2×2 MIMO Soft-Decision Detection With Low-Complexity LLR Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 2×2 MIMO systems face challenges in implementing soft-decision detection for MDCM signals due to high computational complexity, particularly with ML schemes, and the inability of SD schemes to generate necessary LLR values, which hinders efficient detection in high-speed transmission modes like 1280+ Mbps.
Innovation Solution
A method and apparatus for soft-decision detection in 2×2 MIMO systems that calculates LLR values by detecting candidate symbol vector sets with real and imaginary parts, using QR decomposition to convert received signals into upper triangular matrices, and selecting symbols to minimize Euclidean distances, thereby reducing complexity and enabling LLR calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ML scheme is used for soft-decision detection in 2×2 MIMO system, then detection performance is optimized, but computational complexity becomes too high for practical implementation
Solution Approach 1:
The patent segments the detection process into two distinct phases: a forward detection phase that processes symbols in forward order to generate initial LLR values, and a backward detection phase that processes symbols in backward order to refine and correct LLR values. This segmentation divides the complex ML detection into manageable stages, reducing overall computational complexity while maintaining detection performance.
Solution Approach 2:
The forward detection phase performs preliminary detection and generates initial LLR values before the backward detection phase refines them. By performing preliminary action in the forward phase, the system prepares data structures and intermediate results that facilitate more efficient final detection in the backward phase, reducing total computational burden.
2Device complexity
If SD scheme is used for detection in 2×2 MIMO system, then computational complexity is reduced, but ability to generate LLR values for soft-decision is lost
Solution Approach 1:
The patent implements a dynamic detection approach where the system adaptively performs both forward and backward detection phases, dynamically generating and updating LLR values throughout the process. This dynamic mechanism ensures LLR values are properly generated for soft-decision decoding while maintaining computational efficiency through selective processing.
Solution Approach 2:
The detection process maintains continuous useful action by seamlessly transitioning from forward detection to backward detection, with each phase contributing to LLR value generation. The continuous process ensures that LLR values are generated at every stage, preventing information loss while maintaining computational efficiency through optimized processing continuity.
3Measurement precision
If forward and backward detection are performed in 2×2 MIMO system, then LLR values are accurately generated, but processing time increases
Solution Approach 1:
The patent implements periodic action by structuring detection into distinct forward and backward phases that systematically process symbols in different orders. This periodic structure allows efficient utilization of computational resources in each phase, reducing overall processing time while ensuring accurate LLR value generation through the complementary nature of the two phases.
Data Source
AI summary
A method for detecting soft-decisions in a 2×2 MIMO system includes detecting all candidate symbol vector sets S in which there exist all values of a real part and an imaginary part, and calculating a log-likelihood ratio (LLR) with respect to the candidate symbol vector sets S fromLLR(bki)=mins∈S⋂βi,k-ED(s)2σn2-mins∈S⋂βi,k+ED(s)2σn2.