwensie laeruis versterkers in parallel verbind om vanaf 500 tot 700 MHz te werk. Die ontwerp proses bevat verskeie optimalisering skemas om die aanpassings netwerke vir die kaskade topologie te realiseer. Die ruissyfer van die versterker is bevestig deur die gebruik van multi-pad ruisteorie. Die prototipe het beduidende bandwydte verbeterings vertoon in vergelyking met ’n enkel frekwensie versterker ontwerp. ’n Uitstekende ooreenkoms tussen die simulasie en meting was verkry met ’n plat aanwins van 20 dB oor ’n 2:1 bandwydte, met ’n laeruisfiguur van 0.95 dB en ’n uittree-refleksie koëffisiënt van -13 dB oor die bandwydte van 400-800 MHz.

[pt] Neste trabalho, uma análise do fenômeno de mistura de quatro ondas (FWM) em sistemas ópticos multicanais é apresentada. Inicialmente, uma série de simulações determina as limitações impostas aos sistemas ópticos, sendo adotado um nível máximo de crosstalk permitido. Em seguida, os diversos parâmetros que influenciaram a eficiência da geração de FWM são estudados detalhadamente, com resultados apresentados em forma de gráficos. Nestes últimos, o nível de FWM obtido é especificado em função da potência de entrada por canal, comprimento do enlace, número de canais, presença de amplificadores, número de amplificadores e dispersão cromática da fibra utilizada. Um breve estudo discute a deterioração da qualidade do serviço, causada pro FWM, em redes ópticas. O trabalho se encerra com algumas aplicações que utilizam, de forma favorável, o fenômeno de FWM. / [en] This work adresses the four-wave mixing (FWM) issue in optical multichannel systems. First, some simulations are carried out for specifying which are the optical system limitations coping with a maximum allowed crosstalk level. Next, the different parameters affecting the FWM are throughly studied, with these last pertinent results being graphicallly presented. In these, the obtained FWM levels are specified as a function of the channel input power, link span, channel rumber, presence of amplifiers, number of amplifiers and fiber chromatic dispersion. A concise discussion, concerning the quality of service deterioration on optical networks, due to FWM, is also presented. The work is finally closed by presenting some applications that favorably use the FWM phenomena.