Accurate Prediction of Nonlinear Distortion of Multi-Carrier Signals

Cameron M. Pike, Brad Oney, Gabriel Hepner, Animesh Yadav · 2025

Read Accurate Prediction of Nonlinear Distortion of Multi-Carrier Signals by Cameron M. Pike, Brad Oney, Gabriel Hepner, Animesh Yadav (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-23844.txt.

Introduction

Cameron M. Pike and three co-authors sharpen a method for predicting the nonlinear distortion that power amplifiers inflict on multi-carrier signals. Distortion matters because system designers must trade output power against network throughput, and guessing wrong is costly. Their approach uses the series form of the characteristic function method, which computes signal and distortion components separately and efficiently.

On the theory side, they model the signal as a bandpass Gaussian process that gets hard limited. In that case their method reproduces exactly what the classical Price theorem gives, a clean validation. On the practice side, they feed a fragmented-spectrum OFDM signal into an amplifier driven into compression, where the classical theorem is hard to apply. The predicted output spectrum matches laboratory measurements well. The method also yields signal-to-distortion ratios across settings like overdrive level, and it exploits the fast Fourier transform for speed.

Catalog record

CREATORCameron M. Pike, Brad Oney, Gabriel Hepner, Animesh Yadav
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE24 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-23844

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Accurate Prediction of Nonlinear Distortion of Multi-Carrier Signals Cameron M. Pike, Brad Oney, Gabriel Hepner, Animesh Yadav Abstract Nonlinearities in power amplifiers adversely affect multi-carrier modulation techniques. Accurate prediction of nonlinear distortion is essential for making design trade-offs between output power and network throughput. We use the series form of the characteristic function (ch.f.) method to predict distortion spectra for sparse multi-carrier transmissions. This method results in efficient calculations of individual signal and distortion components. The method is validated both theoretically and practically. Theoretical validation is performed by modeling the signal as a bandpass Gaussian process that is hard limited, and it is shown that the series ch.f. method produces results that are identical with the classical Price’s theorem. Practical validation is shown by considering an orthogonal frequency division multiplexing (OFDM) signal with a fragmented spectrum which is then applied to an amplifier driven into compression for which application of Price’s theorem is difficult, and the predicted output spectrum corroborates laboratory measurements. Part of the computational efficiency is realized in that the nonlinearity can be expressed as the fast Fourier transform (FFT) of samples of its forward scattering parameter (i.e., S21) or transconductance function (including AM-PM effects), and distortion contributions of the signal can be expressed as numerical autoconvolutions of the clean spectrum. Signal-to-distortion ratio (SDR) can be easily computed and parameterized across variables of interest, such as overdrive level. I Introduction Multi-carrier modulation (MCM) techniques are essential building blocks for future wireless networks. The evolution from orthogonal frequency division multiplexing (OFDM) to more versatile techniques, such as orthogonal time-frequency and space (OTFS), reflects the continuous innovation at the physical layer, aiming to unlock new applications and ensure robust wireless connectivity in all scenarios. However, in MCM techniques, nonlinear distortion from power amplifiers significantly degrades performance, primarily due to the high peak-to-average power ratio (PAPR) character of their waveforms. The impact of nonlinear distortion on bit error rate (BER) performance for the case of OFDM and OTFS is discussed in [ 1 , 2 , 3 , 4 ] . Therefore, characterizing nonlinearity is crucial as it aids in developing countermeasures for various communication systems [ 5 , 6 ] , and device authentication applications through RF fingerprinting [ 7 ] . In this letter, our focus is on developing and validating an efficient method for accurately predicting the shape and power of the signal and all distortion spectral components at the output of a nonlinear system fed by a multi-carrier signal. These predictions can be used to estimate the impact of clipping on system performance, including signal-to-distortion ratio (SDR) and BER, and thereby inform a trade study to reduce the cost of a power amplifier by relaxing its requirements, or to increase the output power for an existing amplifier, while accounting for the distortion loss in the receiver’s link budget. Particularly, our statistical signal processing approach extends Rice’s characteristic function (ch.f.) method [ 8 , §4.8] by using a Fourier series representation of the nonlinearity [ 9 , 10 ] , (contrast [ 11 ] ). Consequently, distortion terms take the form of autoconvolutions of the signal spectrum, and the total spectrum is found by a weighted sum of signal and distortion terms. While Rice’s method is effective for stochastic processes of any distribution, as the ch.f. of a wide-sense stationary random process always exists [ 12 , p.72] , it requires evaluating many nested improper integrals, which discourages its widespread use. Our approach offers several key benefits: i) it is more amenable to numerical evaluation by eliminating the troublesome nested integrals (which also appear in Price and Bussgang [ 13 , 14 , 15 , 16 ] ), ii) it can be used for multiple signal components, or signals plus channel and/or phase noise, as shown in [ 9 , 10 ] , iii) it can be applied to a wide range of systems, including very-high-frequency (VHF), ultra-high-frequency (UHF), and wideband millimeter-wave amplifiers, and iv) it is broadly applicable to various signals and is especially helpful for signals with complicated spectra that are challenging for other methods, whether spectrum-based or discrete inter-modulation product (IMP) accounting and truncated polynomial models of the nonlinearity , as in [ 17 ] . The summary of the contributions presented in this letter is as follows: • We theoretically validate the extended ch.f. method introduced in [ 9 ] by showing that it produces results identical to the widely-used Price’s theorem [ 13 ] for a classical nonlinearity with a Gaussian bandpass signal. • We apply the method in the laboratory to a case of a real OFDM signal and amplifier, and show that the steps involved in the method are straightforward to apply, utilizing instrument measurements of both the amplifier transfer function and the undistorted spectrum of the signal. The outline of this paper is as follows. Section II presents the signal model and problem statement. Section III details the extended ch.f. method. Section IV compares the classical result for a hard-limited bandpass Gaussian signal. Section V presents and discusses the laboratory-based experimental results. Finally, Section VI concludes the paper. II System Model and Problem Statement II-A Signal Model We model the OFDM signal s(t) as a stationary random process with root-mean-square (RMS) amplitude \sigma and correlation coefficient \rho(\tau) , with time lag \tau . \rho(\tau) incorporates transmit pulse shape, center frequency, subcarrier occupancy, and relative power levels. For example, if the baseband pulse shape has an autocorrelation function \phi(\tau) , and MCM carriers are at center frequencies f_{k} and amplitudes A_{k} , then \rho(\tau)\propto\phi(\tau)\sum_{k}{A_{k}\cos{2\pi f_{k}\tau}} . As we will show in the subsequent section, there is no need to explicitly specify \phi(\tau) or f_{k} , but we work directly with \rho(\tau) or its transform, the spectrum of the signal. The OFDM signal is then passed through a non-ideal power amplifier or general nonlinear system to obtain the amplified signal y(t) for transmission. Our objective for this system is to accurately compute the spectrum, transmit power, and distortion at the output of the power amplifier. To achieve this, we need to determine the output correlation function, as detailed in the next section. II-B Problem Statement As an analysis aid, we treat the input as two different signals in two different nonlinearities, which we then unify as a special case of a single signal and nonlinearity. To this end, we begin by expressing two sample functions from the random process s(t) as s_{1}(t) and s_{2}(t) , stationary jointly random processes with zero mean, variances \sigma_{1}^{2},\,\sigma_{2}^{2} , and correlation coefficient \rho(\tau)=\mathbb{E}\{s_{1}(t+\tau)s_{2}(t)\}/\sigma_{1}\sigma_{2} . Assuming an optional dc bias \beta on the input signal, the inputs to the nonlinear amplifier are x_{1}(t)=s_{1}(t)+\beta;\quad x_{2}(t)=s_{2}(t)+\beta. (1) Next, representing the nonlinear memoryless 1 1 1 See Section V-B distortion by f(x) and g(x) , the output of the power amplifier is y_{1}(t)=f\left(x_{1}(t)\right);\quad y_{2}(t)=g\left(x_{2}(t)\right). (2) The expression for the output correlation can be written by taking the expectation of the product of the outputs as \displaystyle\Psi(\tau) \displaystyle= \displaystyle\mathbb{E}\left\{y_{1}(t+\tau)\,y_{2}(t)\right\}, (3) \displaystyle= \displaystyle\mathbb{E}\left\{f\left(s_{1}(t+\tau)+\beta\right)\,g\left(s_{2}(t)+\beta\right)\right\}. (4) III Proposed Methodology We introduce the method by stating it as a theorem, the proof of which is provided by derivation in this section. Theorem 1 If the general characteristic function of an input signal ensemble \{V(t)\in\mathbb{R}\|\,{|V(t)|<c}\} is

Read the complete text