Signal Envelope Statistics

The Probability Density Function describes the likelihood of a signal's instantaneous amplitude taking on a specific value at any given moment. For a complex baseband signal, the envelope PDF is fundamental to understanding amplifier stress.

• Gaussian signals (e.g., OFDM with many subcarriers) exhibit a Rayleigh-distributed envelope
• The PDF shape directly predicts how often peaks occur and their typical magnitude
• Narrowband single-carrier modulations produce markedly different PDFs than wideband multi-carrier waveforms
• Example: A 4G LTE 20 MHz downlink signal closely approximates a Rayleigh envelope due to the central limit theorem acting on hundreds of active subcarriers

The CCDF curve is the industry-standard visualization for PAPR behavior, showing the probability that a signal's instantaneous power exceeds a given threshold relative to its average power. It is the integral of the PDF tail.

• The x-axis represents the PAPR threshold in dB above average power
• The y-axis represents the probability (often plotted on a logarithmic scale from 10⁰ to 10⁻⁷)
• A 10⁻⁴ probability at 10 dB PAPR means peaks exceed 10 dB above average for 0.01% of the time
• CCDF curves are measured directly on vector signal analyzers and used to specify CFR performance targets
• Key reference point: 3GPP specifications often reference the 10⁻⁴ probability point for conformance testing

The Crest Factor is the ratio of the peak amplitude to the RMS value of a waveform, expressed as a dimensionless ratio or in dB. For voltage signals, it is the square root of the PAPR.

• Crest Factor (dB) = PAPR (dB) / 2 for voltage-domain analysis
• A pure sine wave has a crest factor of √2 (3.01 dB)
• OFDM signals can exhibit crest factors exceeding 12 dB before CFR is applied
• Crest factor directly determines the headroom required in digital-to-analog converters and amplifier stages
• Practical impact: A crest factor of 4 (12 dB) means the PA must handle instantaneous voltages 4× the RMS level without clipping

The Cubic Metric is a figure of merit that estimates the power de-rating (back-off) required for a power amplifier to handle a given signal while maintaining acceptable linearity. It accounts specifically for third-order nonlinearity, the dominant distortion mechanism.

• Derived from the third-order moment of the signal envelope, not just the peak-to-average ratio
• More accurate than PAPR alone for predicting PA back-off because it captures the distortion energy generated by amplitude fluctuations
• Standardized by 3GPP for comparing the PA impact of different modulation and resource block configurations
• Example: Two signals with identical PAPR can have different Cubic Metrics if their envelope distributions differ in the mid-amplitude range where third-order distortion is most active
• Used during waveform design to minimize required PA back-off without full hardware testing

Envelope cross-correlation quantifies the statistical similarity between the amplitude variations of two signals, such as the original waveform and its CFR-processed version, or between signals in different transmit paths.

• A correlation coefficient near 1.0 indicates the processed envelope closely tracks the original, preserving modulation integrity
• Low correlation after CFR suggests excessive distortion that will degrade Error Vector Magnitude (EVM)
• Used in MIMO systems to assess whether beamforming weights alter the composite envelope statistics at each antenna element
• Multi-band DPD applications use cross-correlation to detect and cancel inter-band envelope interactions
• Practical threshold: Correlation above 0.98 is typically required to ensure CFR does not impair demodulation performance

The statistical moments of the signal envelope distribution provide compact numerical descriptors that drive CFR and DPD algorithm design beyond what PAPR alone reveals.

• First moment (mean): The average envelope amplitude, related to average transmit power
• Second moment (variance): The spread of amplitudes around the mean, proportional to average power
• Third moment (skewness): Asymmetry in the amplitude distribution; signals with heavy peak-side tails exhibit positive skew
• Fourth moment (kurtosis): Measures the "peakedness" of the distribution; high kurtosis indicates frequent extreme peaks relative to a Gaussian distribution
• Application: Kurtosis is particularly useful for detecting and characterizing impulsive noise or rare high-amplitude events that dominate PAPR but are invisible to variance-based metrics

The fundamental metric quantifying the ratio of instantaneous peak power to average power in a signal envelope. A high PAPR forces power amplifiers to operate with significant power back-off to avoid nonlinear distortion.

• Expressed in dB: PAPR(dB) = 10·log₁₀(P_peak / P_avg)
• OFDM signals typically exhibit 8-13 dB PAPR
• Directly determines amplifier efficiency degradation
• Measured statistically via CCDF curves rather than single-point values

The statistical curve showing the probability that a signal's instantaneous power exceeds a specified threshold relative to its average power. CCDF is the primary design tool for specifying CFR requirements.

• X-axis: power threshold above average (dB)
• Y-axis: probability of exceeding threshold (10⁻¹ to 10⁻⁷)
• The 10⁻⁴ probability point is commonly used for PAPR specification
• Reveals how often extreme peaks occur, not just their magnitude

The ratio of peak amplitude to RMS value of a waveform, equivalent to the square root of PAPR for voltage signals. Crest factor directly impacts ADC/DAC dynamic range requirements in the transmit and receive chains.

• CF = V_peak / V_rms
• CF(dB) = PAPR(dB) / 2
• Determines headroom requirements in digital signal processing chains
• Critical for preventing numerical overflow in fixed-point FPGA implementations

A figure of merit that estimates the power de-rating required for a power amplifier to handle a given signal's envelope statistics, specifically accounting for third-order nonlinearity. More accurate than PAPR alone for predicting amplifier back-off.

• Derived from the normalized third-order moment of the signal envelope
• Correlates better with measured PA efficiency than raw PAPR
• Used in 3GPP specifications for UE power class determination
• Accounts for the amplifier's cubic gain compression characteristic

The probability density function describing the distribution of instantaneous signal amplitudes. The shape of the PDF determines which CFR techniques will be most effective.

• Gaussian-distributed I/Q components produce Rayleigh-distributed envelope magnitudes
• OFDM signals with many subcarriers approach a complex Gaussian distribution by the Central Limit Theorem
• PDF tails represent the high-amplitude peaks targeted by CFR
• Enables analytical derivation of clipping noise power and EVM degradation

The intentional reduction of input drive level to operate a power amplifier in its linear region, directly proportional to the signal's PAPR. Each dB of back-off reduces efficiency, making PAPR reduction economically critical.

• Output back-off (OBO) = P_sat(dBm) - P_operating(dBm)
• Class-AB PAs may require 6-10 dB back-off for OFDM signals
• Doherty architectures reduce back-off requirements through load modulation
• Envelope tracking can recover efficiency lost to back-off