Error-Feedback Time-to-Digital Converter for Low-Power Noise Shaping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current time-to-digital converters face challenges in achieving high resolution and low power consumption, particularly in low-power mobile applications, due to limitations in time resolution, noise shaping, and sensitivity to process-voltage-temperature variations.
Innovation Solution
The implementation of an error-feedback based time-to-digital converter with a feedback structure, noise-shaping characteristics, and oversampling to improve accuracy, utilizing a time register and digital-to-time converter to push quantization noise to high frequencies, thereby reducing in-band noise and enhancing resolution to 1-2 ps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a Vernier TDC uses two delay lines with inverter delay to achieve time resolution, then time resolution is improved, but a very large number of inverter stages is required which leads to delay mismatch and increased power consumption
Solution Approach 1:
The patent implements an error feedback mechanism where the quantization error from the time quantizer is fed back through a digital-to-time converter to the time register input. This feedback loop allows the system to correct quantization errors iteratively, achieving high time resolution without requiring a large number of delay stages, thus reducing power consumption while maintaining measurement precision.
Solution Approach 2:
The TDC is divided into multiple functional blocks: time register, time quantizer, digital-to-time converter, and feedback path. This segmentation allows each block to perform a specific function efficiently, avoiding the need for a monolithic structure with many inverter stages, thereby reducing overall power consumption while achieving fine time resolution through coordinated operation of the segmented components.
2Adaptability or versatility
If a Vernier TDC uses a very large number of inverter stages to cover a large detection range, then detection range is improved, but delay mismatch increases and power consumption increases
Solution Approach 1:
The patent employs dynamic elements including a gated ring oscillator that can be selectively enabled or disabled, and a feedback mechanism that adapts the quantization error correction based on the input signal characteristics. This dynamic operation allows the TDC to maintain high time resolution across a wide detection range without requiring a fixed large number of delay stages, as the system adapts its operation to the specific measurement requirements.
Solution Approach 2:
The system changes operational parameters dynamically: the gated ring oscillator is activated only when needed to extend the detection range, and the feedback loop adjusts the error correction based on the quantization error magnitude. This parameter changing approach allows the TDC to achieve both large detection range and high time resolution by optimizing the operating parameters rather than relying on a fixed large-scale structure.
3Measurement precision
If a gated-ring oscillator TDC operates in high frequency to achieve noise-shaping characteristic, then noise-shaping is improved, but power consumption becomes very high
Solution Approach 1:
The patent implements a feedback-based error correction mechanism that achieves noise-shaping characteristics without requiring high-frequency operation of a gated ring oscillator. The quantization error is fed back through a digital-to-time converter and subtracted from the input signal, effectively shaping the noise spectrum and pushing quantization noise to higher frequencies while operating at lower, more power-efficient frequencies.
Solution Approach 2:
The patent replaces the mechanical high-frequency oscillation approach with a digital feedback-based noise-shaping mechanism. Instead of relying on the physical oscillation of a gated ring oscillator at high frequencies to achieve noise-shaping, the system uses digital error feedback and subtraction to achieve the same noise-shaping effect at lower frequencies, significantly reducing power consumption while maintaining measurement precision.
4Device complexity
If the time quantizer uses a coarse quantization to reduce complexity, then device complexity is reduced, but quantization error increases
Solution Approach 1:
The patent employs a feedback mechanism where the quantization error from the coarse time quantizer is captured and fed back through a digital-to-time converter. This feedback path allows the system to correct the quantization error by subtracting the reconstructed quantization error from the original input signal, effectively eliminating the quantization error while maintaining the simplicity of the coarse quantizer structure.
Solution Approach 2:
The system discards the quantization error from the coarse quantizer output and recovers it through the feedback path. The digital-to-time converter reconstructs the quantization error from the quantizer output code, and this recovered error is then subtracted from the input signal to produce the final high-precision output. This discard-and-recover approach allows the use of simple coarse quantizers while achieving high measurement precision.
Data Source
AI summary
A time-to-digital converter includes: an input for receiving a time-domain input signal; an output for providing a digital output signal; a time register coupled to the input and to a first node; a time quantizer coupled to the time register for providing the digital output signal at the output; and a digital-to-time converter coupled to the output for providing a feed-back signal at the first node.