Differential QPSK Soft Bit Metrics With Simplified mLLR Computation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communications systems, such as differentially encoded QPSK, face challenges in efficiently generating soft bit metric information, which is crucial for superior performance but requires complex computations, especially in noisy environments.
Innovation Solution
A simplified method to compute modified logarithmic-likelihood ratios (mLLR) by identifying dominant terms and approximating soft metrics, reducing computational complexity while maintaining performance comparable to optimal LLRs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If optimal LLR computation is used for soft decision decoding, then decoding performance is improved, but computational complexity increases significantly
Solution Approach 1:
The patent transforms the optimal LLR computation problem by changing the parameter representation from exact logarithmic-likelihood ratios to modified LLRs (mLLRs) with simplified exponent structures. This parameter transformation maintains the essential decoding information while enabling computationally efficient approximation through dominant term identification.
Solution Approach 2:
The patent employs approximate mLLR values instead of exact LLR computations, sacrificing minimal performance for significant computational savings. The approximation uses simplified exponential terms that are computationally cheaper to evaluate, effectively replacing complex calculations with lighter computational operations.
2Device complexity
If simplified approximation methods are used for soft metric generation, then computational complexity is reduced, but performance degradation occurs
Solution Approach 1:
The patent computes only the dominant terms in the LLR expansion rather than all terms, applying partial action principle. By identifying and computing only the most significant exponential terms that contribute most to the LLR value, the method achieves sufficient accuracy for reliable decoding while avoiding computation of less significant terms.
Solution Approach 2:
The patent modifies the LLR parameter structure to mLLR with a specific form that separates the dominant exponential terms from less significant components. This parameter change enables the approximation method to focus computational effort on the most influential terms while maintaining performance close to optimal LLR computation.
3Reliability
If differential encoding is used for phase noise robustness, then system robustness is improved, but soft bit metric generation becomes more complex
Solution Approach 1:
The patent segments the soft metric generation process into distinct stages: differential decoding to recover bit estimates, identification of dominant terms based on these estimates, and computation of mLLR values. This segmentation allows the complex differential encoding/decoding operation to be handled separately from the soft metric computation, simplifying the overall process.
Solution Approach 2:
The patent introduces mLLR as an intermediary parameter between the differential decoded bits and the final soft metrics required by the decoder. This intermediary representation simplifies the transformation from differential encoded symbols to soft bit metrics, bridging the gap between the robust differential encoding scheme and the soft decision decoding requirement.
Data Source
AI summary
A computer implemented method for generating soft bit metric information of telecommunications systems employing differential encoding of data.