Characterization of power spectral density for nonlinearly amplified OFDM signals based on cross-correlation coefficient

Abstract

Orthogonal frequency division multiplexing (OFDM) has been adopted in many modern communication systems due to its robustness against frequency-selective fading channels as well as its near-rectangular spectrum that can achieve high spectral efficiency. However, its major drawback is the resulting signal with high peak-to-average power ratio (PAPR), which causes severe nonlinear distortion at the power amplifier (PA) unless input backoff is chosen sufficiently large. The effect of the nonlinear distortion is two-fold: out-of-band radiation and signal quality degradation. The former causes adjacent channel interference and thus degrades the bandwidth efficiency. The latter affects the system level performance and is often measured by the error vector magnitude (EVM). It is thus important for the system designer to analyze the nonlinear distortion caused by a given PA in terms of power spectral density (PSD) and EVM, but accurate calculation of these characteristics may be generally involved. In this work, by establishing the link between the cross-correlation coefficient of the input and output signals from PA and the resulting PSD, we characterize the in-band and out-of-band distortion of nonlinearly amplified OFDM signals based exclusively on the cross-correlation coefficient. The accuracy of the proposed approach is confirmed by both simulation and measurement using a real PA.

On Spectral Efficiency of OFDM Signals Based on Windowing

“…We summarize known theoretical expressions over nonlinear channels (see [20,21] and the references therein for details). Under the assumption that s(t) is modeled as a complex Gaussian random process with its PSD given by P s (f ), the PSD of ŝ(t) can be expressed as…”

Performance Analysis and Comparison of Clipped and Filtered OFDM Systems With Iterative Distortion Recovery Techniques

“…In this section, we revisit the performance analysis of clipped and filtered OFDM signals. Specifically, we develop an SDR expression of clipped and filtered OFDM signals based on the classical analysis on nonlinear transformation of Gaussian processes [22], assuming that OFDM signal is modeled by a band-limited complex Gaussian random process following [23], [24]. Furthermore, we derive a simple SER lower bound that serves as a benchmark of the OFDM with CAF (Theorem 1).…”

“…With reference to (3), let z(t) denote a complex Gaussian random process that will be input to the soft-envelope limiter andz(t) denote its output. The power spectral density (PSD) ofz(t), Sz(f ), can be expressed as [24,Eq. (19)]…”

“…As an example, the first several terms of δ n,in can be calculated as 36 288 [24]. The numerical values of δ n,in with n ∈ [5,10] are listed in Table I.…”

Nonlinear Effects in NOMA Systems Using Single Carrier Modulations: Performance Evaluation and Receiver Design

“…The error vector magnitude (EVM) is another measure that evaluates the performance degradation caused by HPA. For instance, if the desired peak-to-average power ratio (PAPR) threshold is low, this leads to an increase in the EVM, penalizing the receiver performance [23].…”