Density-Modulated Feedback Control for Higher SNDR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Feedback control systems face limitations in achieving high performance due to the inherent granularity and resolution of plants and sensors, leading to restricted signal-to-noise ratio (SNR) and signal-to-noise-plus-distortion ratio (SNDR) performance, especially in applications like audio driving where overshoots and undershoots introduce unwanted errors.
Innovation Solution
A feedback control system employing an up-sampling circuit, delta circuit, and sigma circuit to produce a density modulation signal, which improves SNDR performance through noise shaping and energy recycling in the driving circuit, enhancing the system's ability to drive loads effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional feedback control systems use standard ADC resolution and plant processing step size, then the system is economically achievable with reasonable device size and power consumption, but the SNR and SNDR performance is limited by the inherent granularity and resolution
Solution Approach 1:
The patent segments the control signal into multiple quantization levels over time using density modulation. Instead of using a high-resolution ADC, the system uses a low-resolution ADC that samples at a higher rate, with each sample representing a coarser quantization level. Multiple such samples are combined to achieve the equivalent precision of a high-resolution ADC, thereby improving SNDR performance without requiring a complex high-resolution converter.
Solution Approach 2:
The system employs periodic high-frequency sampling at the up-sampling frequency, where the ADC operates at a rate much higher than the Nyquist rate. This periodic oversampling allows the quantization noise to be spread across a wider frequency range, enabling noise shaping techniques to push quantization noise out of the band of interest, thus improving effective SNDR performance.
2Measurement precision
If the ADC resolution is increased to improve SNDR performance, then measurement precision improves, but power consumption and device size increase
Solution Approach 1:
The patent divides the measurement task into multiple low-resolution measurements taken over time, rather than requiring a single high-resolution measurement. By using a low-resolution ADC that operates at high speed and combining multiple samples through density modulation and noise shaping, the system achieves high effective precision while keeping the power consumption of individual ADC operations low.
Solution Approach 2:
The system changes the sampling frequency parameter significantly, operating the ADC at an up-sampling frequency much higher than the Nyquist rate. This parameter change allows the use of lower-resolution quantization at each sample while achieving high effective precision through the combination of oversampling and noise shaping, thereby reducing the power consumption associated with high-resolution conversion.
3Measurement precision
If the ADC resolution is increased to improve SNDR performance, then measurement precision improves, but device size increases
Solution Approach 1:
The patent segments the precision requirement into temporal domains rather than spatial domains. Instead of using a large high-resolution ADC, the system uses a smaller low-resolution ADC that operates continuously at high speed, with precision achieved through the combination of multiple samples over time rather than through a large number of bits in a single sample.
Solution Approach 2:
The system uses periodic high-frequency sampling to achieve high effective precision. The ADC operates continuously at an up-sampling frequency much higher than the signal bandwidth, allowing the use of a compact low-resolution converter whose small size is compensated by its high operating rate and the subsequent processing of multiple samples.
4Measurement precision
If density modulation is implemented with up-sampling and delta-sigma circuits, then SNDR performance improves by 15-20 dB, but device complexity increases
Solution Approach 1:
The patent merges the functions of upsampling, delta modulation, sigma modulation, and noise shaping into an integrated density modulation architecture. The up-sampling circuit, delta circuit, and sigma circuit work together as a unified system where the delta-sigma modulation inherently performs both upsampling and noise shaping, reducing the need for separate complex components and optimizing the overall circuit implementation.
Solution Approach 2:
The system employs feedback through the delta-sigma modulation architecture, where the quantization error is fed back and filtered by the sigma (integrator) circuit. This feedback mechanism shapes the quantization noise spectrum, pushing noise out of the band of interest, and achieves high SNDR performance through the closed-loop control inherent in delta-sigma modulators.
Data Source
AI summary
A feedback control system configured to drive a load is disclosed. The feedback control system includes an up-sampling circuit, configured to perform an un-sampling operation on a source signal and produce an up-sampled signal with an up-sampling frequency according to the up-sampled signal and a feedback signal from the load; a delta circuit, coupled to the up-sampling circuit and configured to produce a delta signal; a sigma circuit, configured to produce a density modulation signal according to the delta signal; and a driving device, configured to drive the load according to the density modulation signal with the up-sampling frequency.


