Reconfigurable Filter Array for Flexible Sampling Rate Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital filter banks are inflexible and cannot be easily configured or expanded, requiring re-development for updated algorithm performance, which prolongs the development period and increases design costs due to fixed filter orders and independent module operations.
Innovation Solution
A filter unit with a control module for analyzing configuration information and controlling data flow, a filtering module for sampling rate conversion, and an output selection module, along with a filter array formed by cascading multiple filter units, allowing for flexible configuration and expansion of filter orders and data paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If filter orders and module operations are fixed in existing digital filter banks, then the system structure is simple and stable, but the system cannot be flexibly configured or expanded when algorithm performance needs to be updated
Solution Approach 1:
The filter bank is divided into multiple independent filter units, each capable of independent configuration and operation. Each filter unit contains its own delay processing unit, frequency shift processing unit, and filter processing unit, allowing individual modules to be configured without affecting others, thereby enabling flexible adaptation while maintaining modular simplicity.
Solution Approach 2:
The filter units are designed with universal functionality to perform multiple operations including delay processing, frequency shift processing, and various types of filter processing. The control module can dynamically configure these universal units to perform different functions based on algorithm requirements, eliminating the need for separate dedicated modules for each function.
2Productivity
If filter coefficient configuration is realized under fixed orders in existing systems, then the design and development process is straightforward, but the development period is prolonged and design cost increases when algorithm performance updates are required
Solution Approach 1:
The filter bank system implements dynamic reconfigurability where filter orders, sampling rate conversion ratios, and data path configurations can be changed in real-time through control module configuration without requiring system re-development. This dynamic capability allows rapid adaptation to algorithm performance updates, significantly reducing development time and costs.
Solution Approach 2:
The system allows dynamic changing of key parameters including filter orders, sampling rate conversion ratios, and data path selections through configuration information processed by the control module. These parameter changes can be made without hardware modification or re-development, enabling rapid algorithm performance updates.
3Adaptability or versatility
If each module in the filter bank operates independently with fixed functions, then the module design is simple and reliable, but calculation sharing among modules cannot be realized
Solution Approach 1:
Multiple filter units are connected in cascade to form an integrated filter bank system where calculation results from one unit can be shared and utilized by subsequent units. The control module coordinates data flow and calculation sharing among units, enabling efficient resource utilization while maintaining operational independence through standardized interfaces and control mechanisms.
Data Source
AI summary
Provided are a filter unit and a filter array, and the filter unit includes: a control module (10) configured to analyze configuration information of the filter unit, and control selection of internal functions and flow direction of data in the filter unit; a filtering module (20) connected to the control module (10), and configured to perform an operation on internal data of the filter unit and perform a sampling rate conversion function under control of the control module (10); and an output selection module (30) connected to the control module (10) and the filtering module (20), and configured to perform a selection function on output data of the filtering module (20) under control of the control module.