Polyphase Filter Banks for Rational Oversampling and Low-Load Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing polyphase filters face challenges in efficiently handling non-maximal decimation/interpolation with oversampling factors, leading to increased computational and power consumption, particularly when A>2, which limits their application in systems with limited resources.
Innovation Solution
Adapting polyphase filters to rational oversampling factors of the form Afs/BM, where A and B are natural numbers, using M-path polyphase-based analysis and synthesis filter banks with Sub-path Fusion/Expansion circuits, reducing computational load and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If polyphase filters use non-maximal decimation/interpolation with oversampling factors A>2, then signal reconstruction quality is improved, but computational complexity and power consumption increase
Solution Approach 1:
The filter bank is divided into M parallel polyphase sub-filters, each handling a specific phase of the input signal. This segmentation allows independent processing of multiple signal phases simultaneously, achieving high-resolution reconstruction while distributing computational load across parallel paths, thereby reducing per-path complexity.
Solution Approach 2:
The patent transitions from sequential single-path filtering to parallel multi-phase processing by introducing the phase dimension. The input signal is decomposed into M phases, processed through parallel polyphase filters, and recombined, effectively adding a temporal phase dimension to the processing architecture that improves reconstruction quality without proportionally increasing total computational complexity.
2Measurement precision
If polyphase filters use non-maximal decimation/interpolation with oversampling factors A>2, then signal reconstruction quality is improved, but power consumption increases
Solution Approach 1:
The computational workload is segmented across M parallel polyphase filter paths, allowing power-efficient parallel processing. Each sub-filter operates at reduced complexity compared to a single full-resolution filter, and modern hardware can execute multiple low-power parallel operations more efficiently than one high-power sequential operation, thereby achieving high reconstruction quality with lower overall power consumption.
Solution Approach 2:
The patent changes the processing parameters by introducing rational oversampling factors (A/B) and polyphase decomposition order M, transforming the filter architecture from traditional single-path to multi-phase parallel structure. This parameter change enables the system to achieve superior reconstruction quality while optimizing power consumption through efficient utilization of hardware resources and reduced per-operation complexity.
3Adaptability or versatility
If polyphase filters handle rational oversampling factors Afs/BM, then adaptability to variable center frequencies and bandwidths is improved, but device complexity increases
Solution Approach 1:
The patent implements a dynamic filter bank architecture where the polyphase decomposition order M and oversampling factors A/B can be adjusted to adapt to different center frequencies and bandwidths. The system dynamically reconfigures the number of parallel phases and decimation ratios based on the specific signal characteristics, enabling flexible adaptation to variable communication standards and frequency allocations without requiring completely different filter designs.
Solution Approach 2:
The polyphase filter bank structure serves multiple functions: it performs decimation, filtering, and frequency shifting operations simultaneously across M parallel paths. The same basic polyphase architecture can handle different oversampling factors (A/B), support variable center frequencies, and accommodate different bandwidth requirements, making it a universal solution for diverse communication applications without requiring separate dedicated hardware for each function.
Data Source
AI summary
An Analysis Polyphase Filter (APPF) for shifting a selected passband of an input signal to a passband signal, having a Q=M·B/A decimation factor, B and A being co-primes, the APPF comprising M·B Paths. Each path includes (i) M·B/A multiplication Sub-paths, and (ii) a Sub-path Fusion Circuit, which is configured to generate fused multiplication products responsively to a sum of the multiplication n products generated by each of the multiplication sub-paths.