Adaptive Filter Branch Updates for Faster Massive MIMO Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Massive Multiple-Input-Multiple-Output (MIMO) systems with numerous antennae face challenges in computation load, resource utilization, and convergence speed due to the complexity of digital pre-distortion (DPD) algorithms in RF circuit compensation.
Innovation Solution
A method is introduced where gradient-based information is shared among all branches of the adaptive filter system, allowing uninterrupted parameter updates, with the similarity between branches influencing the update ratios, thereby accelerating convergence speed and reducing computation load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If gradient-based information is computed independently for each branch, then computational independence is maintained, but convergence speed deteriorates
Solution Approach 1:
The patent merges gradient-based information across multiple branches by computing gradient information for a selected branch and transferring it to update parameters of other branches. This combining approach allows all branches to benefit from the same gradient information simultaneously, achieving fast convergence without requiring complex independent computations for each branch.
Solution Approach 2:
The patent makes the gradient-based information universal across all branches by using it to update parameters for multiple branches simultaneously. The same gradient information serves multiple purposes and multiple branches, eliminating the need for separate gradient computations and achieving both fast convergence and computational efficiency.
2Productivity
If the number of parallel branches is increased, then processing capability is improved, but total computation load increases
Solution Approach 1:
The patent combines the parameter update process across multiple branches by using shared gradient-based information. Instead of computing gradients separately for each branch (which would multiply computation load), the system computes gradient information once and uses it to update all branches, maintaining high processing capability while reducing total computation load.
Solution Approach 2:
The patent uses copying by transferring the computed gradient-based information from one branch to multiple other branches. Rather than independently computing gradients for each branch, the system copies the gradient information and applies it across multiple branches, achieving scalable processing capability without proportional increases in computation load.
Data Source
AI summary
A method in an adaptive filter system is provided. The method comprises obtaining parameters for a plurality of branches of the adaptive filter system. The method further comprises computing gradient-based information for a selected one of the plurality of branches. The method further comprises updating the parameters for the plurality of branches based on the gradient-based information for the selected branch. An adaptive filter system is also provided.