Gated Ring Oscillator ADC Noise Shaping for High-Bandwidth Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing analog-to-digital converters (ADCs) face challenges in meeting power consumption requirements for high-bandwidth telecommunications, particularly due to the power-hungry operational amplifiers needed in high-order continuous time delta sigma modulators and the limited resolution of voltage-controlled oscillator-based quantizers, making them impractical for ultra-high-speed applications.
Innovation Solution
The implementation of noise-shaping enhanced gated ring oscillator (GRO)-based ADCs, which include a delay to feedback an error signal, a voltage-to-time converter to convert the error signal into a time domain signal, a GRO to oscillate based on the time domain signal, a quantization device to generate a front-end output signal, and a time-to-digital converter to digitize the quantization error, providing additional noise shaping and reducing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If high-order continuous time delta sigma modulators are used to achieve high bandwidth, then bandwidth is improved, but power consumption increases due to power-hungry operational amplifiers
Solution Approach 1:
The patent extracts and eliminates the operational amplifiers from the ADC architecture by using a gated ring oscillator-based quantizer. The GRO quantizer replaces the traditional op-amp-based feedback loop, removing the power-hungry components while maintaining high-order noise shaping through a different architectural approach that uses oscillation-based quantization rather than voltage-based feedback.
Solution Approach 2:
The patent substitutes the electrical/voltage-based operational amplifier system with a time-based oscillation system. The gated ring oscillator uses timing and frequency domain operations instead of voltage domain operations, replacing the mechanical/electrical feedback mechanism of op-amps with an oscillation-based quantization mechanism that achieves similar noise shaping without the power consumption.
2Use of energy by moving object
If voltage-controlled oscillator based quantizers are used to eliminate operational amplifiers, then power consumption is reduced, but resolution is limited by VCO nonlinearity
Solution Approach 1:
The patent implements a feedback mechanism where the quantization error from the GRO is fed back through a noise shaping filter to the input summer. This feedback loop corrects the nonlinearity effects by shaping the quantization noise spectrum, allowing the system to achieve high resolution despite the inherent nonlinearity of the VCO-based quantizer. The feedback enables error correction without requiring additional power-hungry op-amps.
Solution Approach 2:
The patent changes the operating parameters and domain of the quantizer by using a gated ring oscillator that operates in the time and frequency domains rather than the voltage domain. By transforming the quantization process from voltage comparison to oscillation timing, the system achieves linearity improvement through the natural properties of the oscillation mechanism and the noise shaping feedback, overcoming the VCO nonlinearity issue.
3Measurement precision
If cascading multiple GRO-based ADCs is used to achieve higher order MASH, then noise shaping order is improved, but quantization noise becomes narrower making it impractical
Solution Approach 1:
The patent merges multiple noise shaping functions into a single integrated GRO-based architecture. Instead of cascading separate ADC stages, the invention combines the quantization process and noise shaping feedback into one unified system where the GRO quantizer works with a noise shaping filter and feedback summer to achieve high-order noise shaping in a single stage, avoiding the complexity of cascaded structures.
Solution Approach 2:
The patent transitions from the traditional time-domain cascaded MASH architecture to a frequency-domain approach using the gated ring oscillator. By operating in the frequency domain and using oscillation-based quantization, the system achieves higher order noise shaping without the bandwidth narrowing issues that plague time-domain cascaded structures, effectively moving the problem to a different dimensional space where the constraints are relaxed.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach results in a lower overall noise floor and improved signal-to-noise ratio, enabling efficient and scalable ADCs suitable for ultra-high-speed applications without the limitations of traditional ADCs, such as Giga sample per second performance and reduced area and power requirements.
Implementation Method 1
a voltage-to-time converter (VTC) configured to convert the error signal to a time domain error signal
Implementation Method 2
a GRO configured to output phase signals from the time domain error signal by oscillating when the time domain error signal is high and inhibiting oscillation when the time domain error signal is not high
Data Source
AI summary
A noise-shaping enhanced (NSE) gated ring oscillator (GRO)-based ADC includes a delay which delays and feedbacks an error signal to an input of the NSE GRO-based ADC. The feedback error signal provides an order of noise-shaping and the error signal is generated at the input of the NSE GRO-based ADC from an input signal, the feedback error signal, and a front-end output. A voltage-to-time converter converts the error signal to the time domain. A GRO outputs phase signals from the time domain error signal by oscillating when the error signal is high and inhibiting oscillation otherwise. A quantization device quantizes the phase signals to generate the front-end output. A quantization extraction device determines a quantization error from the quantized phase signals. A time-to-digital converter digitizes the quantization error to generate a back-end output. An output device generates a second order noise-shaped output based on the front-end and the back-end outputs.