Variable Group Delay Audio Output for Low-Latency Noise Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio output systems in personal devices face challenges in achieving low latency and minimizing out-of-band noise while maintaining a high dynamic range, as conventional filtering methods introduce undesirable delay and power dissipation.
Innovation Solution
A system with a filter configured to receive digital audio signals quantized at specific levels and sampled at high frequencies, featuring a selectable variable group delay, a digital-to-analog converter, and a driver to generate filtered analog audio signals with minimal latency and effective out-of-band noise reduction, utilizing a low-pass filter with a bandwidth of approximately 100 hertz to 10 kilohertz and a group delay of less than 50 microseconds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional out-of-band filtering is applied to reduce ultrasonic noise, then out-of-band noise is reduced, but audio latency increases due to filter delay
Solution Approach 1:
The filter group delay is made variable rather than fixed, allowing the system to dynamically adjust the delay characteristic based on operating conditions. This enables the filter to provide strong out-of-band rejection when needed while minimizing latency when low-latency performance is prioritized, resolving the contradiction between noise reduction and latency.
Solution Approach 2:
The patent changes the group delay parameter of the filter from a fixed value to a variable parameter that can be adjusted between different states. By controlling the group delay parameter, the system can optimize the balance between out-of-band noise filtering effectiveness and audio latency performance according to specific application requirements.
2Reliability
If high sampling rate (at least 500 kHz) is used to reduce quantization noise, then dynamic range is improved, but power consumption increases
Solution Approach 1:
The system uses oversampling at high rates (at least 500 kHz) but applies filtering to remove only the necessary out-of-band components. This partial action approach maintains the benefits of high sampling rate for dynamic range while avoiding the need to process and power all ultrasonic frequencies, thereby reducing overall power consumption compared to continuously operating at full high-rate processing.
Solution Approach 2:
The filter extracts and removes only the harmful out-of-band noise components from the high sampling rate signal, while preserving the beneficial audible band signal. This selective extraction allows the system to maintain high dynamic range benefits from oversampling without the full power penalty of processing all high-frequency content.
3Loss of energy
If aggressive out-of-band filtering is applied to eliminate ultrasonic noise, then power dissipation in drivers and amplifiers is reduced, but audio path delay increases
Solution Approach 1:
The filter's group delay is made dynamically adjustable, allowing the system to optimize the trade-off between power dissipation reduction and audio path delay. When aggressive filtering is needed to reduce power dissipation, the variable group delay compensates to maintain acceptable latency, and vice versa.
Solution Approach 2:
The patent adjusts the filter parameters including group delay to optimize the balance between power dissipation and delay. By changing these parameters dynamically, the system can achieve aggressive out-of-band filtering when power savings are prioritized while maintaining lower delay when audio responsiveness is more critical.
Data Source
AI summary
A system may include an input configured to receive a digital audio input signal having at least four and fewer than 65,000 quantization levels and sampled at at least 500 kilohertz, a low-pass filter configured to receive the digital audio input signal and perform filtering on the digital audio input signal to generate a filtered digital audio input signal having a bandwidth of between approximately 100 hertz and 10 kilohertz, a digital-to-analog converter configured to receive the filtered digital audio input signal and convert the filtered digital audio input signal into an equivalent analog audio input signal, and a driver configured to receive the equivalent analog audio input signal and drive an analog audio output signal to a transducer, wherein a group delay from the input to an output of the driver is less than 50 microseconds.