MIMO Receiver Candidate Generation via Symmetry-Based LUT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
MIMO detection algorithms face complexity and inefficiency in generating candidate symbols due to the large size of the alphabet, leading to high computational burden in finding nearest constellation symbols, especially with the addition of transmit antennas.
Innovation Solution
The use of look-up table (LUT) based candidate generation techniques that exploit symmetric properties of symbol constellations for lossless compression, allowing for a secondary constellation to overlay a primary constellation and reduce storage requirements, enabling efficient candidate list generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a brute-force implementation of candidate generation is used to compute all constellation symbols by their distance from the input, then the complete set of candidate symbols is obtained, but the computational complexity and storage requirements increase significantly when the alphabet size is large
Solution Approach 1:
The constellation is divided into multiple decision regions, with only the relevant decision region containing the input point being processed in detail. This segmentation allows the algorithm to avoid computing distances to all constellation symbols, thereby reducing computational complexity while maintaining accurate candidate generation within the relevant region
Solution Approach 2:
Decision regions are pre-computed and stored based on the constellation geometry before the actual candidate generation process. By pre-establishing the spatial partitioning of the constellation space, the algorithm can quickly identify which region contains the input point and only process symbols within that region, avoiding unnecessary computations
2Productivity
If the number of transmit and receive antennas is increased to increase MIMO channel capacity, then data throughput increases linearly, but the complexity of recovering transmitted information increases
Solution Approach 1:
The detection process is segmented into two stages: first identifying the decision region containing the received signal, then generating candidates only within that region. This segmentation makes the detection complexity grow more slowly with the number of antennas compared to brute-force methods that would examine all possible symbol combinations
Solution Approach 2:
Instead of exhaustively searching all possible transmitted symbols, the algorithm performs partial action by limiting the search to only those symbols within the relevant decision region. This partial search is sufficient to find the nearest neighbors needed for detection while avoiding the excessive computational burden of examining the entire symbol space
3Loss of information
If a look-up table stores candidate lists for all constellation symbols, then complete candidate information is available, but storage requirements become excessive for large alphabets
Solution Approach 1:
The look-up table is segmented to store candidate lists only for symbols within a single decision region rather than all constellation symbols. This segmentation reduces storage requirements by a factor equal to the number of decision regions, while still providing complete candidate information for any input point through the combination of region identification and local table lookup
Solution Approach 2:
The look-up table stores only the partial information needed for one decision region, which is sufficient when combined with the region identification step. This partial storage approach avoids the excessive memory requirements of storing complete candidate lists for all constellation symbols while maintaining full functionality
Data Source
AI summary
A method and system for generating a set of candidate symbols. A system includes a Multiple Input Multiple Output (“MIMO”) receiver. The receiver includes a candidate generation look-up table (“LUT”) that provides a list of candidate values for a transmitted symbol selected from a constellation of symbols. The candidate generation LUT stores candidate lists for a portion of the constellation of symbols. The portion of the constellation for which candidate lists are stored is selected according to a symmetry of the constellation. The LUT preferably provides a secondary constellation superimposed on a decision region of a primary constellation. The LUT also preferably includes an inner point of the primary constellation and outer points of the primary constellation. The primary and secondary constellations are preferably compressed by application of quadrant, mirror, and inner-point symmetries.