Oversampled Low-Delay Filter Bank for Alias-Resistant Subbands
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital filter banks face challenges in reducing aliasing artifacts when subband samples are altered, leading to impairments in band separation and intra-channel pass band ripple, particularly in critically sampled designs.
Innovation Solution
The method involves designing an oversampled filter bank with an asymmetric prototype filter, using a composite objective function that minimizes transfer function errors, aliasing errors, and ripple within subbands, while optimizing coefficients through iterative processes and modulation techniques like cosine and complex-exponential modulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a critically sampled filter bank is used, then the total number of subband samples per unit time equals that of the input signal, but aliasing artifacts occur when subband samples are altered
Solution Approach 1:
The patent applies oversampling (excessive action) by using more subband samples per unit time than the input signal requires. This excess sampling provides redundancy that allows subband processing to occur without generating aliasing artifacts, while maintaining efficient reconstruction of the original signal.
2Object-generated harmful factors
If an oversampled filter bank is used, then aliasing artifacts are reduced, but band separation between subbands deteriorates
Solution Approach 1:
The patent changes the parameter of filter design by using asymmetric prototype filters instead of traditional symmetric ones. This parameter change allows the filter bank to achieve both oversampling (reducing aliasing) and precise band separation simultaneously, resolving the contradiction between the two requirements.
3Reliability
If the number of subband samples is increased, then aliasing robustness improves, but intra-channel pass band ripple increases
Solution Approach 1:
The patent applies local quality by designing filters with different characteristics for different frequency regions. The asymmetric prototype filter provides optimized local properties in the passband to minimize ripple while maintaining the overall oversampled structure for aliasing robustness.
Data Source
AI summary
The present document describes a method (200) for determining N coefficients of an asymmetric prototype filter p0 for use in a low delay M-channel analysis and/or synthesis filter bank (101, 102) comprising M analysis filters hk (103) and/or M synthesis filters fk(106), k=0, . . . , M−1, wherein M is greater than 1, and wherein subband signals which are processed by the analysis and/or synthesis filter bank (101, 102) are decimated by a decimation factor S, with S<M. The method (200) comprises determining (201) a target transfer function of the analysis and/or synthesis filler bank (101, 102) comprising a target delay D; wherein D is smaller or equal to N. Furthermore. the method (200) comprises determining (202) a composite objective function etot comprising a transfer function error term e and an aliasing error term ea; wherein the transfer function error term et is associated with a deviation between a transfer function of the analysis and/or synthesis filter bank (101, 102) and the target transfer function; and wherein the aliasing error term ea is associated with errors incurred