Text preview for : 5990-6534EN.pdf part of HP 5990-6534EN HP Publikacje 5990-6534EN.pdf
Back to : 5990-6534EN.pdf | Home
Digital Pre-distortion and Hardware
Verification using SystemVue
Application Note
by Jinbiao Xu, Agilent Technologies Inc., EEsof EDA
Introduction
Power amplifier (PA) linearization using digital pre-distortion (DPD) techniques
is critical for designers transitioning 3G systems to 3.9G and 4G. This application
note introduces the concept behind DPD, as well as how to use the Agilent
W1716 SystemVue DPD builder to do hardware verification. This software
module implements memory polynomial algorithms to correct wireless
components that have analog memory effects. Both simulation-based and test
equipment-based extraction and verification are possible. For the purposes of this
application note, SystemVue will be used with various Agilent test equipment,
including the ESG/PSG/MXG family of signal sources and PSA/MXA/PXA
family of analyzers.
Digital pre-distortion
Power amplifiers are essential components in the overall performance and
throughput of communication systems, but they are inherently nonlinear. The
nonlinearity generates spectral re-growth, which leads to adjacent channel
interference and violations of the out-of-band emissions standards mandated by
regulatory bodies. It also causes in-band distortion, which degrades the bit-error-
rate (BER) and data throughput of the communication system.
To reduce the nonlinearity, the power amplifier can be operated at a lower power
(that is, "backed off") so that it operates within the linear portion of its operating
curve. However, newer transmission formats, such as wideband code division
multiple access (WCDMA) and orthogonal frequency division multiplexing
(OFDM, 3GPP LTE), have high peak-to-average power ratios (PAPR); that is, large
fluctuations in their signal envelopes. This means that the power amplifier needs
to be backed off well below its maximum saturated output power in order to
handle infrequent peaks, which result in very low efficiencies (typically less than
10%). With greater than 90% of the DC power being lost and turning into heat,
the amplifier performance, reliability and ongoing operating expenses (OPEX) are
all degraded.
To maintain linearity and efficiency, one can apply linearization to the PA through
several techniques such as feedback, feed-forward and DPD.
x (t ) z (t ) y (t )
DPD PA
PA-1
Figure 1. DPD-PA cascade
DPD is one of the most cost-effective linearization techniques. Compared with
feedback and feed-forward linearization techniques, DPD has several advan-
tages. It features an excellent linearization capability, the ability to preserve
overall efficiency, and it takes full advantage of advances in digital signal pro-
cessors and A/D converters. The technique adds an expanding nonlinearity in
the baseband that complements the compressing characteristic of the RF power
amplifier (Figure 1). Ideally, the cascade of the pre-distorter and the power
amplifier becomes linear and the original input is amplified by a constant gain.
With the pre-distorter, the power amplifier can be utilized up to its saturation
point while still maintaining good linearity, thereby significantly increasing its
efficiency. From Figure 1, the DPD can be seen as an "inverse" of the PA. The
DPD algorithm needs to model the PA behavior accurately and efficiently for
successful DPD deployment.
The DPD-PA cascade attempts to combine two nonlinear systems into one lin-
ear result, which allows the PA to operate closer to saturation. Figure 2 shows
the amplitude response of the DPD network, the PA and the DPD-PA cascade,
respectively.
DPD Actual PA Linear PA
Amplitude Response Amplitude Response Amplitude Response
Power Out (dBm)
Power Out (dBm)
Power Out (dBm)
Power In (dBm) Power In (dBm)
Power In (dBm) Power In (dBm)
Figure 2. Amplitude response of DPD, PA and DPD-PA linearized cascade
2
DPD implementations can be classified into memoryless models and models with
memory.
Memoryless models focus on power amplifiers where the output depends only on
the instantaneous input, amplified through a nonlinear mechanism. The complex
values of this nonlinear transfer function are usually characterized by the AM-AM
and AM-PM responses of the power amplifier, where the output signal amplitude
and phase deviation are given as functions of the amplitude of the current input
value. The memoryless polynomial algorithm and the Look-Up Table (LUT) based
algorithm are two key algorithms for memoryless models.
Memory effects begin to be significant as the signal bandwidth widens. This is
especially true for high power amplifiers used in wireless base stations and for
modulation formats with high PAPR such as WCDMA, mobile WiMAXTM and
3GPP LTE. Some of the causes of memory effects include the thermal constants
of the active devices and components in the biasing network that have frequency
dependent behaviors. As a result, the current output value of the power ampli-
fier starts to depend on a history of past input values, thus existing "memory."
Memoryless linearization techniques that cannot account for these effects can
only offer a limited amount of performance improvement. Therefore, practical
DPD algorithms generally include memory structures.
3
There are two main categories of DPD algorithms that account for memory
effects. The first is based on Artificial Neural Networks or Real-Valued Time
Delay Neural Networks. The second DPD algorithm family is based on the
Volterra series and its derivatives, with Volterra algorithms being the more
general of the two approaches. However, the large number of coefficients of
the Volterra series makes it unattractive for practical applications. In order to
make the pre-distortion more computationally efficient, several algorithms based
on Volterra have been developed, including Wiener, Hammerstein, Wiener