Noise Control System Frequency Domain Segmentation Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Active noise reduction systems face challenges in convergence speed due to the need for extensive processing and repeated calculations, which can prolong noise reduction effectiveness.
Innovation Solution
A noise control system comprising a feedforward module, feedback module, error pre-processing module, and signal integrating module, where the error pre-processing module generates different pre-processing signals based on frequency regions to improve calculation efficiency and convergence speed by providing these signals to the feedforward and feedback modules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If active noise reduction is used to offset sounds by generating offset sounds of similar amplitude but opposite phase, then noise reduction effect is improved, but processing speed requirement increases and convergence time becomes too long
Solution Approach 1:
The error signal is segmented into multiple frequency bands using frequency domain decomposition. Each frequency band is processed independently through separate adaptive filters, allowing parallel computation and reducing the convergence time for each individual filter while maintaining overall noise reduction effectiveness.
Solution Approach 2:
The system transforms the time-domain error signal into the frequency domain using Fast Fourier Transform (FFT). This dimensional transformation allows simultaneous processing of multiple frequency components through parallel adaptive filters, significantly reducing convergence time while maintaining noise reduction performance.
2Reliability
If repeated calculation is performed to achieve convergence for best noise reduction effect, then noise reduction quality is improved, but hardware processing speed requirement increases
Solution Approach 1:
The adaptive filter is segmented into multiple parallel filters, each handling a specific frequency band. This segmentation allows independent optimization of each filter's convergence without requiring the entire system to wait for a single slow-converging filter, thereby improving overall processing speed while maintaining noise reduction quality.
Solution Approach 2:
The system performs preliminary frequency domain transformation and signal segmentation before the main noise reduction computation. This preliminary processing organizes the data in a way that enables more efficient parallel computation, reducing the total processing time required for convergence while maintaining high noise reduction quality.
Data Source
AI summary
A noise control system comprises a feedforward module, a feedback module, an error pre-processing module, and a signal integrating module. The feedforward module is configured to receive a reference signal and output a feedforward anti-noise signal by performing feedforward processing to the reference signal. The feedback module is configured to receive an error signal and output a feedback anti-noise signal by performing feedback processing to the error signal. The error pre-processing module is configured to receive the error signal and output a first pre-processing signal to the feedforward module and a second pre-processing signal to the feedback module. The signal integrating module is configured to output an integrated anti-noise signal integrated from the feedforward anti-noise signal and the feedback anti-noise signal. Wherein the first pre-processing signal corresponds to the first part of the error signal which belongs to the first frequency region; the second pre-processing signal corresponds to the second part of the error signal which belongs to the second frequency region.