Subband Audio Processing with Frequency Shifting to Reduce Aliasing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current audio signal processing techniques face challenges with subband aliasing and increased computational resource usage, especially in high-bandwidth applications, which can lead to degraded performance and increased complexity.

Innovation Solution

The approach involves subband processing with reduced aliasing components by shifting subbands to specific frequency intervals and using power complementary filters to minimize aliasing, allowing for efficient processing and adaptation to different frequency characteristics without gaps in the frequency spectrum.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If subband processing is applied to reduce computational resource usage, then computational efficiency is improved, but subband aliasing components are introduced which degrade processing performance

Engineering Contradiction:
Improvecomputational efficiencyVSAvoidprocessing performance
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent converts the harmful aliasing components into beneficial elements by using them as input to adaptive filters that selectively remove aliasing while preserving desired signal components. The aliasing that would normally degrade performance is instead utilized to drive adaptation processes that improve overall system performance.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The patent introduces intermediate processing stages including adaptive filters and frequency shifters that act as mediators between the subband decomposition and the final reconstruction. These intermediaries process the subband signals to reduce aliasing components before reconstruction, thereby protecting the final output from aliasing degradation.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If the bandwidth of audio signals is increased to improve audio quality, then audio quality is improved, but computational resource usage is substantially increased

Engineering Contradiction:
Improveaudio qualityVSAvoidcomputational resource usage
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent divides the wideband audio signal into multiple narrower subbands, each processed independently at lower computational rates. This segmentation allows the system to handle wideband signals by breaking them into manageable pieces that require fewer computational resources while maintaining overall audio quality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different processing characteristics to different frequency subbands, allowing optimization for each band's specific requirements. This local adaptation enables efficient resource usage by tailoring processing intensity to the actual needs of each frequency region rather than applying uniform high-cost processing across the entire bandwidth.

Inventive Principle:
Principle #3Local quality

3Measurement precision

If subband processing is applied to adapt to different frequency characteristics, then processing accuracy is improved, but device complexity is increased

Engineering Contradiction:
Improveprocessing accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the frequency spectrum into multiple subbands that can be processed with band-specific characteristics. This segmentation enables accurate adaptation to different frequency properties while keeping each individual processing block relatively simple, distributing complexity across multiple manageable units rather than requiring one complex full-band processor.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3353785B2Audio signal processing
Publication Date: 2021.09.22 KONINKLIJKE PHILIPS NV
  • EP3353785B2 patent drawingFigure 1
  • EP3353785B2 patent drawingFigure 2
  • EP3353785B2 patent drawingFigure 3

AI summary

An audio signal processing apparatus comprises a receiver (403) receiving an audio signal sampled at a first sampling frequency, the audio signal having a maximum frequency below half the first sampling frequency by a first frequency margin. A filter bank (405) generates subband signals for the digital audio signal using overlapping sub-filters. A first frequency shifter (407) applies a frequency shift to at least one subband of the set of subbands and a decimator (409) decimates the subband signals by a decimation factor resulting in a decimated sampling frequency being at least twice a bandwidth of each of the overlapping sub-filters. The frequency shift for a subband is arranged to shift the subband to a frequency interval being a multiple of a frequency interval from zero to half the decimated sample frequency. The subband may be individually processed and the processed subbands may subsequently be combined to generate a full band output signal.