# «SYAZANA BASYIRAH BINTI MOHAMMAD ZAKI UNIVERSITI SAINS MALAYSIA 2015 DESIGN OF 10 GHz NEGATIVE RESISTANCE DIELECTRIC RESONATOR OSCILLATOR by SYAZANA ...»

the oscillator circuit will be measured by a spectrum analyzer, observed, and recorded.

Environmental effects might be the major factor that limits the accurate measurement which can be reduced with the help of metal casing since it can block the excessive noise.

The effect of frequency pulling also was not discussed as there was no significant change in output frequency that could justify the effect of frequency pulling.The effects of implementing single transistor and dual transistors towards oscillation frequency, output power and phase noise have been discussed further in the discussion.

1.5 Dissertation Organization This dissertation was organized into five main chapters. Chapter 1 began with the general overview of an oscillator and proceeded with a DR and a DRO, problem statement, objectives, scope of research and dissertation organization.

Chapter 2 covers the literature review, which is related mostly to the DRO design. In this chapter, theoretical background of microwave oscillator, negative resistance oscillator, feedback oscillator, DR, DRO and WPD are presented. There are also reviews about the coupling of DR and previous published works on the negative resistance DRO are presented.

Chapter 3 discusses the methodology of this research, which has been employed in order to complete the negative resistance DRO design. The DR is modeled in Computer Simulation Technology Microwave Studio (CST MWS) and also the modeling of a DRO circuit is presented. The modeling of other parts of the negative resistance DRO such as DC bias circuit, RF choke network and the integration of all related parts into a complete negative resistance DRO is presented. The fabrication and measurement process is also discussed in this chapter.

measurement results of a dual transistors negative resistance DRO. The simulation results obtained are compared with the measurement results in order to analyze factors that affected the results differences. Brief analysis based on the comparisons is discussed in this chapter.

Finally, Chapter 5 contains conclusion based on the experimental results. The contributions of this research are clearly highlighted. Some recommendations for future works are suggested at the end of this chapter in order to improve the oscillator design for future innovations.

This chapter evolves towards the overview of the research for designing a negative resistance DROat 10 GHz. Here, the theoretical background has been divided into several sections which include topics and subtopics in detail. In order to achieve objectivesin this dissertation, the theoretical background part is focused mainly on microwave oscillator, negative resistance oscillator, dielectric resonator (DR), and dielectric resonator oscillator (DRO) where the theories and concepts such as characteristics, principle of operation, various oscillators‟ configuration, and performance have been studied. Furthermore, the related parameters such as the quality factor, coupling coefficient and phase noise also have been discussed in the subtopics while in literature review part, published research that related or similar to this dissertation will be reviewed and summarized.Some ideas and understanding on the design of negative resistance DRO at 10 GHz can be realized in terms of its operation, performance, limitations and suggestion to improve those designs.

2.2 Theoretical Background 2.2.1 Microwave Oscillator Oscillator is an essential and fundamental system block in every wireless or wired product(Butt et al., 2009). Hence, it becomes a logical choice for many fixed-frequency receiver/transmitter local oscillator applications (Pavio and Smith, 1985). It functions as a signal generator in a transmitter while the local oscillator in a receiver is used together with a mixer to convert the received RF signal to an IF signal. The oscillator can be modulated by a low frequency analog or digital signal, where its signal serves as a carrier and the modulating low

efficiency, low noise, good stability and good frequency tenability are indispensable (Mahyuddin, 2006). The design goals have a significant impact on development of oscillator technology. Apart from the cost, size and power consumption, another important parameter that have been actively investigated in oscillator design is the phase noise. Oscillator phase noise can be effectively reduced by incorporating high-Q resonators (Butt et al., 2009).

Oscillators represent the basic microwave energy source for all microwave systems such as radars, communications, navigation, electronic warfare missile(Wan, 2008). A typical microwave oscillator consists of an active device (a diode or a transistor) and a passive frequency-determining resonant element.

All microwave oscillators are designed by adding resonating elements (L, C, or R) in various configurations to different ports of a transistor. These elements generate a negative resistance at a certain resonant frequency and set the device into oscillation. In the case of a DRO, the resonating element is the DR, which can be modeled electrically as an L, C, R network (MITEQ, 2013).

2.2.2 Characteristics of an Oscillator Recently, wireless communication services have been developed rapidly. Microwave oscillators are very important components in these high frequency systems where the higher frequency signals are required to carry the information and to transmit huge amount of information (Tanaka and Aikawa, 2009, Han-li et al., 2012).

There are several parameters or characteristics which are essential in part of an oscillator. The major aspects that have to be analyzed are the output power, operating frequency and tuning range, efficiency, stability, noise, phase noise, harmonic suppression, frequency

the characteristics of an oscillator is the output power which is defined as continuous power generated by an oscillator in watts or milliwatts for continuous wave operation and the peak power or average power generated by an oscillator for pulse operation.

Operating frequency and tuning range are a fixed output frequency or a tunable output frequency range which is used either mechanically or electronically tunable oscillator. The frequency range is determined by the architecture of the oscillator. Tuning for an oscillator can be divided into several categories, which are post-tuning drift, tuning linearity, tuning sensitivity, tuning performance and tuning speed.Stability is the ability of an oscillator to return to the original operating point after experiencing a slight electrical or mechanical disturbance which refers to both short term and long term stability. The oscillator should be clean in the sense that it does not pick up unwanted signals and noise in the circuit (Mahyuddin, 2006).

Another important characteristic of an oscillator is its noise comprised of amplitude modulation noise or amplitude variations of the output signal, FM noise, unwanted frequency variations and phase noise or phase variations. Various noise sources contribute to oscillator noise including the loss of the resonator, the noise sources inside the transistor; noise modulated on the power supply and the noise contributions from the tuning diodes. FM noise is usually measured at about 100 kHz from the carrier in unit of dBc, which means decibels below the carrier level, in a specified bandwidth of 1 Hz. The various noise sources in and outside of the transistor modulate the oscillator, resulting the energy or spectral distribution on both sides of the carrier. Another type of noise that describes an oscillator is a harmonic suppression. The oscillator has a typical harmonic suppression of more than 15 dB. For high performance applications, a low pass filter at the output will reduce the harmonic contents to a desired level (Rohde et al., 2005a).

the output frequency which appears, in the frequency domain, as FM energy around the carrier frequency. It is specified in dBc/Hz and measured at specified offsets from the carrier frequency which is typically 10 kHz and 100 kHz. The phase noise (or FM noise) is produced by thermal noise, shot noise, and flicker noise.

Frequency stability or phase stability refers to the ability of the oscillator to maintain constant frequency of oscillation. In achieving frequency stability, series or parallel resonant circuits are generally used in the terminating network, such as lumped elements, cavities and DRs. A requirement for all resonators is to acquire high Q factor or low loss. By appending a high Q tuning network to the oscillator, this will result in the enhancement of the oscillator stability (Mahyuddin et al., 2006).

As well known, an amplifier and a tuned circuit that arebuilt in an oscillator are functioned to transform the dc energy into RF energy at the desired frequency and acceptable power added efficiency. By depending on the frequencies and configurations, the efficiency of a low noise oscillator varies between the range of 10% to 70%(Rohde et al., 2005b). According to Rohde, the main objective of the amplifier and tuned circuit that is built in an oscillator is to gain a frequency output signal which is stable with low phase noise and free of spurious signals at sufficient level.

The unloaded quality factor is the parameter of the DR which influences the oscillator‟s phase noise. It can be defined as a ratio of the stored energy within the puck to the amount of energy which is dissipated by the resonator and serves as a figure of merit for the device.

Unloaded quality factors of 20,000 are now possible to 10 GHz. However, the loaded quality factor will determine the oscillator‟s phase noise response. The loaded quality factor can be defined as a ratio of the energy stored within the puck per cycle to that delivered to the load per

reduction in the loaded Q, resulting from the inclusion of voltage variable frequency control, in the traditional manner, results in a degradation of the oscillator phase noise performance (Warburton, 2005).

The effects of the output frequency of an oscillator can be classified into three categories known as jumping, pushing and pulling frequency. Jumping frequency is a term to describe a discontinuous change in oscillator frequency due to nonlinearities in the device impedance.

Meanwhile, frequency pushing is a measure of the sensitivity of the oscillator output frequency to the supply voltage (MHz/volt) which is tested by varying the dc supply voltage of an oscillator with its tuning voltage held constant, where it is typically around ±1V. In other words, frequency pushing characterizes the degree to which an oscillator‟s frequency is affected by its supply voltage(Rohde et al., 2005a). On the other hand, frequency pulling is a measure of the change in frequency to a non-ideal load (load mismatch over 360° of phase noise variation).

Frequency pulling is also the change of frequency resulting from partially reactive loads, which is an important oscillator characteristic (MINICIRCUIT, 2013, Mahyuddin, 2006).

Spurious outputs are signals found around the carrier of an oscillator which are not harmonically related. A good, clean oscillator needs to have a spurious free range of 90 dB, but this requirement makes it expensive. Typically, an oscillator have no spurious frequencies other than possibly 60 Hz and 120 Hz pickup (Rohde et al., 2005a).

2.2.3 High Frequency Oscillator Design Both negative resistance and feedback DROs design in this dissertation contain the implementation of a transistor oscillator which concern with nonlinear analysis of high frequency FET and bipolar oscillators. The method of designing a transistor oscillator involves

bias circuits and S-parameters, with the exception of the transistor oscillator is designed to be unstable. There are some requirements have to be taken into considerations concerning generating an oscillation for transistor amplifier and transistor oscillator design. As for transistor amplifier, S11 and S22 must be less than unity while for transistor oscillator to be unstable, S11 and S22 must be greater than unity.

Bipolar, MOSFET, MESFET, HEMT and HBT are examples of transistors which can be implemented into the oscillator design method based on the S-parameters datasheet of the transistor correspondingly. Normally, common source or common gate FET configurations are frequently used since it meets the condition of implementing a high degree of unstable device in an oscillator.

By referring to the previous section, resonators play a major part in an oscillator which has been implemented into the oscillator design towards accomplishes a low noise and high frequency stability. They are made of lumped element, a distributed transmission line, a cavity or a dielectric disc.

The approach of S-parameters in small and large signal designs which has been implemented into the oscillator design as a requirement for oscillation to start up. Figure 2.1 shows a two-port transistor oscillator where ZL is the load impedance and ZT is the terminating impedance seen by the transistor. The condition for oscillation to start is by the used of terminating network in order provide the |Гout| 1. The frequency of oscillation and power delivery to the load are determined by the load network(Maas, 2003, Mahyuddin, 2006).

Where;

Rout = real part of output resistance Xout = imaginary part of load resistance RL = real part of load resistance XL = imaginary part of load impedance V = voltage across the output resistance (V) f0 = carrier centre frequency (Hz) The above conditions need to be satisfied to make sure that the oscillation occurs at the desired frequency. It is stated that Rout is negative. Equation 2.1 ensures that |Гout| 1 and the Equation 2.2 determines the oscillation frequency. The network has the potential for oscillation as long as the first equation is satisfied. As an elaboration, the oscillations start to build from the noise level when the power supply voltages of an oscillator are turned on. As a result, the output amplitude will continue to grow until the saturation effects of the device limit it. Decisively, the