Vernier Time Difference Circuit for High-Resolution Low-Power TDCs

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing time-to-digital converters (TDCs) face challenges in achieving high resolution and low power consumption, particularly in fractional-N All-Digital Phase-Locked Loops (ADPLLs), due to limitations in delay line technology and high power consumption from high-frequency feedback signals.

Innovation Solution

A time difference determining device utilizing a Vernier technique with dual delay lines and samplers, where one delay line has faster cells and the other slower cells, allowing for high resolution and low power consumption by using a lower frequency sampling signal, and an output stage to determine the leading signal.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a standard delay line TDC architecture is used, then the device is simple to implement, but the resolution is limited by the minimum delay of buffers which is too high for many applications

Engineering Contradiction:
ImproveTDC resolutionVSAvoidTDC architecture complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The TDC is divided into two separate branches (first and second time difference determining branches), each with its own delay lines and samplers. This segmentation allows independent optimization of each branch's delay characteristics, enabling high resolution measurement while maintaining manageable complexity in each individual branch.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from a single delay line approach to a two-dimensional structure with two branches having different delay characteristics. The first branch uses delay cells with first delay values while the second branch uses delay cells with second delay values, creating an additional dimension in the measurement space that enables higher resolution without proportionally increasing complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Speed

If the DCO feedback frequency is kept high, then the PLL operates at high frequency, but the power consumption is very high due to the high frequency of the clock DCOFB

Engineering Contradiction:
ImprovePLL operating frequencyVSAvoidTDC power consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The patent employs periodic sampling where the second digital signal (lower frequency) periodically samples the delayed versions of the first digital signal. This periodic action at a lower frequency reduces the power consumption of the sampling operation while still capturing the time difference information, as the sampling occurs only at necessary intervals rather than continuously at the high DCO feedback frequency.

Inventive Principle:
Principle #19Periodic action

3Measurement precision

If a Vernier TDC is used to improve resolution, then the TDC resolution is significantly improved and independent from technology, but there is no power consumption reduction

Engineering Contradiction:
ImproveTDC resolutionVSAvoidTDC power consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The Vernier TDC is segmented into two independent branches with different delay characteristics. The first branch uses delay cells with first delay values and the second branch uses delay cells with second delay values. This segmentation allows each branch to operate at optimized frequencies, with the second branch sampling at a lower frequency to reduce power consumption while the first branch maintains high resolution capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the operating parameters of the two branches differently - the first branch operates with delay cells having first delay values at one frequency, while the second branch operates with delay cells having second delay values at a lower frequency. This parameter differentiation enables the system to achieve high resolution through the Vernier effect while reducing overall power consumption by operating one branch at lower frequency.

Inventive Principle:
Principle #35Parameter changes

4Use of energy by moving object

If DCO downconversion is used to reduce power consumption, then the power consumption is reduced by shifting down the DCO feedback frequency, but the resolution decreases by a factor of 2 and matching of delays becomes difficult

Engineering Contradiction:
ImproveTDC power consumptionVSAvoidTDC resolution
Core Design Contradiction:
Use of energy by moving objectVSMeasurement precision

Solution Approach 1:

Instead of downconverting a single TDC input, the patent segments the system into two branches where the first branch processes the original high-frequency signal for resolution-critical measurements, while the second branch processes a downconverted lower-frequency signal for power-efficient sampling. This segmentation allows the system to maintain high resolution in the first branch while achieving power reduction in the second branch, avoiding the resolution loss that would occur in a single downconverted system.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4026246B1Time difference determining device
Publication Date: 2025.07.30 HUAWEI TECH CO LTD
  • EP4026246B1 patent drawingFigure 1
  • EP4026246B1 patent drawingFigure 2~3
  • EP4026246B1 patent drawingFigure 4~5

AI summary

A time difference determining device determines a time difference between a first digital signal and a second digital signal. It comprises a first time difference determining branch, comprising a first delay line, in turn comprising N first delay cells, each of the N first delay cells having a first delay, and a second delay line, comprising N second delay cells, each of the N second delay cells having a second delay, the second delay (f) being higher than the first delay. Moreover, it comprises a second time difference determining branch, in turn comprising a third delay line comprising N third delay cells, each of the N third delay cells having a second delay, and a fourth delay line, comprising N fourth delay cells, each of the N fourth delay cells having the first delay.