The novel part of this implementation is that the amplifier bias conditions are determined from the error signals within the closed loop system. For the polar loop case, these are the phase and amplitude error signals from the detectors. This error signal can be related to the distortions in the amplifier. For example, the phase error relates to error-vector magnitude (EVM) while the amplitude error relates to adjacent-channel power modulation (ACPM). By processing the magnitude of these errors (by analog or digital methods), it is possible to determine the distortion products generated within the amplifier. When these error signals fall outside the predefined window the biasing arrangements for the amplifier is not optimum and need to be adjusted. Therefore, these error signals can be used to determine the amplifier dc bias conditions. It is assumed that the ac load line is also maintained within reasonable limits.

Figure 5 shows a simplified 3G waveform along with the distortion artifacts following amplification (these are the red blocks). These artifacts occur as a result of phase and amplitude distortions within the amplifier, which results in a spreading of energy into the adjacent channels. To provide correction for these distortions, the linearizer will produce a comparison signal following downconversion to a low frequency. The algorithm in Figure 6 show this comparison signal, which results from comparing input reference and output signals.

Assume that the power amplifier is required to deliver a certain power with a predefined EVM and ACPM distortion. Initially, the quiescent and supply voltages are too low to deliver the RF power required so the amplifier produces excessive distortion that fails specification. The linearizer error signal will exceed the predefined limit indicating that the distortion is outside specification. This error signal is monitored as the supply voltage and quiescent current are increased for the amplifier. When the error signal falls within the permitted range, the bias is then correct. In practice, this happens dynamically resulting in the best efficiency over a wide range of RF powers for varied modulation types.

Conversely, if the quiescent current and supply voltages are high for the required RF power, the amplifier will produce minimal distortion and the error signal from the linearizer will be too small. This error signal is monitored as the supply voltage and quiescent current are decreased for the amplifier until the error signal lies within the predefined window (Figure 6).

Under these conditions the linearizer is working at full capacity over a range of RF powers. Returning to the load line, the improvement is clear with reference to the load line where constant efficiency is maintained.

From Figure 7 it can be seen that the benefits of linearization are maintained while the RF is backed off. In addition, the amplifier efficiency is constant over a 10 dB RF power range. These figures are based on a Class A gallium arsenide (GaAs) amplifier.

The hybrid combination of adaptive bias and linearization offers significant efficiency benefits including:

• Optimum efficiency over a wide range of RF output powers for different signals.

• Near constant efficiency over a range of powers with slow or dynamic bias adaptation.

• Provides a method for determining acceptable distortion artifacts so that the adaptive bias can be optimally set.

• Linearizer reduces the potential for modulation due to distortions occurring from adaptive bias techniques.

• Potentially permits the supply voltage and quiescent current of the amplifier to preserve the load line in accordance with the processed error signals.

• No need for significant engineering margins to be built into the design.

I would like to thank all my colleagues in the Wireless Technology Practice for their assistance and helpful comments during this research.

For the PDF version of this article, click here.

1. Tim Fergus, EDGE modulation — How Linearization Improves Amplifier Performance, RF Design, October 2002.

2. S. C, Cripps, RF Power Amplifers for Wireless Communications, Artech House 1999, ISBN 0-89006-989-1.

3. P B Kenington, High Linearity RF Amplifier Design, Artech House, 2000, ISBN 1-58053-143-1.

Tim Fergus is a consultant at the Wireless Technology Practice of PA Consulting Group, Melbourn, Hertfordshire, U.K. He holds a First Class B. Eng. in Electronic and Electrical Engineering from Brunel University and has worked in consultancy for more than six years specializing in RF and high-speed analog circuits. He has worked on a broad range of communication systems from bespoke systems to digital cellular systems including 3G, EDGE, GSM for handset and base station applications. He can be contacted at Tim.Fergus@paconsulting.com.

Previous 1 2 3

RSS    Save to Del.icio.us  Digg This