Antenna tuning for WCDMA RF front end

Abstract

Modern mobile handsets or so called Smart-phones are not just capable of communicating over a wide range of radio frequencies and of supporting various wireless technologies. They also include a range of peripheral devices like camera, keyboard, larger display, ash-light etc. The provision to support such a large feature set in a limited size, constraints the designers of RF front ends to make compromises in the design and placement of the antenna which deteriorates its performance. The surroundings of the antenna especially when it comes in contact with human body, adds to the degradation in its performance. The main reason for the degraded performance is the mismatch of impedance between the antenna and the radio transceiver which causes part of the transmitted power to be reflected back. The loss of power reduces the power amplifier efficiency and leads to shorter battery life. Moreover the reflected power increases the noise floor of the receiver and reduces its sensitivity. Hence the over performance of the radio module in terms of Total Radiated Power and Total Isotropic Sensitivity, gets substantially degraded in the face of these losses. This thesis attempts to solve the issue of impedance mismatch in RF front-ends by introducing an adaptive antenna tuning system between the radio module and the antenna. Using tunable reactive components and by intelligently controlling them through a tuning algorithm, this system is able to compensate the impedance mismatch to a large extent. The improvement in the output power and the reduction in the Return Loss observed in the measurements carried out for WCDMA, as part of this thesis work, confirm this. However, the antenna tuner introduces an insertion loss and hence degrades the performance in perfect match conditions. The overall conclusion is that the adaptive antenna tuner system improves the performance much more than it degrades it. Hence it is an attractive solution to be included in mobile terminals on a commercial scale.