A Time-to-Digital Converter with Steady Calibration through Single-Photon Detection

Basic Information

• Paper ID: 2406.01293
• Title: A time-to-digital converter with steady calibration through single-photon detection
• Authors: Matías R. Bolaños Wagner, Daniele Vogrig, Paolo Villoresi, Giuseppe Vallone, Andrea Stanco
• Classification: eess.SP physics.ins-det quant-ph
• Publication Date: June 4, 2024
• Paper Link: https://arxiv.org/abs/2406.01293

Abstract

This paper presents an FPGA-based time-to-digital converter (TDC) with 27 ps residual jitter and scalability for multi-channel operation. The design is specifically tailored for quantum key distribution (QKD) applications and employs a novel calibration method based on single-photon detection that requires neither interruption of data acquisition nor interpolation techniques, thereby improving accuracy and eliminating data loss. The calibration method was tested across a temperature range of 5°C to 80°C, with the device capable of continuously streaming data at rates up to 12 Mevents/s for approximately one week without TDC overflow.

Research Background and Motivation

Problem Context

1. Importance of TDC in Quantum Communication: Time-to-digital converters are critical tools in multiple domains, particularly in quantum communication, where the performance of time tagger systems significantly impacts the quality of the entire application.
2. Advantages and Challenges of FPGA-TDC: FPGA-based TDCs offer advantages over ASIC solutions in terms of flexibility, high-speed performance, and cost-effectiveness, but must address inherent nonlinear behavior of the devices.
3. Limitations of Traditional Calibration Methods: Existing calibration approaches typically rely on interpolation techniques that require periodic interruption of data acquisition for recalibration, resulting in data loss and reduced efficiency.

Research Motivation

1. Specific Requirements of Quantum Key Distribution: In satellite QKD applications, communication time windows are limited, necessitating maximization of data reception while maintaining performance. The data loss problem of traditional calibration methods is particularly acute in this context.
2. Need for Real-Time Calibration: A method is required that can continuously calibrate during data acquisition, avoiding performance degradation due to calibration interruptions or calibration failure.

Core Contributions

1. Proposed "Steady Calibration" Method: A real-time calibration technique based on single-photon detection that requires no interruption of data acquisition.
2. Designed and Implemented the MARTY TDC System: A multi-channel TDC based on Zynq7020 FPGA with excellent performance metrics.
3. Verified Temperature Adaptability: System performance was tested across a temperature range of 5°C to 80°C, demonstrating the effectiveness of steady calibration.
4. Provided QKD Application Validation: Comparative testing with commercial TDC verified the system's practicality for quantum key distribution.

Methodology Details

Task Definition

Design an FPGA-based time-to-digital converter capable of:

• Providing picosecond-level time resolution
• Supporting multi-channel operation
• Implementing real-time calibration without data loss
• Applicability to quantum communication systems

Model Architecture

1. TDC Hardware Architecture

Coarse Counting Module:

• Binary counter implemented using DSP48 components
• Clock frequency: fs = 412.5 MHz
• 48-bit encoding with total runtime: 2^48/fs ≈ 1 week

Fine Counting Module:

• Delay line implemented using fast carry chains (CARRY4 elements)
• 36 CARRY4 elements (144 fast carry components total)
• Coverage of coarse clock period τcoarse = 1/fs ≈ 2.42 ns

Decoding Algorithm:

• Pipeline adder tree algorithm proposed by Adamič et al.
• High robustness against "bubble errors"
• Thermometer code to binary code conversion

2. Steady Calibration Method

Traditional Calibration Process:

Minimum event count: Nevents > (zα/2/β)² · (M-1) = 116110
where: zα/2 corresponds to 99% confidence interval, β=0.02 tolerance, M=8 bits

Code Density Test:

• Propagation time calculation: δti = (wi/Σwj) · τsample
• Calibration time: t(c)i = (1/2)δti + Σj<i δtj

Steady Calibration Innovation:

• Ordered storage of code density histogram
• For each new event: remove oldest event, add new event
• Real-time recalculation of calibration curve
• Assumption of uniform distribution of detected events on TDL

Technical Innovation Points

1. Lossless Calibration: Unlike traditional methods requiring 30 ms downtime for switching to ring oscillator, steady calibration utilizes detection data itself for calibration.
2. Uniform Distribution Assumption: Ensured through non-commensurate clock frequencies and drift effects guaranteeing uniform distribution of input events.
3. Temperature Self-Adaptation: Calibration curve automatically adjusts with temperature variations without manual intervention.

Experimental Setup

Test Environment

• Temperature Control: Peltier unit + RAL 9006 climate chamber
• Temperature Range: 5°C to 80°C, 1°C increments
• Temperature Monitoring: XADC sensor measuring chip center temperature

Test Equipment

• Light Source: 1550 nm attenuated laser
• Detector: PMD-IR single-photon detector (MPD)
• Commercial Reference: QuTAG time tagger (QuTools)

Performance Metrics

1. Resolution: τres = t(c)Nc/Nc = 18.22 ps
2. Jitter: Gaussian fitting of dual-channel time difference FWHM
3. Differential Nonlinearity (DNL): DNLi = (wi - ⟨w⟩)/⟨w⟩
4. Quantum Bit Error Rate (QBER): QKD application performance indicator

Experimental Results

Primary Performance Metrics

• Resolution: 18.22 ps
• Jitter: 27.11 ps (room temperature)
• DNL Range: -0.97, 2.86
• Data Transfer Rate: Maximum 15 Mevents/s (Ethernet limited)
• Continuous Operation Time: Approximately 1 week without overflow

Temperature Characteristic Analysis

1. Propagation Time vs. Temperature Relationship

• Average propagation time inversely related to temperature
• Attributed to CARRY4 element characteristics of Zynq7020 chip (28 nm process)

2. Comparison of Four Calibration Schemes

1. 5°C Ring Oscillator Calibration + Hold: Jitter monotonically increases with temperature
2. 5°C Laser Calibration + Hold: Similar trend, confirming temperature dependency
3. Ring Oscillator Calibration at Each Temperature Point: Improvement before 30°C, then stabilization
4. Steady Laser Calibration: Best performance with standard deviation ⟨σsteady⟩ = 0.64 ps vs ⟨σRO⟩ = 1.33 ps

QKD Application Validation

• Test Configuration: Polarization-encoded HVDD sequence
• QBER Results: Both MARTY and QuTAG yield 2.2%
• Pulse Width: MARTY 107±3 ps, QuTAG 108±3 ps
• Conclusion: Despite higher MARTY jitter, pulse widths are comparable, primarily limited by detector jitter (~100 ps)

Related Work

FPGA-TDC Development

1. Early Work: Wang et al. (2010) full-featured FPGA-TDC implementation
2. High-Precision Solutions: Markovic et al.'s 17 ps single-stage vernier delay line
3. Commercialization Trends: Garzetti et al.'s IP core approach

Calibration Technology Evolution

1. Traditional Methods: Periodic interpolation-based calibration
2. Automated Calibration: Ito et al.'s random TDC self-calibration architecture
3. Real-Time Calibration: Chen et al.'s dual delay line real-time calibration

Quantum Communication Applications

1. QKD Time Synchronization: Calderaro et al.'s Qubit4Sync algorithm
2. Satellite Quantum Communication: Pirandola et al.'s quantum cryptography advances
3. Integrated Photonics: Avesani et al.'s all-photonic QKD system

Conclusions and Discussion

Main Conclusions

1. Steady Calibration Effectiveness: Successfully achieved real-time calibration without data loss, outperforming traditional methods
2. Temperature Adaptability: Maintained stable performance across 75°C temperature range
3. QKD Applicability: Performance comparable to commercial equipment, meeting quantum communication requirements
4. Scalability: Supports multi-channel operation and long-term continuous operation

Limitations

1. FPGA Process Constraints: 28 nm process limits further performance improvements
2. Uniform Distribution Assumption: Relies on clock non-commensurateness and drift effects
3. Temperature Range: While covering a wide range, extreme environment adaptability remains unverified
4. QKD Optimization: QKD system in testing was not fully optimized

Future Directions

1. Process Upgrade: Migration to Ultrascale+ (16 nm FinFET) technology
2. Online Calibration: Implementation of fully online steady calibration version
3. Multi-Channel Expansion: Further optimization of multi-channel parallel processing
4. Application Extension: Expansion to other fields requiring high-precision time measurement

In-Depth Evaluation

Strengths

1. Strong Technical Innovation: The steady calibration method represents an important innovation in the field, addressing the long-standing data loss problem
2. Comprehensive Experimental Design: Broad temperature testing coverage, comprehensive comparison schemes, and practically meaningful QKD validation
3. Excellent Engineering Implementation: High system integration, support for long-term stable operation, and high data transfer efficiency
4. Sufficient Theoretical Analysis: In-depth analysis of calibration principles, temperature effects, and performance metrics

Weaknesses

1. Theoretical Foundation: Theoretical proof of uniform distribution assumption is insufficient, relying on empirical observation
2. Performance Comparison: Limited comparison with TDCs implemented using latest FPGA processes
3. Application Scope: Primarily targeted at QKD applications; applicability to other scenarios requires further verification
4. Long-Term Stability: While one-week operation was tested, long-term drift characteristics require deeper analysis

Impact

1. Academic Contribution: Provides new perspectives for TDC calibration technology, potentially influencing subsequent research directions
2. Practical Value: Directly applicable to quantum communication systems, particularly satellite QKD applications
3. Reproducibility: Sufficient design details based on standard FPGA platforms, facilitating reproduction and improvement
4. Commercialization Potential: Good cost-effectiveness, high technical maturity, and commercial viability

Applicable Scenarios

1. Quantum Communication: Time tagging for QKD systems, particularly suited for satellite quantum communication
2. High-Energy Physics: Time measurement in particle detection experiments
3. Light Detection and Ranging: LiDAR systems requiring high-precision time measurement
4. Industrial Measurement: Time measurement applications requiring continuous operation and high reliability

References

The paper cites 27 important references covering key domains including TDC technology, FPGA implementation, quantum communication, and calibration methods, providing a solid theoretical foundation and technical benchmarks for the research.

Overall Assessment: This is a high-quality technical paper with the proposed steady calibration method possessing significant theoretical and practical value. The experimental design is rigorous, results are credible, and it has direct application value for the quantum communication field. While there is room for improvement in theoretical analysis and performance comparison, it represents an important contribution to the field overall.