.png){kind=icon}
.png){kind=icon}
Return
Nonlinear Equalization by Linearizing Cascaded Amplitude–Phase–Frequency Characteristic Function
.png){kind=icon}
PRE
AI
X
DOI：10.1109/tcomm.2026.3717000.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Conventional nonlinear equalization (NLE) techniques typically rely on iterative procedures to refine equalizer coefficients and/or signal estimates, which limits their practicality in high-speed, low-complexity, and low-latency systems. This paper overcomes these limitations by linearizing the amplitude-phase-frequency (APF) characteristic function (CF) of the complete signal chain. Based on expanded memory polynomial (EMP) modelling and APF characterization, a cascade of a nonlinear channel and a nonlinear equalizer is shown to be an EMP, and the frequency responses in its APF CF expansion are linear combinations of the channel’s expanded frequency responses weighted by the equalizer coefficients. This linear relationship enables a novel non-iterative NLE technique. Several inverse nonlinear system models are analytically developed to further validate the proposed concept. With further proposed practical implementation methods, the non-iterative NLE can be realized at only widely-linear complexity. Simulation results for both single-carrier and multi-carrier systems confirm the effectiveness of the proposed approach, demonstrating its strong potential for emerging sixth-generation millimeter-wave and terahertz communication systems.
Keywords:
Nonlinear equalization
expanded memory polynomial
amplitude–phase–frequency characterization
millimeter wave
terahertz
Journal
IF:
8.3
Papers:
1.2W
Citations:
3.6W
Organization
Cited Papers