Includes bibliographical references (p. 113-115). / Under certain conditions Iterative Feedback Tuning (IFT) may produce a controller that cancels the poles of the process and as a result can give a closed loop that has poor internal performance. The disadvantage of this is that the closed loop will have poor input disturbance rejection. A solution for ensuring that IFT does not have poor internal performance is to make sure that the disturbance rejection is adequate. However an adequate input disturbance may lead to other undesirable dynamics in the closed loop performance. These are such as overshoot in the response for setpoint tracking and that for output disturbance rejection. On the other hand the advantage of pole shifting is that for a one degree of freedom control structure all the characteristic equations of the loop transfer functions will be the same. Four methods are proposed for avoiding pole-zero cancellation by concentrating on the input disturbance. These methods are using: a model for input disturbance rejection, time-weighted IFT for disturbance rejection, a setpoint-tracking model with overshoot and approximate pole placement IFT. Approximate pole placement IFT was chosen as the best method. The reason is that the dynamics of the closed loop can be specified with the choice of characteristic equation. This method was then investigated further to establish its feasibility on a physical system. After the evaluation of this method, it was applied on a DC motor for speed control to show that is viable in practice. Multiple experiments were done to show that this method does not produce a controller that cancels the process poles, confirming it as a good solution to prevent poor internal performance.

The quality of control provided by what is termed "short-sighted" stochastic optimal controllers controlling a linear heating system subjected to a random disturbance is evaluated by computer simulation and by experimentation.

Includes bibliographical references. / A market has been identified for 23 GHz, short-haul, low-capacity, digital radio. The dissertation presents the development of the varactor controlled Gunn oscillators that constitute the crystal locked microwave sources of the radio. An accurate description of a design procedure for Gunn oscillators at 23 GHz is presented. With reference to advanced modulation methods which require constant modulation indices, a method of linearising the voltage/frequency characteristic of the varactor controlled Gunn oscillator is described, which allows direct modulation of the source at 23 GHz. Due to the wide operating temperature of the radio a technique to temperature compensate the oscillator is presented. The dissertation ends with an investigation how the semiconductor device's spread affects the oscillator characteristics and an evaluation of the noise performance of the Gunn oscillator.

Bibliography:leaves 203-205. / Decentralised control schemes are becoming more common in industry as the advantages of decentralised control become more apparent. These advantages include fewer tuning parameters than centralised controllers, the simplification and cost reduction of hardware requirements and greater reliability. In addition the application of decentralised controller design to large scale systems allows established CAD methods to be implemented easily and efficiently. When the control engineer designs a distributed controller the system is divided up into a number of subsystems and a controller designed for each subsystem. The controllers are designed independently for each subsystem ignoring any interaction that may occur between the different subsystems. In terms of the input-output representation of the system this means that the matrix representing the controller will be in a block diagonal form. In general the interactions between the different subsystems will not be negligible. In some cases the interactions will be such that stabilising the individual subsystems will not be sufficient to stabilise the system as a whole. Stability theorems are required to enable the designer to check if the decentralised controller that he has designed will in fact stabilise the system as a whole. Such stability theorems have been devised although at present they are too conservative. However even with such theorems available the designer must still select the subsystems to be controlled in such a way as to satisfy the conditions laid down for stability. The stability theories usually are based on a particular matrix structure. If the matrix representing the system possesses a structure detailed by the stability theorem in question then, subject to various conditions, the system as a whole will be stable under decentralised control. In this thesis a number of different matrix structures are considered that give information as to the stability of the closed loop system. Methods are developed that allow the designer to rearrange the matrix in such a way as to obtain a particular structure, if this is possible.

Bibliography: p. 102-103. / It was the objective of this project to design and build a transmitter/receiver unit that indicates when the ore level in a vertical ore chute falls below the level of the signal path. The design was commissioned by De Beers Consolidated Mines to be implemented in South Africa's largest diamond producing mine, the Finsch Mine at Limeacres in the north-western Cape. The system requirements laid out'by de Beers stressed that the unit is to be used as a backup system, and thus its absolute reliability was of highest importance.

Bibliography: leaves 78-79. / Partial response signaling ( PRS) has been used successfully to improve the spectral properties of Pulse Amplitude Modulated (PAM) digital transmission systems. This thesis investigation studied the effect of PRS on frequency- and phase-modulated carrier systems, in particular on their spectral performance and their maintenance of constant envelope.

Includes bibliographical references. / This project arose as a result of inefficiencies in the diamond recovery process at Premier Mine. A considerable amount of barren waste rock, gabbro, is mined along with the diamond bearing kimberlite. No automated method exists for separating the kimberlite from the waste rock and a device was required to effect ore sorting on a rock by rock basis. Experimentation with a microwave oven indicated that samples of kimberlite were more attenuative than samples of gabbro. The possibility of using microwave heating for rock differentiation was investigated but was impractical to implement. A study of low power microwave attenuation and reflection measurements was undertaken. Reflection measurements were found to be impractical due to the similar amounts of reflected signal from the different rock types. Microwave signal attenuation through rock samples was studied over a broad frequency spectrum. A detectable difference in signal attenuation was found through the gabbro and kimberlite. The difference in signal attenuation increased with increasing frequency. Different techniques to implement signal attenuation measurements through rock samples were investigated. The passing of rock samples through waveguide structures was found to be impractical in this application. Microwave signal attenuation measurements were successful when rock samples were placed between a transmitting and a receiving antenna. Equipment was designed and constructed with an operating frequency of 35GHz chosen due to the small antenna aperture area and the large attenuation difference at this frequency. Static measurements with this equipment revealed the problems with signal scattering and reflection from some irregularly shaped samples of low loss gabbro. The importance of these phenomenon could only be gauged from dynamic measurements. Dynamic measurements were performed using a laboratory test system with a conveyor belt capable of moving at speeds of up to 5 m/s. It was found that 93% of the kimberlite could be correctly detected whilst rejecting 67% of the gabbro. The system functioned satisfactorily and led to the filing of several patents.

Bipolar complementary metal-oxide semiconductor (BiCMOS) technology is the platform of choice for near-Infrared (IR) detector research because of low power consumption, increased operating speed and a high fill-factor. The drawback is poor noise performance which can be attributed to the readout circuitry of the detector. Conventional near-IR detector design is an iterative process. While recognising the value of this approach, rapid prototyping can be achieved by using mathematical modelling that would ensure design repeatability. Heterojunction bipolar transistor (HBT) and metal-oxide field-effect transistors (MOSFET) models for SiGe process technologies have been documented extensively. However, mathematical modelling of BiCMOS near-IR detectors has not been implemented in a complete working system before. This proposed model can be used to determine the output voltage as well as the noise performance of near-IR detectors. The focus of this research is to determine how process independent parameters and detector performance can be mathematically modelled. Secondly, and associated to this, is determining how the model can be extended to accommodate multiple feature sizes including short-channel MOSFETs. An implementation of this model on the three-transistor pixel structure, using reverse-biased diode-connected HBTs as pixels, was done as part of the experimental verification process of this research. The implementation was done in a 2 × 2 gated array detector configuration. The validity of the proposed modelling procedure was verified through comparison of simulations and measured results. The simulations were done in an iterative fashion to show how a deviation in one process independent parameter affects the noise performance, while the other process independent parameters are kept constant. The detector design with optimal noise performance can be achieved in this manner, thereby minimising design time and developing optimised detectors without the need for extensive prototyping. The main contribution of this research is that a designer can use this mathematical model to tune a detector to achieve desired performance. By changing the temperature, biasing voltage and biasing current and choosing the aspect ratio, noise performance changes. An iterative process in the mathematical model development can achieve optimised parameters for noise performance. Two approaches, namely DC analysis and y-parameter representation, were used to develop the mathematical model. Feedback was taken into account using the y-parameter representation. The measured results show that the output voltage behaviour follows the mathematical model developed. The output voltage behaviour also shows that the mathematical model parameters can determine noise performance. As an extension to this work, the same modelling process can be used to develop mathematical models for other detecting structures such as the four-transistor pixel structure. / Thesis (PhD (Electronic Engineering))--University of Pretoria, 2020. / Optronic Sensor Systems, Defence, Peace, Safety and Security (DPSS), Council for Scientific and Industrial Research (CSIR), South Africa / Armscor, Armaments Corporation of South Africa Limited Act, Act No 51 of 2003 / Department of International Relations, University of Pretoria, South Africa / National Research Foundation (NRF), South Africa / Electrical, Electronic and Computer Engineering / PhD (Electronic Engineering) / Unrestricted

Data traffic of future 5G telecommunication systems is projected to increase 10 000-fold compared to current rates. 5G fronthaul links are therefore expected to operate in the mm-wave spectrum with some preliminary International Telecommunication Union specifications set for the 71-76 and 81-86 GHz bands. Processing 5 GHz as a single contiguous band in real-time, using existing digital signal processing (DSP) systems, is exceedingly challenging. A similar challenge exists in radio astronomy, with the Square Kilometer Array project expecting data throughput rates of 15 Tbits/s at its completion. Speed improvements on existing state-of-the-art DSPs of 2-3 orders of magnitude are therefore required to meet future demands. One possible mitigating approach to processing wideband data in real-time is to replace some DSP blocks with analog signal processing (ASP) equivalents, since analogue devices outperform their digital counterparts in terms of cost, power consumption and the maximum attainable bandwidth. The fundamental building block of any ASP is an all-pass network of prescribed response, which can always be synthesized by cascaded first- and second-order all-pass sections (with two cascaded first-order sections being a special case of the latter). The monolithic integration of all-pass networks in commercial CMOS and BiCMOS technology nodes is a key consideration for commercial adaptation of ASPs, since it supports mass production at reduced costs and operating power requirements, making the ASP approach feasible. However, this integration has presented a number of yet unsolved challenges. Firstly, the state-of-the-art methods for synthesizing quasi-arbitrary group delay functions using all-pass elements lack a theoretical synthesis procedure that guarantees minimum-order networks. In this work an analytically-based solution to the synthesis problem is presented that produces an all-pass network with a response approximating the required group delay to within an arbitrary minimax error. This method is shown to work for any physical realization of second-order all-pass elements, is guaranteed to converge to a global optimum solution without any choice of seed values as an input, and allows synthesis of pre-defined networks described either analytically or numerically. Secondly, second-order all-pass networks are currently primarily implemented in off-chip planar media, which is unsuited for high volume production. Component sensitivity, process tolerances and on-chip parasitics often make proposed on-chip designs impractical. Consequently, to date, no measured results of a dispersive on-chip second-order all-pass network suitable for ASP applications (delay Q-value (QD) larger than 1) have been presented in either CMOS or BiCMOS technology nodes. In this work, the first ever on-chip CMOS second-order all-pass network is proposed with a measured QD-value larger than 1. Measurements indicate a post-tuning bandwidth of 280 MHz, peak-to-nominal delay variation of 10 ns, QD-value of 1.15 and magnitude variation of 3.1 dB. An active on-chip mm-wave second-order all-pass network is further demonstrated in a 130 nm SiGe BiCMOS technology node with a bandwidth of 40 GHz, peak-to-nominal delay of 62 ps, QD-value of 3.6 and a magnitude ripple of 1.4 dB. This is the first time that measurement results of a mm-wave bandwidth second-order all-pass network have been reported. This work therefore presents the first step to monolithically integrating ASP solutions to conventional DSP problems, thereby enabling ultra-wideband signal processing on-chip in commercial technology nodes. / Thesis (PhD)--University of Pretoria, 2018. / Square Kilometer Array (SKA) project - postgraduate scholarship / Electrical, Electronic and Computer Engineering / PhD / Unrestricted

Thesis (MScIng)--University of Stellenbosch, 2007. / ENGLISH ABSTRACT: The quadrifilar helix antenna has frequently been employed to provide the wide angle, circularly polarized radiation pattern which is preferable for mobile satellite communications. A detailed study of this antenna and its varieties is presented. When used as an element in wide angle scanning arrays for aeronautical satellite tracking applications, the additional requirement of a low profile with light weight applies. This further complicates the realization of good circular polarization close to the horizon. A top disk is incorporated into various antenna structures in an effort to improve the radiation performance when used within a closely spaced array while reducing the height of the radiating element. The array performance of these elements are investigated which hints at a supported traveling mode aiding the ability to accomplish the wide angle beam steering. Subsequent study of leakywave antennas initiated the study and implementation of leaky-wave structures in phased arrays of quadrifilar helix elements. / AFRIKAANSE OPSOMMING: Die vierledige heliksantenna word dikwels gebruik vir toepassings wat ’n wye, sirkulêr gepolariseerde stralingspatroon benodig. ’n Verskeidenheid van hierdie antennas is noukeurig bestudeer en hul toepaslikheid vir mobiele sattelietkommunikasie ondersoek. Wanneer hierdie antennas in fase gestuurde skikkings gebruik word, word dit ook vereis dat die eenhede liggewig is met ’n lae profiel. Dit bemoeilik die verkryging van goeie sirkulêre polarisasie naby die horison aansienlik. In ’n poging om die wyehoek straling met kort antennas te verbeter, is ’n oorhoofse skyf tot verskillende heliksstrukture toegevoeg. In nou gespasieerde skikkings maak hierdie skywe ’n positiewe bydrae to wyehoekstralingeienskappe. Die invloed van lekmodusse wat binne die struktuur voortplant is ondersoek en gebruik in skikkings waarvan die bundel naby aan die horison gestuur word.