Broadcast Transmitter Mapping Using Non-Uniform QAM Constellations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current broadcasting systems, such as DVB-T2, utilize uniform QAM constellations that leave a significant gap from the theoretical Shannon limit, resulting in suboptimal bit error rate (BER) and frame error rate (FER) performance.
Innovation Solution
The implementation of non-uniform constellations (NUC) that relax the rectangular shape and uniform spacing properties of traditional uniform QAM, optimizing constellation design for specific signal-to-noise ratios (SNR) and fading channels to enhance BER and FER performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If uniform QAM constellation is used, then mapping and demapping is easy, but BER and FER performance is suboptimal with significant gap from Shannon limit
Solution Approach 1:
The patent applies local quality by transitioning from uniform spacing to non-uniform spacing in the QAM constellation, where different regions of the constellation have different point densities optimized for specific SNR ranges. This allows the system to achieve better BER performance in specific channel conditions while maintaining manageable complexity through targeted optimization rather than universal uniformity.
Solution Approach 2:
The patent changes the fundamental parameter of constellation point spacing from uniform to non-uniform, and further to probabilistic shaping where the distribution of constellation points follows a specific probability distribution. This parameter change enables the system to approach the Shannon limit by optimizing the information density across different constellation regions, thereby improving reliability without completely sacrificing operational simplicity.
2Reliability
If non-uniform constellation is used, then BER and FER performance is improved, but mapping and demapping complexity increases
Solution Approach 1:
The patent manages complexity by changing the parameter representation from geometric coordinates to probability-based indexing. The transmitter uses a probabilistic mapping table that stores pre-computed constellation point assignments, while the receiver uses corresponding lookup tables. This parameter transformation maintains manageable device complexity through efficient data structures while achieving the reliability benefits of non-uniform constellations.
3Productivity
If QAM size is increased to increase throughput, then throughput increases, but gap from Shannon limit remains significant
Solution Approach 1:
The patent applies local quality by optimizing different regions of the QAM constellation for different purposes: inner regions with higher reliability for critical data bits and outer regions with higher throughput contribution for less critical bits. This regional optimization allows the system to simultaneously achieve high throughput and reduced gap to Shannon limit by strategically allocating constellation points based on their information importance and channel conditions.
Solution Approach 2:
The patent introduces dynamics by making the constellation mapping adaptive to channel conditions through probabilistic shaping. The mapping strategy dynamically adjusts the probability of selecting different constellation points based on current SNR conditions, allowing the system to optimize both throughput and reliability in real-time rather than being fixed to a static uniform QAM configuration.
Data Source
AI summary
A transmitting apparatus is disclosed. The transmitting apparatus includes an encoder to perform channel encoding with respect to bits and generate a codeword, an interleaver to interleave the codeword, and a modulator to map the interleaved codeword onto a non-uniform constellation according to a modulation scheme, and the constellation may include constellation points defined based on various tables according to the modulation scheme.