PDM Filter Structure for Low-Latency Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The Signal to Noise Ratio (SNR) of digitally processed output in systems handling Pulse-Density Modulated (PDM) signals is often dependent on the oversampling ratio, and existing systems face challenges in reducing high-frequency noise when converting digital signals to analog, particularly in audio signal processing and noise cancellation applications.
Innovation Solution
A signal processing structure comprising a first digital filter and a pre-processing digital filter that directly low-pass filters sigma-delta modulated input signals, optionally using a moving average filter and upsampler to enhance the SNR by increasing the sampling rate and reducing noise, specifically designed for audio signal processing and noise cancellation in headphones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a digital filter operates directly on PDM signals without pre-processing, then the device complexity is reduced, but the Signal to Noise Ratio (SNR) deteriorates due to high-frequency noise
Solution Approach 1:
The filtering process is segmented into two distinct stages: a pre-processing digital filter that performs initial noise reduction on the PDM signal, followed by a main digital filter that processes the pre-processed signal. This segmentation allows each filter to be optimized for its specific function, improving overall SNR while maintaining manageable complexity through functional division.
2Reliability
If the oversampling ratio is increased to improve SNR, then the Signal to Noise Ratio improves, but the processing speed and latency worsen
Solution Approach 1:
The system optimizes the oversampling ratio parameter to achieve an optimal balance between SNR improvement and processing speed. Rather than using excessively high oversampling ratios that would slow processing, the pre-processing filter enables effective noise reduction at moderate oversampling ratios, thus maintaining processing speed while improving SNR.
3Reliability
If a pre-processing digital filter is added to reduce high-frequency noise, then the SNR improves, but the device complexity increases
Solution Approach 1:
High-frequency noise is extracted and removed from the PDM signal in the pre-processing stage before the main filtering operation. By separating the noise reduction function into a dedicated pre-processing filter, the main filter can focus on the primary signal processing task, and the overall system achieves better SNR with organized functional separation.
4Reliability
If digital filtering is performed on PDM signals before DAC conversion, then the output characteristics improve, but the hardware requirements increase
Solution Approach 1:
The pre-processing digital filter is designed to perform multiple functions: high-frequency noise reduction, signal conditioning for the main filter, and optimization of the signal for subsequent DAC conversion. This multi-functionality allows a single component to address multiple requirements, improving output characteristics without proportionally increasing hardware complexity.
Data Source
AI summary
A signal processing structure and method are presented. A first digital filter operates on received sigma-delta modulated (SDM) input signals. A second pre-processing digital filter receives a SDM input signal, directly low pass filter the SDM input signal and provides an output SDM signal. The output sigma-delta modulated signal is provided as an input for said first digital filter. In standard digital systems operating with digital microphones, filtering of the microphones' output signal requires to first convert the signal into pulse code modulation (PCM), then filter and finally convert back to pulse density modulation (PDM). This approach increases the latency of the system because decimation and interpolation must be performed in order to pass from PDM to PCM. By using filters that operate directly on the oversampled PDM output of the digital microphones it is possible to reduce the latency of the system and minimize the hardware area.