Distribution Matching via Reversed Compression for Shaping Gap Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication technologies face challenges in achieving optimal spectral efficiency and achievable information rate due to the 'shaping gap' in signal distribution, particularly in noisy channels like AWGN, where uniform signaling falls short of Shannon Capacity.
Innovation Solution
Implementing a distribution matcher that converts uniformly distributed information bits into non-uniformly distributed symbols using a target probability mass function, such as Maxwell-Boltzmann Distribution, with techniques like probabilistic amplitude shaping and variable-to-fixed schemes to reduce overhead and improve spectral efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If uniform signaling is used for transmission, then the system is simple to implement, but spectral efficiency falls short of Shannon Capacity due to shaping gap
Solution Approach 1:
The patent applies parameter changes by transforming the probability distribution of transmitted symbols from uniform to non-uniform using a target probability mass function (such as Maxwell-Boltzmann distribution). This changes the statistical parameters of the signal to better match the channel capacity characteristics, thereby improving spectral efficiency while maintaining implementation feasibility through established mathematical transformations.
2Productivity
If probabilistic amplitude shaping is implemented to reduce shaping gap, then spectral efficiency improves, but device complexity increases
Solution Approach 1:
The patent introduces a distribution matcher as an intermediary component that sits between the uniform bit source and the modulation scheme. This mediator transforms uniformly distributed bits into non-uniformly distributed symbols according to a target probability mass function, enabling probabilistic amplitude shaping without requiring fundamental changes to the underlying modulation and coding structures.
3Loss of substance
If variable-to-fixed scheme is used to convert information bits to shaped symbols, then overhead is reduced, but processing complexity increases
Solution Approach 1:
The patent employs an inverted processing approach where instead of directly mapping variable-length information bits to fixed-length shaped symbols, it first generates fixed-length shaped symbols from information bits and then truncates or pads as needed. This reverse approach simplifies the variable-to-fixed conversion by leveraging the determinism of fixed-length processing followed by simple truncation operations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Apparatus, methods, and computer-readable media for facilitating distribution matching via reversed compression are disclosed herein. An example method for wireless communication at a wireless transmitter includes decompressing a first sequence of information bits for wireless transmission to output a sequence of shaped symbols. The example method also includes compressing the sequence of shaped symbols to output a second sequence of compressed information bits. The example method also includes transmitting, to a receiver, a signal comprising the sequence of shaped symbols.