Multi-band Signal Processor Dynamic Update Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-band dynamic range compressors for hearing aids suffer from undersampling and oversampling issues due to a fixed update rate for all frequency bands, leading to aliasing distortion and inefficient use of computational resources.
Innovation Solution
A multi-band signal processor that allows separate and flexible update rates for different frequency bands, enabling optimal sampling rates based on perceptual performance criteria, thereby avoiding aliasing and reducing computational waste.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a fixed block rate is used for FFT-based frequency analysis in all frequency bands, then the implementation is simple and computationally efficient, but high frequency components are undersampled causing aliasing distortion
Solution Approach 1:
The frequency spectrum is divided into multiple bands, and each band is updated at a different rate appropriate to its frequency characteristics. High frequency bands are updated more frequently than low frequency bands, preventing aliasing while maintaining computational efficiency.
Solution Approach 2:
The update rate for frequency bands is made dynamic rather than fixed. The system adapts the block rate for each frequency band based on its specific requirements, allowing high frequency bands to be sampled more frequently while low frequency bands use lower update rates.
2Reliability
If a high block rate is used to accommodate high frequency components, then aliasing is avoided, but low frequency bands are oversampled wasting computational resources
Solution Approach 1:
The frequency spectrum is segmented into multiple bands with different update rate requirements. Low frequency bands are updated less frequently since they require lower sampling rates, reducing unnecessary computational work and power consumption.
Solution Approach 2:
Different quality levels (update rates) are applied to different frequency bands based on their specific needs. High frequency bands receive higher update rates for accurate representation, while low frequency bands use lower update rates, optimizing the overall system efficiency.
3Measurement precision
If a high block rate is used to avoid undersampling, then high frequency components are accurately represented, but computational resources are wasted on oversampling low frequency bands
Solution Approach 1:
The computational load is segmented across different frequency bands with tailored update rates. Each band is processed at the minimum necessary rate for its frequency characteristics, reducing the overall computational burden while maintaining accuracy where needed.
Solution Approach 2:
The block rate parameter is changed dynamically for different frequency bands. Instead of using a single fixed block rate for all bands, the system applies different block rates to different bands, optimizing the balance between accuracy and computational complexity.
Data Source
AI summary
A method includes: processing the digital audio input signal to generate M delayed digital audio signal samples; converting the delayed digital audio signal samples to frequency domain representation in N number of frequency bands to compute respective signal spectrum values; determining respective signal level estimates; computing respective frequency domain gain coefficients based on the respective signal level estimates and band gain laws; transforming the frequency domain gain coefficients to time domain representation to produce M time-varying filter coefficients of a processing filter; convolving the M delayed digital audio signal samples with the time-varying filter coefficients to produce the processed digital output signal; and updating the delayed digital audio signal samples in accordance with a sample-by-sample or a predetermined block rate; wherein two of the signal spectrum values for at least two of the frequency bands are updated at different rates; and wherein M and N are positive integer numbers.