Complementary Cumulative Distribution Function (CCDF)

The CCDF curve shows the probability that a signal's instantaneous power exceeds a given threshold relative to its average power. For any power level x dB above average, the CCDF value represents the fraction of time the signal envelope spends above that level. A steeper curve indicates a lower PAPR, while a shallower tail reveals frequent high-amplitude excursions that stress power amplifier linearity.

CCDF is measured using a vector signal analyzer or high-speed oscilloscope capturing I/Q samples over millions of data points. The instrument computes the instantaneous power for each sample, normalizes to the average power, and constructs a histogram of power levels. The complementary cumulative sum of this histogram produces the CCDF curve. Modern analyzers display this in real-time, overlaying reference curves for common modulation formats like OFDM and 256-QAM.

Engineers use the CCDF to determine the required power amplifier back-off. A common reference point is the 0.01% probability level (10⁻⁴ on the CCDF y-axis). The power level at this probability defines the practical PAPR that the amplifier must accommodate. For example, an LTE downlink signal might show 9.5 dB at 0.01%, meaning the PA needs at least 9.5 dB of back-off to avoid compressing 99.99% of signal peaks.

CCDF curves are the primary tool for evaluating Crest Factor Reduction effectiveness. By overlaying CCDF plots before and after CFR processing, engineers quantify PAPR reduction gain at specific probability points. A well-designed CFR algorithm shifts the CCDF curve leftward—reducing the power level at the 10⁻⁴ probability point—while monitoring EVM degradation and ACLR regrowth to ensure the trade-off between efficiency and signal quality remains acceptable.

Different modulation schemes produce distinct CCDF signatures:

• Single-carrier QPSK: Relatively low PAPR, narrow CCDF spread
• OFDM with 1024 subcarriers: High PAPR, broad CCDF tail extending beyond 12 dB
• DFT-spread-OFDM: Lower PAPR than conventional OFDM due to precoding
• Filtered OFDM: Sharper CCDF roll-off from reduced out-of-band emissions These curves guide waveform selection for power-constrained applications.

While CCDF captures the statistical distribution of instantaneous power, the Cubic Metric (CM) extends this analysis by weighting the envelope distribution according to third-order nonlinearity. CM estimates the actual power de-rating required for a given PA technology by correlating the CCDF shape with measured amplifier distortion characteristics. This provides a more accurate efficiency prediction than PAPR alone, especially for GaN Doherty amplifiers with complex nonlinear behavior.