.png){kind=icon}
.png){kind=icon}
Return
Low-Complexity Demapping Algorithm for Two-Dimensional Non-Uniform Constellations
.png){kind=icon}
OA
AI
DOI：10.1109/TBC.2015.2492477.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Non-uniform constellations (NUCs) have been recently introduced in digital broadcasting systems to close the remaining gap to the unconstrained Shannon theoretical limit. Compared to uniform quadrature amplitude modulation (QAM) constellations, NUCs provide a signal-to-noise ratio (SNR) gain (i.e., a reduction in the required SNR), especially for high-order constellations. One-dimensional NUCs (1D-NUC) have a squared shape with non-uniform distance between the constellation symbols. Since the I and Q components remain as two independent signals, a 1D-demapper as for uniform QAM constellations is feasible. Two-dimensional NUCs (2D-NUC) provide a better performance than 1D-NUCs, since they are designed by relaxing the square shape constraint, with arbitrary shape along the complex plane. However, the main drawback of 2D-NUCs is the higher complexity at the receiver, since a 2D-demapper is needed. In this paper, we propose a demapping algorithm that reduces from 69% to 93% the number of required distances when using 2D-NUCs. The algorithm discards or replicates those constellation symbols that provide scarce information, with a performance degradation lower to 0.1 dB compared to the optimal maximum likelihood demapper.
Keywords:
Broadcasting
non-uniform constellations
demapping complexity
ATSC 3.0
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.8
Papers:
2.1K
Citations:
3.0K