No description available.

No description available.

No description available.

No description available.

A new general three-port systems model of the thickness-mode piezoelectric transducer is proposed. This model is considered to have several substantial advantages over existing modelling techniques. In particular, it may be readily used to explain the underlying physical operation of ultrasonic transducers. The model has been developed in such a manner that the effects of arbitrary electrical load and source elements and mechanical matching layers may be incorporated. The use of z-transforms in the calculation of the transducer transient response has been investigated. This has resulted in a fast, efficient and accurate method for calculating the transducer response to arbitrary transient excitation. In the course of this work, the model has been verified extensively by computer simulation and experimental measurement. Excellent agreement was obtained between the theoretical and practical results for a comprehensive range of electrical and mechanical configurations. Techniques for controlling piezoelectric transmitters electrically have been investigated. This involves the generation of a precisely defined force transient, by exciting the transducer system with a calculated voltage. These methods were verified by simulation and experiment, and it is concluded that they offer a significant improvement over existing transducer control strategies. Finally, it is considered that the techniques and models proposed have made a significant contribution to the investigation of thickness-mode piezoelectric devices.

No description available.

In the recently evolved fields of satellite and space communications as well as in a number of related areas, a vital requirement is an accurate knowledge of the radiating and receiving characteristics of the transmitting and receiving antennas as they appear at a large distance (in the so called far-field region). It is often impossible to obtain a direct measurement of the performance of an antenna and in such cases where it is possible, the accuracy obtainable is frequently limited by the many difficulties associated with the process. Over recent years, a number of techniques have begun to appear which allow measurement of data close to the test antenna (in the near-field region) and then by mathematical processing (the transformation) predict what the far-field performance will be. The earlier techniques while being basically simple from a mathematical viewpoint, were not completely general and tended to involve special, sophisticated, hardware. The later techniques use the most general spherical scanning system but involve much more complicated processing. A new approach to the problem is presented in which much of the computational burden is pre-processed so that the size and complexity of the ultimate prediction task is reduced. The various measurement systems are considered briefly and the spherical system is formulated in detail. Simulated and experimental predictions are carried out and studies are included of the various errors likely to be present and their effects. The important parameters, including the sampling criterion, are discussed in some detail. It is shown that this technique has the potential for producing rapid and accurate predictions of antenna far-field patterns including the facility of compensation for the characteristics of the measurement probe.

No description available.

No description available.

This thesis surveys linearisation techniques for implementing monolithic MOS active resistors and transconductors, and investigates the design of linear tunable resistors and transconductors. Improving linearity and tunability in the presence of non-ideal factors such as bulk modulation, mobility-degradation effects and mismatch of transistors is a principal objective. A family of new non-saturation-mode resistors and two novel saturation-mode transconductors are developed. Where possible, approximate analytical expressions are derived to explain the principles of operation. Performance comparisons of the new structures are made with other well-known circuits and their relative advantages and disadvantages evaluated. Experimental and simulation results are presented which validate the proposed linearisation techniques. It is shown that the proposed family of resistors offers improved linearity whilst the transconductors combine extended tunability with low distortion. Continuous-time filter examples are given to demonstrate the potential of these circuits for application in analogue signal-processing tasks.