MIMO Demodulation With Auxiliary Dimensions for Probabilistic Shaping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing linear MIMO demodulation techniques fail to fully realize the shaping gain from probabilistic shaping due to wireless fading, leading to performance loss and wasted spectral efficiency in non-uniform modulation constellations.
Innovation Solution
Implementing a transformation of the modulation constellation to a subspace using dimensions associated with a non-uniform probabilistic distribution and adding auxiliary dimensions to account for probabilistic shaping, allowing improved demodulation performance and spectral efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If independent SISO demodulation is used for each layer, then device complexity is reduced, but shaping gain is lost leading to performance degradation
Solution Approach 1:
The patent transforms the modulation constellation to a subspace by adding auxiliary dimensions associated with the non-uniform probabilistic distribution. This dimensional transformation allows the receiver to process multiple layers jointly in an expanded space, recovering the shaping gain that would otherwise be lost in independent SISO demodulation, while maintaining computational feasibility through structured transformation.
2Productivity
If probabilistic shaping is applied to MIMO signals, then spectral efficiency is improved, but existing linear demodulation techniques fail to realize the full shaping gain
Solution Approach 1:
By transforming the modulation constellation to a subspace with added auxiliary dimensions, the patent enables linear MIMO demodulation techniques to properly account for the non-uniform probabilistic distribution. This transformation allows the demodulator to operate in an expanded dimensionality where the shaping gain can be fully realized while maintaining the spectral efficiency benefits of probabilistic shaping.
Solution Approach 2:
The patent modifies the demodulation approach by incorporating dimensions associated with the non-uniform probabilistic distribution parameters. This parameter change enables the linear demodulation technique to adapt to the probabilistic shaping characteristics, allowing full realization of shaping gain without requiring complex non-linear processing.
3Reliability
If modulation constellation is transformed to subspace with auxiliary dimensions, then demodulation performance is enhanced, but computational complexity increases
Solution Approach 1:
The patent segments the demodulation process into distinct stages: transformation of the modulation constellation to subspace, projection of received signals onto the transformed constellation, and determination of nearest constellation points. This segmentation allows each stage to be optimized independently, managing computational complexity while achieving enhanced demodulation performance through the auxiliary dimensions.
Data Source
AI summary
Some wireless communications systems modulate signals for transmission via a channel. Probabilistic shaping may utilize non-uniformly distributed modulation constellations to increase spectral efficiency. In some cases, wireless fading may occur in a multiple-input and multiple-output (MIMO) channel, which may cause a loss of the shaping gain from probabilistic shaping. Some examples of approaches to linear MIMO demodulation for probabilistic shaping using auxiliary dimensions are provided. A wireless device may receive information indicating a non-uniform probabilistic distribution of a modulation constellation for a MIMO signal. One or more symbols of the MIMO signal may be transmitted or received via a channel. The wireless device may demodulate the symbol based at least in part on a transformation of the modulation constellation to a subspace using dimensions associated with the non-uniform probabilistic distribution and a point of the transformed modulation constellation nearest to a projected symbol in the subspace.