Calibration enhancement of non-linear VNA system

Communication systems are generally found wherever data is to be transmitted from point-to-point or from a point to many points. It is impossible to imagine modern life without communication systems such as radio, telephone, TV, Satellite and etc. The transmission of information from one place to another requires an operation or other alteration to be sent through an electrical signal; the same principle applies for the receiving terminal. Modern life requires that efficient wireless communication systems for long range transmission be built, therefore, base-stations must use high power transistors almost exclusively. Furthermore, modern cellular communication systems also need to transmit across long distances, hence, to achieve this aim successfully, a radio frequency (RF) power amplifier is employed. That is considered to be a key element of any wireless communication system. Efficient communication systems must have minimum spectral re-growth, interference and employ linear. Signal amplification is one of the core circuit functions in modern microwave and RF systems. The Power Amplifier (PA) is a key element in the construction of all wireless communication systems. The PA uses the most current because it is the last stage in the transmission chain. In modern PA design, the RFPA designer must have accurate S-parameter data for the DUT, thereby allowing the creation of an accurate system model and to reduce the re-design and rework effort. The ultimate aim of the research work presented in this thesis is to achieve improvement in the accuracy of a waveform measurement system by increasing the accuracy of the small-signal calibration used. This involved removing the phase reference from the NVNA during calibration and operation, which in turn removes the bandwidth and frequency limitations that the phase reference imposes, as well as reducing the complexity of the overall system. Essential contributions to this research work concentrated in two areas; firstly, developments that allow for Enhanced Vector Calibration of Load-pull measurement systems, especially near the edge of the Smith Chart, and secondly, the operation and Anoor Aldoumani Abstract IV calibration of a VNA-based large-signal RF I-V waveform measurement system without using a harmonic phase reference standard. The first research area described in this thesis involved investigating the prospect of improving vector measurement accuracy, especially near the edge of the Smith Chart, by using load pull technology. Increased measurement error near the edge of the Smith chart was observed during calibration. To help correct this, the realisation of novel optimization that increases the accuracy of all the raw data which was collected during calibration process and therefore increases the accuracy of calibration at reflection coefficients close to unity. This research work focuses on taking advantage of the load-pull capability during calibration, this reduces the effect of measurement errors on the raw data when measuring the calibration standards before being applied in traditional LRL/TRL calibration algorithms. Leading to time proved measurement accuracy and eliminates the requirement to use complex optimisation algorithms post calibration. The second concept developed simplifies the NVNA architecture and removes the complexities and bandwidth limitations introduced when employing a harmonic phase reference generator. A key capability of the Rohde and Schwarz ZVA-67 VNA is that it incorporates internal signal and local oscillator sources and employs direct digital synthesis was exploited to advantage allows the Vector Network Analyzer to be operated as a NVNA without the need for a harmonic phase reference generator. This is due to such a Vector Network Analyzer based NVNA configuration provided a system with both coherent receivers and sources. This feature combined with a modified calibration procedure, means that during calibration only the internal signal sources and an external phase meter are required during measurement. All the internal signal sources and receiver port are available to measure, also now since no phase reference required, bandwidth and functionality issues and avoided.