Segmented DAC Noise Shaping for MSB Quantization Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Segmented digital-to-analog converters (DACs) used in high-fidelity audio systems suffer from noise issues due to quantization of most significant bits, and existing solutions fail to effectively suppress this noise, limiting dynamic range and signal quality.
Innovation Solution
The design incorporates a multi-segment approach with delta-sigma modulators for both most significant and least significant bits, where quantization noise from the most significant bits is extracted, filtered, and injected into the least significant bit path to suppress noise, and a DAC controller uses fixed transition weighting average to reduce inter-symbol interference and resistance mismatches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If segmented DAC architecture is used to convert digital signal to analog signal, then conversion functionality is achieved, but quantization noise from most significant bits degrades signal quality
Solution Approach 1:
The patent extracts quantization noise from the MSB path and reinjects it into the LSB path, converting the harmful noise into a beneficial cancellation mechanism. The noise shaping filter processes the extracted noise to enhance cancellation effectiveness, transforming the originally harmful quantization noise into a tool for improving signal quality and reducing distortion.
Solution Approach 2:
The DAC is divided into separate MSB and LSB paths, each with independent delta-sigma modulators and DACs. This segmentation allows independent processing of different bit significance levels, enabling selective noise extraction from the MSB path and reinjection into the LSB path, thereby addressing quantization noise issues without compromising overall conversion functionality.
2Measurement precision
If noise shaping filter is added to process extracted quantization noise, then noise cancellation effectiveness is improved, but circuit complexity increases
Solution Approach 1:
The noise shaping filter acts as an intermediary component between the MSB and LSB paths. It processes the extracted quantization noise to optimize its characteristics for effective cancellation, serving as a mediator that enhances the noise cancellation mechanism while maintaining a clear functional separation between the MSB and LSB processing paths.
3Reliability
If fixed transition weighting average control method is used, then inter-symbol interference and resistance mismatches are reduced, but control logic complexity increases
Solution Approach 1:
The control method pre-calculates optimal switching transitions and applies weighting averages to anticipate and compensate for potential inter-symbol interference and resistance mismatch effects before they occur. This preliminary action approach reduces the need for complex real-time corrections while improving signal accuracy.
4Measurement precision
If multiple delta-sigma modulators and DACs are used in multi-segment architecture, then dynamic range is extended, but device complexity increases
Solution Approach 1:
The patent merges the MSB and LSB paths through a unified noise cancellation mechanism where extracted noise from the MSB path is reinjected into the LSB path. This merging approach allows the multiple modulators and DACs to work cooperatively, extending dynamic range while using the interaction between paths to reduce overall system complexity compared to fully independent designs.
Data Source
AI summary
Circuits and methods for converting digital input signals into the analog domain are described. Such circuits may perform the conversion in a segmented fashion. For example, a circuit may include a most significant bit (MSB) path and a least significant bit (LSB) path. The MSB path may include a first delta-sigma modulator having first and second outputs and a first digital-to-analog converter coupled to the first output of the first delta-sigma modulator. The LSB path comprises a second delta-sigma modulator comprising a loop filter and a quantizer. The quantizer may have an input coupled to the loop filter and to the digital filter. The LSB path may further include a second digital-to-analog converter coupled to an output of the quantizer. The circuit may further include a digital filter and/or a gain stage interposed between the MSB path and the LSB path.