Intermodulation Distortion in Active and Passive Components

Licentiatavhandling, 2023

Wireless communication is growing rapidly, and the demand for higher data rate, low latency, and availability leads to a more complex infrastructure with new challenges. Multiple communication systems have to co-exist in a densely populated frequency-spectrum, where higher power levels and more sensitive receivers are required. However, this also increases the potential risk of interference from weakly spurious signals that are detectable by these sensitive receivers, leading to a degradation in the performance of the communication systems. The unwanted spurious signals are generated when two or more signals of different frequencies pass through a nonlinear element, resulting in mixing products. If the mixing products are produced in an active device, they are referred to as intermodulation distortion (IMD). In passive devices, they are referred to as passive intermodulation (PIM) distortion. The requirements for highly linear devices are increasing because of these new challenges. Therefore, a better understanding of the nonlinear behavior is imperative to develop accurate nonlinear models that will aid in mitigating IMD.

This thesis analyze and model IMD in periodic structures. The periodic structure employed in this thesis is a loaded-line phase shifter that is periodically loaded by varactor-diodes, which are highly nonlinear elements that generate IMD in the forward and backward direction. In a multi-source environment, the IMD from each source creates an interference pattern that can add constructively or destructively. The IMD generated in loaded-line phase shifters have been investigated for several design factors, e.g., periodicity, bias condition, input power, number of unit cells, and capacitance per unit cell length. Moreover, it was demonstrated that the capacitance-voltage relationship in a hyperabrupt varactor-diode has to be accurately modeled to predict the sensitive nonlinear behavior. A polynomial varactor model was employed and experimentally validated with satisfactory results. The model was scaled and evaluated further in a circuit simulator to investigate if there is an optimum design of loaded-line phase shifters in terms of phase-shift/loss and linearity. It was demonstrated that evenly distributing the varactor capacitance for the same varactor capacitance per unit length improves the phase shift/loss. Additionally, it showed that there is a trade-off between low loss and low IMD.