MIMO Signal Detection Using Dynamic Candidate Symbol Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MIMO systems face high computational complexity in signal detection, making maximum likelihood detection impractical for systems with multiple inputs or high modulation orders, and near ML detection methods fail to match ML performance due to fixed candidate symbol selection and suboptimal thresholds.
Innovation Solution
An optimal maximum likelihood signal detection method that dynamically sets the number of surviving candidate symbols based on channel environment and SNR using QR decomposition, ensuring the smallest metric is used as the optimal threshold for each detection layer, thereby reducing computational complexity and maintaining ML performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If maximum likelihood detection is applied to achieve optimal signal detection performance, then bit error rate performance is improved, but computational complexity increases exponentially
Solution Approach 1:
The detection process is divided into multiple detection layers, where candidate symbols are progressively selected and refined through successive layers. This segmentation transforms the single-step exhaustive search of ML detection into a multi-stage process, reducing computational complexity while maintaining detection performance.
Solution Approach 2:
Candidate symbols are pre-selected based on metrics calculated from received signals and channel information before final detection. By preliminarily identifying promising candidate symbols using QR decomposition and metric computation, the system avoids evaluating all possible symbol combinations, thereby reducing computational complexity.
2Device complexity
If a fixed number of candidate symbols M is selected in each detection layer to reduce complexity, then computational complexity is reduced, but bit error rate performance degrades below ML detection
Solution Approach 1:
The number of candidate symbols M is made dynamic rather than fixed. M varies according to the detection layer index and the specific channel conditions, allowing the system to allocate more computational resources when needed and fewer when sufficient performance is already achieved. This dynamic adaptation enables the system to maintain ML-level performance with reduced average complexity.
Solution Approach 2:
The parameter M (number of candidate symbols) is changed based on detection layer and performance requirements. By adjusting M as a variable parameter rather than a constant, the system can optimize the trade-off between computational complexity and detection performance for each specific detection stage and channel condition.
3Productivity
If an adaptive threshold method is applied to select M candidate symbols variably, then computational efficiency is improved, but optimal threshold cannot be applied to each detection layer resulting in suboptimal performance
Solution Approach 1:
Different thresholds are applied to different detection layers based on their specific requirements and characteristics. Each detection layer receives a locally optimized threshold value rather than a single global threshold, allowing each layer to operate at its optimal performance point. This local quality approach ensures that early layers use stricter thresholds while later layers can use more lenient thresholds.
Solution Approach 2:
The threshold parameter is changed for each detection layer index. By making the threshold a layer-dependent variable rather than a constant, the system can optimize detection performance at each stage of the multi-layer process, achieving both computational efficiency and optimal bit error rate performance.
Data Source
AI summary
An optimal maximum likelihood signal detection method and apparatus for a MIMO system are disclosed. An embodiment of the invention can provide an optimal signal detection method for a multiple-input multiple-output system that includes: determining the smallest metric of the final layer expanded from the candidate symbols of the i-th detection layer as the optimal threshold of the i-th detection layer; and selecting surviving candidate symbols from among the candidate symbols of the i-th detection layer by using the optimal threshold.