Adaptive Filter Offset Injection for Faster Shared-Resource Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Adaptive filters face challenges in achieving fast convergence of tap-weight coefficients, particularly when using resource sharing, which leads to reduced power efficiency and increased costs in cost-sensitive markets.
Innovation Solution
The implementation of a bootstrapping technique that injects offsets into the iterative convergence algorithm for adjusting filter coefficients, combined with computational resource sharing, to enhance the convergence rate of adaptive filters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If computational resource sharing is used in adaptive filter, then power efficiency and cost are improved, but convergence rate deteriorates
Solution Approach 1:
The patent applies preliminary action by injecting an offset into the filter coefficient before the adaptive convergence algorithm begins. This offset is determined based on the sharing factor and serves as a head start for the convergence process, allowing the system to achieve faster convergence despite using reduced computational resources. The offset preparation is done in advance, enabling the resource-sharing adaptive filter to catch up in convergence speed without requiring additional computational blocks.
Solution Approach 2:
The patent changes the parameter of the filter coefficient by adding an offset that is a function of the sharing factor. This parameter modification allows the system to compensate for the reduced convergence rate caused by resource sharing. By adjusting the filter coefficient with this predetermined offset, the system maintains faster convergence performance while still benefiting from the power efficiency and cost savings of computational resource sharing.
2Device complexity
If computational resource sharing is used in adaptive filter, then device complexity is reduced, but convergence rate deteriorates
Solution Approach 1:
The patent uses preliminary action by calculating and injecting an offset based on the sharing factor before the adaptive convergence algorithm executes. This offset injection compensates for the reduced computational resources, enabling the simplified device to achieve faster convergence without requiring additional computational blocks or increased device complexity.
Solution Approach 2:
The patent modifies the filter coefficient parameter by adding an offset that depends on the sharing factor. This parameter change allows the reduced-complexity device to compensate for its limited computational resources and achieve convergence rates comparable to or faster than systems with more computational blocks, thereby resolving the contradiction between device complexity and convergence rate.
3Use of energy by moving object
If fewer computational blocks are used, then power consumption is reduced, but convergence rate deteriorates
Solution Approach 1:
The patent applies preliminary action by determining and injecting an offset into the filter coefficient before the adaptive convergence algorithm begins execution. This offset, calculated based on the sharing factor, provides a head start that compensates for the reduced number of computational blocks, enabling faster convergence with lower power consumption.
Solution Approach 2:
The patent changes the filter coefficient parameter by adding an offset that is a function of the sharing factor. This parameter modification allows the system to achieve faster convergence despite using fewer computational blocks and consuming less power, thereby resolving the contradiction between power consumption and convergence rate.
Data Source
Figure 1~3
Figure 4~5
Figure 6
AI summary
The present application relates to an adaptive filter using resource sharing and a method of operating the adaptive filter. The filter comprises at least one computational block, a monitoring block and an offset calculation block. The computational block is configured for adjusting a filter coefficient, ci(n), in an iterative procedure according to an adaptive convergence algorithm. The monitoring block is configured for monitoring the development of the determined filter coefficient, ci(n), during the performing of the iterative procedure. The offset calculation block is configured for determining an offset, Offi, based on a monitored change of the filter coefficient, ci(n), each first time period, T1, and for outputting the determined offset, Offi, to the computational block if the determined filter coefficient, ci(n), has not reached the steady state. The computational block is configured to accept the determined offset, Offi, and to inject the determined offset, Offi, into the iterative procedure.