MIMO Decoding Using Reduced Bit Representation and Bounded Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MIMO decoding methods, such as standard Maximum Likelihood Decoding (MLD), face high complexity and power consumption due to their exponential growth with the number of antennas and modulation order, making them inefficient in terms of processing time and power usage.
Innovation Solution
The proposed solution involves using a MIMO decoding process that employs successively smaller bounded regions, such as spheres or hyper-cubes, to eliminate candidate solutions, reducing the complexity by representing points with a reduced bit representation and using logical bit operations to determine if candidates fall within these regions, thereby decreasing the number of candidates to evaluate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard Maximum Likelihood Decoding (MLD) is used, then decoding performance is optimal, but computational complexity grows exponentially with the number of antennas and modulation order
Solution Approach 1:
The patent segments the search space by dividing it into multiple regions (e.g., inner region, middle region, outer region) based on distance from the received point. Instead of evaluating all candidate points uniformly, the algorithm processes regions in sequence, eliminating candidates outside the current search region. This segmentation reduces the number of candidates that need to be fully evaluated, thereby reducing computational complexity while maintaining decoding performance.
Solution Approach 2:
The patent applies partial action by evaluating only a subset of candidate points rather than all possible candidates. The algorithm performs partial evaluation by checking distance metrics and eliminating candidates that fall outside bounded regions before completing full evaluation. This partial evaluation approach reduces computational complexity while still identifying the correct transmitted symbols with high probability.
2Reliability
If standard Maximum Likelihood Decoding (MLD) is used, then decoding accuracy is maximized, but power consumption increases due to high computational requirements
Solution Approach 1:
The patent segments the computational workload by dividing the candidate evaluation into multiple stages corresponding to different regions. Candidates are eliminated in successive stages based on distance criteria, so that fewer candidates require full evaluation. This segmentation reduces the total number of computational operations, thereby reducing power consumption while maintaining decoding accuracy.
Solution Approach 2:
The patent performs partial evaluation of candidate points by computing distance metrics and eliminating candidates that cannot be optimal before completing full evaluation. This partial action reduces the number of expensive computational operations required, thereby reducing power consumption while still achieving accurate decoding results.
3Productivity
If sphere decoding is used to reduce complexity, then processing time is reduced, but the algorithm still requires significant computational resources
Solution Approach 1:
The patent segments the search space into multiple bounded regions and processes candidates in sequence through these regions. By dividing the search space and eliminating candidates in successive stages, the algorithm reduces the number of candidates that require full evaluation, thereby reducing computational resources while maintaining processing speed improvements.
Solution Approach 2:
The patent applies partial evaluation by computing distance metrics and eliminating candidates that fall outside bounded regions before completing full candidate evaluation. This partial action reduces the computational resources required compared to evaluating all candidates, while still achieving fast processing speeds.
Data Source
AI summary
A method, a system, a wireless device and a computer program product are provided. A process for decoding transmitted symbols in MIMO systems is provided. The received point coordinates and estimated candidate coordinates are represented by a reduced bit representation naturally dividing each dimension into uniformly sized bit segments. A bounded region surrounding the received point is defined in the total number of dimensions, the bounded region being a hypercube. For each candidate, an elimination process is carried out including: determining a position of the candidate to be inside or outside the bounded region, eliminating the candidate if it is outside, storing it if it is inside, adding a further bit thereby reducing a size of the bounding region; repeating the elimination process for the candidates stored as potential solutions until a single solution is obtained; determining the transmitted symbols from the single solution found.