The digital cellular systems era started about thirty years ago with the release of the first digital cellphones. These first digital cellphones were very different from today’s slim and esthetic cellular pocket computers. They were not mass-produced in million units a day, and they were designed for radio performance rather than appearance. Today, all components are integrated inside the mobile phone to enable a product for the masses and not only the lucky few. For the radio performance this makes a large difference, especially the cellphones interaction with the user, which has a tendency to load the integrated antennas. This loading of the antennas means that the electronics inside the cellphone works sub-optimally, and a decrease in radio performance is inevitable, resulting in increased power consumption and reduced data rates. This problem can, however, be reduced by a concept called adaptive antenna impedance matching. This compensates for antenna loading effects, so that the electronics inside the cellphone can still operate with a close to optimum impedance, although the antenna impedance is changed due to user interaction. For adaptive impedance matching, the key component is the so called impedance tuner, which is studied, designed, and evaluated in this thesis. The requirements on this impedance tuner are very high, with low insertion loss, in-band distortion, out-of-band distortion, high tunability, and good power handling. The cost should also be as low as possible, which means that it should be implemented in a CMOS based technology suitable for mass-production. In this thesis, an impedance tuner is therefore designed and implemented in a CMOS-SOI technology. It has been verified to fulfill the requirements for use in a modern cellphone, with all measurements of key merits indicating high performance. Finally, it is worth to mention that this impedance tuner has also been used in a different project, where adaptive impedance matching was used in MIMO channel measurements with real cellphone users, but that project is outside the scope of this thesis.
The range of frequencies used for cellular communication has increased over the years, and today a large part of the so called sub-6 GHz frequency range is used. Most of the wireless services we have today use this decade wide frequency range (~600-6000 MHz), and although it is a wide frequency range, the spectrum is congested with a high density of communication. The circuits used to communicate in the sub-6 GHz bands must therefore have high RF-performance, and they should also be low cost since a large number of circuits is used to cover the complete frequency range. Difficulties reducing the cost per frequency band has drastically increased the cost of today’s cellphones. This thesis therefore proposes an alternative concept for the power amplifier, the key component in the transmitter of the mobile phone, with the goal to reduce the cost of and the number of power amplifier circuits required to cover the complete sub-6 GHz range. The core of the concept was first designed and verified by measurements, an injection-locked power amplifier with supply modulation and dynamic transistor bias, resulting in high efficiency and bandwidth. To further reduce the cost of the cellphone more of the transmitter functionality, i.e. the frequency up-conversion, was added to the power amplifier circuit, which also improved the overall transmitter performance. Furthermore, a new version of polar modulation is proposed, to reduce the baseband signal bandwidth expansion, which polar modulation is notorious for. The reduction in bandwidth expansion decreases the overall power consumption of the transmitter, since the baseband circuits can then have lower bandwidth and clock-frequency. To further reduce the number of power amplifier circuits needed to cover the entire sub-6 GHz range, the bandwidth of the circuit was improved using a new higher order output matching network, together with a dual output power amplifier, resulting in a circuit that can operate with high performance over the complete sub-6 GHz frequency range. The proposed solutions in this thesis can reduce the number of ICs in cellular devices, which benefits not only the production cost, but also has positive effects on the environment.
- Sjöland, Henrik, handledare
- Törmänen, Markus, Biträdande handledare
|2019 nov. 8
|Published - 2019 okt. 15
Place: Lecture hall E:1406, building E, Ole Römers väg 3, Lund University, Faculty of Engineering LTH, Lund
Name: Rahkonen, Timo
Affiliation: Oulu University, Finland