OFDM PAPR Signature

By the Central Limit Theorem, an OFDM time-domain signal with a large number of independent subcarriers converges to a complex Gaussian distribution. The in-phase and quadrature components become normally distributed, producing a Rayleigh-distributed envelope. Single-carrier modulations like QPSK or GMSK exhibit fundamentally different amplitude histograms—constant-envelope or discrete-level distributions—making this statistical signature a primary discriminative feature.

• OFDM: Continuous Rayleigh envelope distribution with long tails
• Single-carrier PSK: Constant envelope (theoretical PAPR of 0 dB)
• Single-carrier QAM: Discrete amplitude levels with bounded peaks

The CCDF curve plots the probability that the PAPR exceeds a given threshold, providing a compact visual signature for waveform identification. OFDM signals exhibit a characteristically slow-rolling CCDF due to the high probability of large amplitude excursions. A typical OFDM signal has a PAPR exceeding 10 dB at the 10⁻³ probability level, while filtered single-carrier signals show a much steeper drop-off.

• OFDM CCDF: Gradual decay, PAPR > 10 dB at 10⁻³ probability
• SC-QAM CCDF: Sharp knee, PAPR typically 4–7 dB at 10⁻³
• GMSK CCDF: Near-vertical drop, PAPR < 3 dB

The PAPR distribution scales predictably with the number of active subcarriers. As the subcarrier count increases, the peak power grows linearly while the average power remains constant, shifting the CCDF curve rightward. This relationship enables blind estimation of the FFT size from the PAPR signature alone. A 64-subcarrier OFDM signal (WiFi 802.11a/g) exhibits a measurably different PAPR profile than a 2048-subcarrier LTE downlink signal.

• 64 subcarriers (WiFi): Median PAPR ~8.5 dB
• 2048 subcarriers (LTE 20 MHz): Median PAPR ~10.5 dB
• 4096 subcarriers (5G NR 100 MHz): Median PAPR ~11.2 dB

Many operational OFDM systems employ PAPR reduction techniques that leave identifiable artifacts in the amplitude distribution. Clipping and filtering creates a hard ceiling in the CCDF curve with spectral regrowth. Selected Mapping (SLM) and Partial Transmit Sequence (PTS) methods produce subtle side-information signatures. The presence and type of PAPR reduction can itself serve as a protocol fingerprint, distinguishing LTE (which uses DFT-s-OFDM for uplink) from pure CP-OFDM systems.

• Hard clipping: Abrupt CCDF truncation at a specific dB threshold
• DFT-s-OFDM: Single-carrier-like PAPR with OFDM frame structure
• Tone Reservation: Reserved subcarriers visible in spectral analysis

The excess kurtosis of the time-domain signal magnitude provides a single scalar feature for discriminating OFDM from single-carrier waveforms. OFDM signals exhibit kurtosis values near 3.0 (mesokurtic, consistent with Gaussian processes), while constant-envelope modulations show kurtosis significantly below 3.0. This computationally efficient metric can be calculated on short observation windows, making it suitable for real-time classification front-ends.

• OFDM (N ≥ 64): Kurtosis ≈ 2.95–3.05
• QPSK: Kurtosis ≈ 1.0 (sub-Gaussian)
• 16-QAM: Kurtosis ≈ 1.32
• 64-QAM: Kurtosis ≈ 1.38

The Clipping Ratio (CR), defined as the peak amplitude divided by the RMS amplitude, produces a characteristic distribution for OFDM signals that differs markedly from single-carrier schemes. For a given observation window, OFDM's CR exhibits higher variance and a larger mean value. Statistical hypothesis testing on the CR distribution—using the Kolmogorov-Smirnov test against known Gaussian references—provides a robust blind classifier that is agnostic to symbol rate and carrier frequency.

• OFDM CR (99.9th percentile): 3.7–4.2× RMS
• QPSK CR (99.9th percentile): 1.4–1.6× RMS
• GMSK CR (99.9th percentile): 1.2–1.4× RMS

The unique spectral correlation pattern generated by the cyclic prefix and pilot subcarriers in OFDM signals. Unlike stationary noise, OFDM exhibits periodicity in its autocorrelation function at specific cycle frequencies.

• Exploited for robust signal detection under low SNR conditions
• Enables blind estimation of symbol rate and cyclic prefix length
• Discriminates OFDM from single-carrier modulations with high confidence

A blind OFDM detection method that exploits the autocorrelation introduced by the cyclic prefix. Since the CP is a copy of the end of the OFDM symbol, correlating the received signal with a delayed version of itself reveals a peak at the symbol duration lag.

• Estimates symbol timing and carrier frequency offset without prior knowledge
• The correlation lag profile reveals the CP length for normal vs. extended mode classification
• Forms the basis of the Schmidl-Cox algorithm for synchronization

A two-dimensional function measuring the correlation between spectral components of an OFDM signal at different frequencies. This reveals cyclostationary features used for blind parameter estimation.

• Peaks in the SCD plane correspond to subcarrier spacing and cyclic prefix rate
• Robust against stationary noise and narrowband interference
• Computationally intensive but provides a rich feature set for deep learning classifiers

A neural network model, typically a convolutional neural network (CNN), trained on IQ samples or spectrograms to automatically identify OFDM variants and their physical-layer parameters without explicit feature extraction.

• Learns hierarchical representations from raw IQ data or time-frequency transforms
• Classifies LTE, 5G NR, WiFi, and proprietary OFDM waveforms
• Requires large labeled datasets but generalizes across varying channel conditions

The identification of specific OFDM implementation details—such as pilot patterns, preamble structures, and frame timing—to determine the wireless standard or vendor-specific configuration of a transmitter.

• Distinguishes between CP-OFDM and DFT-s-OFDM uplink schemes
• Identifies numerology (subcarrier spacing, slot format) in 5G NR signals
• Enables Physical Cell Identity (PCI) extraction from synchronization signals

A time-frequency analysis technique that identifies the energy ridges corresponding to individual subcarriers in an OFDM signal. By analyzing the spacing and duration of these ridges, key parameters can be visually estimated.

• Enables estimation of subcarrier spacing and OFDM symbol duration
• Effective for signals with high PAPR where individual subcarriers are distinguishable
• Often used as a preprocessing step before FFT size detection