How to Characterize RF Distortion – Part 2: Making Distortion Measurements
2019-10-08 | 6 min read
In an earlier post, “How to Characterize RF Distortion – Part 1: CCDF Measurements,” I discussed how to evaluate waveform designs and statistical analysis of the signal power levels with complementary cumulative distribution function (CCDF) power curves. This is an essential step in characterizing the power distributions of digital modulation signals.
In this post, I will discuss common distortion types and their measurements in the frequency domain.
Distortion is the alteration of the original waveform. For a linear device, the input and output frequencies are the same; there are no additional frequencies created. The output signal only has amplitude and phase change. For a nonlinear device, the output may have a frequency shift or additional frequencies. Nonlinear distortion is typically unwanted and R&D engineers strive to minimize it. There are three major types of nonlinear distortion measurements in the frequency domain — harmonic, intermodulation distortion, and adjacent channel power (ACP) measurements.
The amplitude transfer characteristics of a circuit or device cannot precisely track the input signal. The amplitude shifts generate higher frequency components at integer multiples of the input signal.
The most straightforward method for measuring harmonic distortion is to use a continuous wave (CW) tone as an input signal, and measure the output signal with a signal analyzer; see Figure 1. A device under test (DUT) might be an RF amplifier or mixer.
Figure 1. Harmonic distortion measurement setup
The signal analyzer sets to zero-span, which enables a time-domain power measurement to measure the power level at the fundamental and harmonic frequencies as shown to the right of Figure 2. Problems arise when the DUT has distortion product levels that approach the internally generated distortion product levels of the signal analyzer, from the sixth to tenth harmonics in this example. For more information on optimizing dynamic range for distortion measurements, please refer to the application note, “Optimizing Dynamic Range for Distortion Measurements.”
Figure 2. Harmonics measurement with Keysight X-Series signal analyzer
Third-order intermodulation distortion
Two-tone, third-order intermodulation (TOI) distortion is a common test for RF distortion measurements. When two or more signals are present in a non-linear system, they can interact and create additional components at the sum and difference frequencies of the original frequencies, and at sums and differences of multiples of those frequencies. Figure 3 below shows the two-tone third-order Intermodulation measurement setup. The DUT could be an amplifier or a mixer.
Figure 3. Two-tone intermodulation distortion measurement setup
Figure 4 illustrates TOI measurements with a signal analyzer. The two test tones are at frequencies 995 MHz and 1005 MHz. The third-order intermodulation products occur at frequencies 985 MHz and 1015 MHz and the TOI measurement results are 31.7 dBm (lower) and 32.2 dBm (upper).
Figure 4. TOI measurement with a Keysight X-Series signal analyzer
For production testing, a vector signal generator alone can be used to generate two test tones using the internal baseband generator to save costs. Keysight offers an advanced correction routine which can suppress distortion products generated by the signal generator itself or an external pre-amplifier. To learn about how to configure distortion-free two-tone and multitone test signals, download the technical overview “N7621B Signal Studio for Multitone Distortion.”
Adjacent channel power (ACP)
Wider bandwidths and multi-carrier techniques are used broadly to increase data throughput for the latest wireless standards. The two-tone, third-order intermodulation technique does not completely characterize the behavior of wide-bandwidth components. Digital modulation, which uses both amplitude and phase shifts, generates distortion, also known as spectral regrowth. Figure 5 shows the spectral regrowth (red curve) of a digital modulation signal.
Figure 5. Spectrum regrowth due to intermodulation
Like third-order intermodulation (TOI), the spectrum regrowth also interferes with the adjacent channels and spreads energy outside the main channel. ACP measurement describes the ratio of power in a modulated signal vs. power emitted into an upper or lower adjacent channel. ACP measurements provide useful information for spectrum regrowth and emissions to characterize the transmitter design, including baseband filter and nonlinear distortion. Figure 6 shows the ACP measurement results of a 5G new radio (NR) signal with a 100 MHz channel bandwidth. Both the lower and upper ACP ratios are -50.28 dBc.
Figure 6. ACP measurement with Keysight X-Series signal analyzer
Characterizing the distortion performance of an RF component requires a test system with low distortion that will not mask the DUT’s actual performance. Examine the distortion’s impacts on the frequency spectrum with harmonic distortion, TOI distortion, and ACP measurements.
In the next post, I will discuss demodulation analysis that can help you perform advanced characterization and troubleshoot your RF designs.