automatic antenna tuners - theory and design
Ettore Firrao is a PhD student in the research group Integrated Circuit Design (ICD). His supervisor is prof.dr.ir. B. Nauta from the Faculty of Electrical Engineering, Mathematics and Computer Science.
In chapter 1, an introduction to wireless communication was given. Antennas and RF front ends, the basic blocks in wireless communication, are analysed. It is explained how the antenna impedance (assumed to be 50 ohm) can be different from 50 ohm depending on the unpredicatable electromagnetic environment. Since the electromagnetic environment can be varied in an unpredictable way, it is not possible to optimise the antenna or to design a static matching network. The effect of the unpredicatable antenna impedance is then analysed on the RF front end. An RF front end is typically made of a switch or a diplexer, an RF power amplifier and a low noise amplifier. The RF power amplifier, or simply RF-PA, is used to drive the antenna and the low noise amplifier, or simply LNA, amplifies the weak signal coming from the antenna. In case of unpredictable antenna impedance, the RF power amplifier has to work under load mismatch conditions resulting in a lower maximum output power, lower efficiency, stability issues, compromised ruggedness, etc. Therefore there is the need for automatic antenna tuners in order to "tune" the antenna impedance to its optimum value (typically 50 ohm).
In chapter 2, automatic antenna tuners are introduced. They are comprised of sensors, a tunable matching network and a control network. Each of these is described (with most detail for the matching network), after which the state-of-art is discussed. Finally, the methodology of the work is described: to investigate theoretical limits and shows how they can be reached.
In chapter 3, a practical implementation of an automatic antenna tuner was proposed. The feature of this implementation is the measurement of only RF magnitudes requiring low frequency control networks. The control loop only requires low-frequency electronics which makes it low cost, low power and relatively easy to integrate. Measurements on a demonstrator system show correct behavior for VSWR up to 10.
In chapter 4, a theoretical derivation on the minimum number of states for switchable matching networks was analysed. The word "state" denotes the status (open or closed) of each switch of the switchable components used inside the switchable matching network. Generally speaking the number of states is equal to 2n where n is the number of the switches. The theoretical result is only dependent on the load VSWR and the required input VSWR without referring to a specific implementation. In other words no specific implementation was taken into account. The results were based on a mathematical derivation and not on brute-force simulations. Although some other authors have proposed a minimum-number-of-states concept as well, all these are based on numerical simulation as opposed to the analytical approach in this chapter. Configurations were analyzed and benchmarked: single-stage one-ring configuration, single-stage two-ring configuration, two-stage one-ring configuration and three-stage one-ring configuration showing that single-ring configurations are optimum. An extension towards the required number of states for lossy matching networks was also provided. The results can be used to optimise the design of switchable matching networks.
In chapter 5, the proof of concept of the theoretical result of chapter 4 was discussed. In order to compare each switchable matching network, a parameter was introduced: the hardwareoverhead. This parameter is defined as the ratio between the actual number of states and the theoretical minimum number of states given in chapter 4. Several switchable matching networks were analysed: PI networks, T networks, loaded transmission lines, branch line coupler based switchable matching networks, single circulator and cascaded circulator based switchable matching networks.
The loaded transmission line was proposed to achieve a lower number of states compared to classical matching networks like a PI matching network or a T matching network. In fact in order to achieve each optimum state, switchable capacitors were used that are equally spaced on a transmission line. The result is quite large number of states because not all the combinations are used and only a small number are really needed: in case of eight states on the ring (where 9 is the total number of states), eight switchable capacitors are needed but the overall number of states is 28 = 256 resulting in an overdesigned switchable matching network. The hardwareoverhead is 256/9 = 28.44.
According to the analysis carried out, the cascaded circulator based switchable matching networks achieve the closest value to the lowest possible number of states as defined in chapter 4. Referring to the example of nine states depicted above, the overall number of states for the cascaded circulator based switchable matching networks is 16 (16 times less than the loaded transmission line). The hardwareoverhead is 16/9 = 1.77. This value is also the best value that can be achieved with binary switchable matching networks since 9 is not an integer power of 2. The main limitation of this matching network topology is the integration. Other topologies such as PI and T matching networks can be integrated.