Multi-Branch Sampling Converter for Wideband High-Resolution ADC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional analog-to-digital converters (ADCs) face limitations in achieving high instantaneous bandwidth and resolution, particularly at very high sample rates, due to issues such as quantization noise, clock jitter, thermal noise, and component tolerances, which restrict their utility in applications requiring high-resolution, high-speed signal processing.
Innovation Solution
The proposed solution involves a Multi-Channel Bandpass Oversampling (MBO) converter that employs continuous-time Diplexing Feedback Loops (DFLs) for noise shaping and Bandpass Moving-Average filters for signal reconstruction, along with nonlinear bit-mapping to mitigate quantization errors and improve resolution and bandwidth, allowing for more accurate conversion of continuous-time signals to discrete-time signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional ADC approaches are used, then implementation is straightforward, but instantaneous bandwidth and resolution are limited
Solution Approach 1:
The converter is divided into multiple parallel processing branches, each handling a different frequency band. Each branch includes its own quantization-noise-shaping circuit, sampling/quantization circuit, and digital bandpass filter, allowing simultaneous processing of multiple signal components to achieve high bandwidth while maintaining resolution
Solution Approach 2:
The invention transitions from time-domain oversampling to frequency-domain parallel processing by introducing multiple processing branches that operate simultaneously on different frequency bands. This dimensional shift from sequential time-based processing to parallel frequency-based processing enables both high resolution and wide instantaneous bandwidth
2Measurement precision
If oversampling is used to improve resolution, then quantization noise is reduced, but instantaneous bandwidth is limited by the Nyquist criterion
Solution Approach 1:
The signal bandwidth is segmented into multiple frequency bands, with each processing branch handling a specific band. This allows the system to achieve high effective sample rate for each branch while the combined output provides wide instantaneous bandwidth, resolving the Nyquist limitation
Solution Approach 2:
The invention changes the operating parameters by using bandpass filtering instead of lowpass filtering, and by operating each branch at a lower sample rate relative to its bandwidth while the parallel combination achieves high overall bandwidth. This parameter transformation allows exceeding the conventional Nyquist limit
3Productivity
If Nyquist-rate conversion is used, then bandwidth is maximized, but resolution is limited by quantization noise and implementation impairments
Solution Approach 1:
Each processing branch implements quantization-noise-shaping circuits that use feedback mechanisms to push quantization noise away from the signal band. This feedback-based noise shaping maintains high resolution despite high-speed operation by reducing the impact of quantization errors in the passband
Solution Approach 2:
Each processing branch is optimized for its specific frequency band with dedicated bandpass filters and noise-shaping circuits. This local optimization ensures that each branch achieves high resolution for its assigned band, and the combined output provides high overall resolution across the wide bandwidth
4Speed
If flash converters are used for high bandwidth, then speed is improved, but resolution is limited by clock jitter and component tolerances
Solution Approach 1:
The invention introduces digital bandpass filters as intermediary stages between the sampling/quantization circuits and the final output. These filters serve as mediators that clean up quantization noise and signal imperfections, improving resolution without compromising the high-speed performance of the flash converter architecture
Solution Approach 2:
The invention changes the filtering approach from lowpass to bandpass, and operates each branch at a sample rate matched to its bandwidth rather than the overall system bandwidth. This parameter optimization reduces the impact of clock jitter and component tolerances while maintaining high conversion speed
Data Source
AI summary
Provided are, among other things, systems, apparatuses, methods and techniques for converting a continuous-time, continuously variable signal into a sampled and quantized signal. One such apparatus includes an input line for accepting an input signal that is continuous in time and continuously variable, multiple processing branches coupled to the input line, and an adder coupled to outputs of the processing branches. Each of the processing branches includes a continuous-time quantization-noise-shaping circuit, a sampling/quantization circuit coupled to an output of the continuous-time quantization-noise-shaping circuit, a digital bandpass filter coupled to an output of the sampling/quantization circuit, and a line coupling an output of the digital-to-analog converter circuit back into the continuous-time quantization-noise-shaping circuit. A center frequency of the digital bandpass filter in each the processing branch corresponds to a minimum in a quantization noise transfer function for the continuous-time quantization-noise-shaping circuit in the same processing branch.