MIMO Demapper Using QR Decomposition and L1 Norm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multiple-input, multiple-output (MIMO) communication systems, decoding transmitted symbols from received signals corrupted by channels and noise is complex due to the high number of possible symbol combinations in constellations, requiring significant matrix calculations that are computationally intensive, especially for large constellations like 64-QAM.
Innovation Solution
A demapper system is developed that includes a QR decomposer, normalizer, candidate hypothesis generator, and error calculators to minimize multiplications and operations, using L1 norm calculations and table lookups to efficiently determine hypothesis and counter-hypotheses for received symbols, thereby reducing computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If maximum likelihood sequence estimation is used to decode MIMO symbols, then decoding accuracy is improved, but computational complexity increases significantly
Solution Approach 1:
The patent segments the complex MIMO decoding problem into multiple independent stages: QR decomposition of the channel matrix, hypothesis generation for each transmitted symbol, counter-hypothesis determination, and L1 norm calculation. This segmentation allows each stage to be processed separately and efficiently, reducing overall computational complexity while maintaining decoding accuracy
Solution Approach 2:
The patent performs preliminary QR decomposition of the channel matrix before symbol decoding. This preliminary action transforms the complex MIMO detection problem into a simpler form where hypothesis generation and evaluation can be performed more efficiently, reducing the computational burden during actual decoding operations
2Productivity
If large constellations like 64-QAM are used to increase bit rate, then transmission capacity is improved, but the number of matrix calculations increases exponentially
Solution Approach 1:
The patent divides the constellation decoding into independent hypothesis generations for each symbol position. For 64-QAM, instead of evaluating all possible symbol combinations simultaneously, the system generates hypotheses for one symbol at a time based on QR decomposition results, then determines counter-hypotheses separately. This segmentation reduces the number of matrix calculations from exponential to linear complexity
Solution Approach 2:
The patent changes the norm parameter from L2 (Euclidean distance) to L1 (Manhattan distance) for hypothesis evaluation. This parameter change simplifies the calculations significantly, as L1 norm requires only absolute value operations rather than square root and multiplication operations, making large constellation decoding computationally feasible
3Ease of operation
If conventional demapping methods are used for MIMO systems, then demapping functionality is achieved, but processing time and computational operations are excessive
Solution Approach 1:
The patent performs preliminary QR decomposition and normalization of the channel matrix before demapping operations. This preliminary action prepares the system for faster hypothesis evaluation by pre-computing transformation matrices and normalization factors, significantly reducing processing time during actual demapping operations
Solution Approach 2:
The patent generates a limited set of candidate hypotheses based on QR decomposition results rather than evaluating all possible constellation points. By focusing computational resources on the most likely hypotheses and their counter-hypotheses, the system achieves adequate demapping performance with significantly reduced processing time
Data Source
AI summary
A method, implementable on a multiple-input, multiple-output symbol receiver, includes selecting a hypothesis for a second symbol value U2 from among the set of fixed constellation points, calculating a hypothesis for a first symbol value U1 from the resultant selected U2 value, and generating a first half of counter-hypotheses from interim results of calculating the hypotheses values.