MIMO Receiver SNR-Adaptive Detector Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In MIMO communication systems, the evaluation of all possible combinations of symbols transmitted from multiple antennas becomes infeasible due to higher order modulation and a large number of antennas, leading to inefficiencies in data transfer and increased bit error rates.
Innovation Solution
A system employing a K-best detector and a depth-first ML detector, where the K-best detector is enabled for low signal-to-noise ratios (SNR) and the depth-first ML detector with pruning is enabled for high SNR, allowing for adaptive symbol detection by estimating the SNR and selecting appropriate detection methods to reduce bit error rates and computational effort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all possible symbol combinations are evaluated for maximum-likelihood detection, then detection accuracy is improved, but computational complexity becomes infeasible for higher order modulation and large number of antennas
Solution Approach 1:
The patent implements dynamic detector selection that adapts to channel conditions. The system switches between K-best detector and depth-first ML detector based on estimated SNR levels, making the detection complexity dynamic rather than static. This resolves the contradiction by allowing full ML detection only when channel conditions warrant it, while using reduced-complexity K-best detection under typical conditions.
Solution Approach 2:
The patent changes the detection parameter (detector type) based on the SNR parameter. By monitoring channel quality and adjusting the detection strategy accordingly, the system optimizes the balance between accuracy and complexity. When SNR is high, full ML detection is used for maximum accuracy; when SNR is low, K-best detection reduces computational load while maintaining acceptable performance.
2Productivity
If K-best detector is used for low SNR conditions, then computational effort is reduced, but bit error rate increases compared to full ML detection
Solution Approach 1:
The system dynamically adjusts detection strategy based on SNR conditions. By switching to depth-first ML detector when SNR exceeds the threshold, the system ensures that reliability is maintained when channel conditions permit, while accepting the K-best detector's higher error rate only when computational resources are constrained by low SNR conditions.
Solution Approach 2:
The detector type parameter is changed based on the SNR parameter. This parameter change allows the system to optimize the trade-off between processing speed and bit error rate by selecting the appropriate detector for current channel conditions, rather than using a fixed detection strategy.
3Reliability
If depth-first ML detector is used for high SNR conditions, then bit error rate is reduced, but the number of potential choices to evaluate increases
Solution Approach 1:
The patent applies partial action by using depth-first ML detector selectively only when SNR exceeds a predetermined threshold, rather than always performing exhaustive ML detection. This partial application of full ML detection achieves the reliability benefits when needed while avoiding the computational burden during typical low-SNR operating conditions.
Solution Approach 2:
The system makes the detection complexity dynamic by adapting to SNR conditions. The depth-first ML detector's exhaustive search is enabled only when channel conditions (high SNR) justify the computational expense, while the K-best detector handles the majority of lower-SNR cases with reduced complexity.
Data Source
Figure 1
Figure 2~3
Figure 4A~4B
AI summary
A system detects symbols communicated from multiple transmitting antennas (102) to multiple receiving antennas (104). An evaluator (108) estimates the SNR of the signals received at the receiving antennas. The evaluator enables a first detector (110) in response to the SNR being lower than the threshold (114), and the evaluator enables a second detector (112) in response to the SNR being higher than the threshold (114). The first detector is preferably a K-best detector, the second detector is preferably an ML depth-first sphere detector. The second detector has a lower bit error rate than the first detector. The computational complexity of the first detector is constant, while the computational complexity of the second detector is lower at high SNR and higher at low SNR.