Time-to-digital converter based on vernier double-chain structure
By adopting a vernier double-chain time-to-digital converter, the area and power consumption problems of TDC at high resolution are solved, realizing high-precision time measurement with low dead zone, which is suitable for scenarios such as lidar and quantum computing.
Patent Information
- Application Number
- CN202512001286.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-29
- Publication Date
- 2026-05-01
AI Technical Summary
Existing time-to-digital converters (TDCs) face challenges in achieving high resolution, including increased chip area and power consumption, as well as measurement dead zones and nonlinear errors, making it difficult to meet the demands of modern electronic systems for high precision, low power consumption, and small size.
A time-to-digital converter based on a vernier double-chain structure is adopted. Through the combination of a signal holding circuit, a vernier double-chain coarse measurement module, a data selector, a time amplifier module, and a digital encoding unit, high-precision, low-dead-zone, small-area, and low-power time measurement is achieved.
It achieves a system resolution of 3.0 ps, eliminates measurement dead zone, reduces chip area by more than 80%, reduces power consumption by 75%, and has a nonlinear error of <0.5%, making it suitable for high-precision applications such as lidar and quantum computing.
Smart Images

Figure CN121956475A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of integrated circuit technology, and in particular to a time-to-digital converter based on a vernier double-chain structure. Background Technology
[0002] With the rapid development of technologies such as 5G / 6G communication, LiDAR, and quantum computing, the demand for high-precision time measurement is becoming increasingly urgent. Time-to-digital converters (TDCs), as key circuit modules that convert time intervals into digital quantities, play a crucial role in precision measurement, communication, and signal processing. However, as application scenarios increasingly demand higher time resolution, existing TDC technologies face significant challenges.
[0003] In advanced electronic systems, the resolution requirements for time-delay line (TDC) have increased from the traditional 100-200 ps to sub-10 ps or even single-digit picosecond levels. However, while achieving high resolution, existing TDC technologies often come at the cost of a significant increase in chip area and power consumption. For example, in a traditional single-chain delay line structure, hundreds of delay units are typically required to achieve 10 ps resolution, resulting in a significant increase in chip area. Furthermore, when further resolution needs to be improved, simply increasing the number of delay units is limited by the minimum process delay and cannot effectively solve the problem. Although the vernier structure achieves time amplification through mutual calibration of two delay chains, it has an inherent measurement dead zone problem. Within certain specific time intervals, TDC cannot accurately measure or output stable results, typically accounting for 10-20% of the measurement range. This discontinuity can cause measurement jumps in applications such as lidar, seriously affecting system reliability.
[0004] Nonlinear errors caused by process variations are also a major problem in existing time-delay measurement (TDC) technologies. Due to process fluctuations during manufacturing, the actual delay of the delay unit deviates from the design value, resulting in nonlinear characteristics in the measurement results. Traditional TDCs typically require complex calibration circuits to compensate for these errors, further increasing system complexity and power consumption. To be compatible with modern electronic systems, TDCs are required to maintain high resolution while also having a small area, low power consumption, and good scalability. However, existing technologies struggle to meet these requirements simultaneously: high-resolution TDCs are usually large in area and consume a lot of power; small-area, low-power TDCs have limited resolution. Therefore, developing a TDC architecture that can achieve high precision, low dead time, small area, low power consumption, and easy scalability is of great significance for meeting the high-precision time measurement needs of modern electronic systems. Summary of the Invention
[0005] The purpose of this invention is to address the problems of insufficient resolution, measurement dead zone, excessive area and power consumption, significant nonlinear error and poor scalability of existing TDC technology, and to provide a time-to-digital converter based on a vernier double-chain structure to achieve high-precision, low-dead-zone, small-area and low-power time measurement.
[0006] To achieve the above objectives, the present invention adopts the following technical solution:
[0007] A time-to-digital converter based on a vernier double-chain structure includes a signal holding circuit, a vernier double-chain coarse measurement module, a data selector, a time amplifier module, and a digital encoding unit.
[0008] The signal holding circuit is used to sample and hold the START and STOP input signals, and output a stable sampled signal to the vernier double-chain coarse measurement module.
[0009] The vernier dual-chain coarse measurement module includes a main delay chain and a vernier delay chain. Each delay chain consists of a series-connected inverter delay unit and a D flip-flop. The START signal is input to the main delay chain, and the STOP signal is input to the vernier delay chain. By comparing the output states of the corresponding D flip-flops in the two chains, a coarse measurement result of the time difference between the START and STOP signals is obtained. The inverter delay unit adjusts the width-to-length ratio of the MOS transistor to control the unit delay difference between the main delay chain and the vernier delay chain to within 10 ps.
[0010] The data selector is used to select the phase difference node that is closest to the time difference between the main delay chain and the vernier delay chain based on the coarse measurement result output by the D flip-flop of the vernier double-chain coarse measurement module, and output the residual time difference signal to the time amplifier module.
[0011] The time amplifier module includes a transmission gate, an RC network, and a current source, which is used to perform fine measurement on the phase difference node signal (i.e., the residual time difference after coarse measurement) output by the data selector, and output a voltage signal that is linearly related to the residual time difference.
[0012] The digital encoding unit is used to receive the coarse measurement data output by the vernier double-chain coarse measurement module and the fine measurement data output by the time amplifier module, integrate and encode the two types of data, and output the final digital time measurement result.
[0013] Furthermore, the inverter delay unit consists of two inverters connected in series to ensure that the logic state remains unchanged after the signal passes through the delay unit; in the main delay chain and the vernier delay chain, the parameters of the second-stage inverters of the delay unit are completely identical, while the width-to-length ratio of the MOS transistors of the first-stage inverters differs. By adjusting the width-to-length ratio (W / L) of the MOS transistors, the unit delay difference is adjusted so that the unit delay time difference between the main delay chain and the vernier delay chain is controlled within 10 ps.
[0014] Furthermore, the data selector selects the closest phase difference node based on the coarse digital code output by the D flip-flop and outputs it to the time amplifier module for fine measurement.
[0015] Furthermore, the transmission gates in the time amplifier module include two parallel transmission gates, which are used to control the charging switch and discharging circuit of the RC network, respectively; the resistor R in the RC network is 20-30Ω, the capacitor C is 400-500fF, and the current source provides a 10mA bias current; this time amplifier module can provide a linear time-to-voltage conversion gain of 11 mV / ps in the 0-10 ps input range, with a nonlinearity error of <0.5%. Preferably, R=30Ω and C=500fF.
[0016] Furthermore, the time constant of the RC network is matched with the coarse measurement resolution of the vernier double-chain coarse measurement module to ensure that the fine measurement range covers the residual time difference after the coarse measurement.
[0017] Furthermore, the present invention is configured with only one time amplifier module, which uses time-division multiplexing to perform fine measurements on each phase difference node selected by the data selector in sequence, thereby significantly reducing chip area and power consumption.
[0018] Furthermore, the present invention adopts a modular design, using a 4-bit TDC as the basic functional unit, and achieving higher bit expansion by cascading multiple 4-bit TDC units; two inverters with high fan-out capability are connected in series between two adjacent 4-bit TDC units to reshape the signal, eliminate timing jitter and delay errors accumulated in the previous module, and realize integral linearity error reset.
[0019] Furthermore, 8-32 bit TDC expansion is achieved by cascading 2-8 TDC units; the modules are connected via a front-stage output → strong-drive inverter → rear-stage input, and the strong-drive inverter is used to reset the cascaded accumulated integral linearity error.
[0020] Furthermore, the fan-out capability of the strong drive inverter is ≥8, which can reshape the previous stage signal and ensure signal integrity; the cascaded modules share a set of time amplifier modules and digital encoding units, and the timing synchronization of each unit is achieved through the cascade controller.
[0021] Compared with the prior art, the beneficial effects achieved by the technical solution of this invention are:
[0022] 1. Two-stage measurement architecture: The system adopts a two-stage architecture of "vernier dual-chain coarse measurement + time amplifier fine measurement". The vernier dual-chain provides a coarse time resolution of 10 ps, and the time amplifier performs fine measurement on the residual time difference after coarse measurement, ultimately achieving a system resolution of 3.0 ps.
[0023] 2. Dead Zone Elimination: This invention uses a two-stage measurement architecture of "vernier double-chain coarse measurement + time amplifier fine measurement" to ensure that the measurement range is continuous and without dead zones, thus fundamentally solving the dead zone problem of traditional vernier TDC.
[0024] 3. Area and power consumption optimization: Using only one time amplifier module, a data selector is used to select specific nodes for fine measurement. Compared with the traditional solution (equipping each node with an amplifier), the chip area is reduced by more than 80% and the power consumption is reduced by 75%.
[0025] 4. High linearity: The dedicated time amplifier module provides a linear time-to-voltage conversion gain of 11 mV / ps with a nonlinearity error of <0.5%, which is significantly better than existing technologies (typically <5 mV / ps).
[0026] 5. Wide applicability: Low power consumption design achieved under 180nm CMOS process, system resolution up to 3.0 ps, meeting the needs of high-precision application scenarios such as LiDAR (distance accuracy <0.45 mm) and quantum computing. Attached Figure Description
[0027] Figure 1 This is a schematic diagram illustrating the working process of the time-to-digital conversion circuit of the present invention.
[0028] Figure 2 This is a design diagram of the 4-bit vernier double-chain coarse measurement module described in Example 1 (including an inverter delay module and a D flip-flop).
[0029] Figure 3 This is a schematic diagram of the overall structure of the 4-bit time-to-digital converter described in Example 1.
[0030] Figure 4 The waveform diagram of the 4-bit vernier double-chain coarse measurement module in Example 1 under a test signal with a time difference of 10 ps is shown (Q1, Q2, Q3, and Q4 are the output results of the first to fourth flip-flops, respectively). Figures 4-7 (This is true for all of them).
[0031] Figure 5 The waveform diagram of the 4-bit vernier double-chain coarse measurement module in Example 1 under a test signal with a time difference of 20ps is shown.
[0032] Figure 6 The waveform diagram of the 4-bit vernier double-chain coarse measurement module in Example 1 under a test signal with a time difference of 30ps is shown.
[0033] Figure 7 The waveform diagram of the 4-bit vernier double-chain coarse measurement module in Example 1 under a test signal with a time difference of 40ps is shown.
[0034] Figure 8 This is a schematic diagram of the time amplifier in Example 1.
[0035] Figure 9 The waveform diagram for the linearity test of the time amplifier in Example 1 is shown.
[0036] Figure 10 This is a schematic diagram of the structure of the 16-bit TDC in Example 2.
[0037] Figure 11 This is a schematic diagram of the time amplifier in Comparative Example 1, including the control circuit and the amplification circuit.
[0038] Figure 12 The circuit diagram is shown in Comparative Example 1.
[0039] Figure 13 This is the circuit schematic of the amplifier circuit in Comparative Example 1.
[0040] Figure 14 The graph shows the change in the output time difference and the slope of the output time difference from the input of the time amplifier provided in Comparative Example 1, from -200ps to 200ps. Detailed Implementation
[0041] To make the technical problems, technical solutions and beneficial effects of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.
[0042] Example 1: Implementation of 4-bit TDC (180 nm CMOS process)
[0043] like Figure 3 The diagram shown is a schematic representation of the overall circuit structure of the 4-bit time-to-digital converter in Embodiment 1 of the present invention. This diagram illustrates the complete implementation scheme of the "coarse measurement-fine measurement" two-level architecture of the present invention at the system level, specifically including the following key modules and their connections:
[0044] Signal Holding Circuit: Located on the left side of the diagram, this circuit receives the externally input START and STOP signals. Its function is to sample and stabilize these two signals during the measurement period, ensuring signal integrity in subsequent processing stages. Simultaneously, this circuit also generates delayed START and STOP signals, which are then re-inputted to the main delay chain and secondary delay chain, respectively, to achieve accurate time difference measurement.
[0045] The vernier dual-chain coarse measurement module consists of two parallel delay chains: an upper-level main delay chain and a lower-level secondary delay chain. Each chain contains four cascaded delay units, each consisting of two inverters to ensure that the logic state of the signal remains unchanged after delay. The START signal is delayed and then input to the main delay chain, and the STOP signal is delayed and then input to the secondary delay chain.
[0046] D flip-flop array: Located between corresponding nodes of the main delay chain and the secondary delay chain, its clock input (CLK) is connected to the output of the secondary delay chain, and its data input (D) is connected to the corresponding node of the main delay chain. When the signal of the secondary delay chain arrives at a certain node, the D flip-flop captures the signal state of the main delay chain at that node and outputs coarse measurement results such as Q1, Q2, Q3, and Q4, forming a 4-bit binary code to represent a rough value of the time difference.
[0047] Data selector: Receives coarse measurement results from the D flip-flop array and selects the specific node requiring fine measurement based on these results. For example, if the coarse measurement results show that the time difference falls within the second delay unit interval, the data selector will select the residual time difference signal for that interval and send it to the time amplifier.
[0048] Time amplifier module: Receives the residual time difference signal selected by the data selector and amplifies it with high gain. Based on RC network and transmission gate design, this module can convert minute time differences into quantifiable voltage signals, achieving sub-picosecond precision resolution.
[0049] Signal path and workflow: As shown in the figure, the workflow of the entire system is as follows: The START and STOP signals are first sampled by the signal holding circuit; then they enter the dual-chain TDC for coarse measurement; the D flip-flop outputs the coarse measurement result for the data selector to judge; the selected residual time difference signal is sent to the time amplifier for fine measurement; finally, the fine measurement result is combined with the coarse measurement result to form a complete high-precision time measurement value.
[0050] This embodiment implements the TDC system of the present invention using a 180nm CMOS process, specifically including:
[0051] 1) System preparation:
[0052] A dual-chain TDC structure was designed using a 180nm CMOS process with a power supply voltage of 1.8V. Each chain contains 8 MOS transistor delay units and D flip-flops. A dedicated time amplifier module was designed, including transmission gates and RC networks.
[0053] 2) Vernier delay double-chain design:
[0054] like Figure 2 As shown, the main delay chain (upper chain) consists of four sets of delay units, each set composed of two inverters connected in series to ensure the signal logic state remains unchanged. The secondary delay chain (lower chain) also consists of four sets of delay units with the same structure. The START signal is input to the beginning of the main delay chain, and the STOP signal is input to the beginning of the secondary delay chain. The delay difference between the two chains is controlled by precisely designed MOS transistor parameters. A D flip-flop array connects the corresponding nodes of the two chains, and a coarse value of the time difference is determined by sampling the output state, encoded as a 4-bit coarse measurement result.
[0055] 3) Delay Unit Design
[0056] like Figure 2 As shown, a delay unit consists of two inverters. The main delay chain consists of inverter 1 and inverter 3, and the secondary delay chain consists of inverter 2 and inverter 3. The parameters of the second inverter (inverter 3) in both delay chains are identical; the parameters of the first inverter are different. The delay difference is controlled by adjusting the width-to-length ratio (W / L) of the MOS transistor in the first inverter. Main chain inverter 1 MOS transistor parameters: NMOS transistor W / L = 2μm / 0.18μm; PMOS transistor W / L = 4μm / 0.18μm; Secondary chain inverter 2 MOS transistor parameters: NMOS transistor W / L = 3μm / 0.18μm; PMOS transistor W / L = 6μm / 0.18μm; Unit delay difference: approximately 10ps (measured value, 180nm CMOS process, 25℃).
[0057] 4) Time amplifier module design:
[0058] The schematic diagram of the time amplifier in Example 1 is shown below. Figure 8 As shown, the core of the time amplifier includes two transmission gates, an RC network, and a current source: the upper transmission gate 1 controls the capacitor charging switch; the lower transmission gate 2 forms the capacitor discharge circuit; the RC network includes a resistor R1 (20Ω) and a capacitor C0 (400fF). The current source provides a 10mA bias current.
[0059] Working principle: The phase difference signal (i.e., the residual time difference after coarse measurement) output by the dual-chain TDC is input to the transmission gate. The transmission gate controls the charging and discharging process of the RC network, and the output voltage is linearly related to the input time difference. The time constant of the RC network τ = RC = 8ps, which matches the 10ps coarse measurement resolution.
[0060] The waveform test diagram of the time amplifier linearity in Example 1 is shown below. Figure 9 As shown, the horizontal axis represents the input residual time difference (in ps), and the vertical axis represents the output voltage (in mV). This test introduces a test signal at 800 ps to begin charging the capacitor. Figure 9 It can be seen that the time amplifier maintains excellent linearity within 30 ps, and the output voltage is linearly related to the input time difference: Vout = 11 mV / ps × Δt. This indicates that the time amplifier can perform good secondary measurements on the coarse measurement signal (10 ps time resolution), thereby outputting fine measurement results.
[0061] 5) Workflow:
[0062] Figure 1A schematic diagram of the time-to-digital conversion circuit of this invention is provided. It can be seen that the pulse signals to be measured (START and STOP signals) are divided into two parts: one part is directly input to the vernier double chain, and the other part is sampled and held by a signal holding circuit. The test signal directly input to the vernier double chain generates a coarse measurement result (generated by signals Q1, Q2, Q3, and Q4) through a D flip-flop array. The data selector selects the specific node requiring fine measurement based on the coarse measurement result and connects that node to the time amplifier module. The signal delayed by the holding circuit is input again to the vernier double chain coarse measurement module. The regenerated coarse measurement structure is then finely measured by the time amplifier to measure the residual time difference, outputting a voltage signal. Integrating the coarse and fine measurement results, the final time value is output.
[0063] Figure 4 The waveform diagram of the 4-bit vernier double-chain coarse measurement module in Example 1 under a test signal with a time difference of 10 ps is shown (Q1, Q2, Q3, and Q4 are the output results of the first to fourth flip-flops, respectively). Figures 4-7 (This is true for all of them). Figure 5 The waveform diagram of the 4-bit vernier double-chain coarse measurement module in Example 1 under a test signal with a time difference of 20ps is shown. Figure 6 The waveform diagram of the 4-bit vernier double-chain coarse measurement module in Example 1 under a test signal with a time difference of 30ps is shown. Figure 7 The image shows the waveform of the 4-bit vernier double-chain coarse measurement module in Example 1 under a test signal with a time difference of 40 ps. From... Figures 4 to 7 As can be seen from the results, when the interval between the START and STOP signals is 10ps, the test result is Q1 = 1 and Q2-Q4 = 0; when the interval is 20ps, the test result is Q1 and Q2 = 1 and Q3 and Q4 = 0; when the interval is 30ps, the test result is Q1-Q3 = 1 and Q4 = 0; and when the interval is 40ps, the test result is Q1-Q4 = 1. Figures 4-7 This indicates that the time resolution of the 4-bit TDC is 10ps. Further explanation is that every 10ps, one bit of the output signal Q1-Q4 will sequentially change from 0 to 1, thus forming a coarse measurement result, which is used to locate the subsequent fine measurement signal, which is output from a single D flip-flop.
[0064] Example 2: 16-bit TDC Extension Implementation
[0065] like Figure 10 The diagram shown is an extended structure schematic of the 16-bit time-to-digital converter in Embodiment 2 of the present invention. This diagram illustrates how a TDC system with higher resolution and a wider measurement range can be achieved by cascading multiple 4-bit time-to-digital converter modules. Specifically, it includes the following key components:
[0066] Basic Unit Module: The "4-bit Time-to-Digital Converter" that appears repeatedly in the figure is the basic TDC unit described in Embodiment 1 of this invention, and its internal structure is as follows: Figure 2 As shown, it includes a vernier delay double chain consisting of 4 sets of delay units and a D flip-flop, forming a 4-bit TDC module with a time resolution of 10ps.
[0067] Cascading Connection: Four 4-bit Time-to-Digital Converter (TDC) modules are cascaded through two consecutive inverters (3) to form a complete 16-bit TDC system. The output of each 4-bit TDC module is connected to the input of the next module through a pair of inverters (3), forming a serial cascade structure. This design utilizes inverters (3) as signal buffers and delay matching units to ensure signal integrity and timing synchronization between stages.
[0068] Signal transmission path: As shown in the diagram, the input signal first enters the first 4-bit TDC module for preliminary time difference measurement; its output is transmitted to the second 4-bit TDC module via inverter 3 to continue the subsequent measurement stages; and so on, until the fourth 4-bit TDC module completes the full 16-bit time encoding. This step-by-step processing method enables segmented measurement of long-delay signals.
[0069] 1) Module Cascading Architecture:
[0070] This invention adopts a modular design, using a 4-bit TDC as the basic unit, and cascading four identical units to achieve a 16-bit TDC; each 4-bit TDC module contains a complete main delay chain, a vernier delay chain, a D flip-flop array, and a data selector; the modules are connected by high-speed signal lines to form a cascaded structure.
[0071] 2) Module connection method:
[0072] The first-level module receives the original START and STOP signals; the output of each level module (coarse measurement result) serves as the input signal for the next level module; the connection method is "pre-stage output → driver → post-stage input"; two inverters with strong driving capability are inserted between every two 4-bit TDC modules.
[0073] 3) Integral error reset mechanism:
[0074] Since multi-stage cascading may lead to the accumulation of delay errors, this invention introduces a strong-drive inverter at the module connection point. The inverter has a high fan-out capability (capable of driving ≥8 subsequent units) to ensure signal integrity. Through signal reshaping, the timing jitter and delay errors accumulated in the preceding modules are effectively eliminated, achieving "integral linear error reset" and preventing the error from accumulating as the number of stages increases.
[0075] 4) 16-bit TDC workflow:
[0076] The START and STOP signals are input to the first-stage 4-bit TDC module. The signals are processed by the first-stage module to generate a 4-bit coarse measurement result. After being reshaped by a strong-drive inverter (i.e., inverter 3), the signal is input to the second-stage module. The above process is repeated until the fourth-stage module completes the 16-bit coarse measurement. The data selector selects the specific node that needs to be measured in detail based on the 16-bit coarse measurement result. The phase difference signal of the selected node is led out to a single time amplifier for fine measurement. The digital encoding unit integrates the 16-bit coarse measurement result and the fine measurement result and outputs the final time value.
[0077] 5) Advantages of modular design:
[0078] Error control: Inter-module inverters effectively reset the integral linearity error, ensuring consistent measurement accuracy at each stage.
[0079] Architecture reuse: The 4-bit TDC module can be directly reused, simplifying the design process.
[0080] Flexible expansion: The number of bits can be expanded by adding modules without redesigning.
[0081] Resource efficiency: Only one time amplifier module is still needed, and fine measurements can be achieved through a 16-to-1 data selector.
[0082] Signal integrity: The strong drive inverter ensures the quality of cascaded signals and avoids signal attenuation caused by long-distance transmission.
[0083] Through the above embodiments, this invention achieves high-precision, low-dead-zone, small-area, and low-power time measurement. Using a 180nm CMOS process, the system resolution reaches 3.0 ps, making it particularly suitable for high-precision applications such as lidar and quantum computing. This invention resolves the inherent contradiction between area and power consumption in high-resolution time-displacement measurement (TDC), achieving the design goal of "obtaining the highest measurement accuracy with minimal hardware resources."
[0084] Comparative Example 1
[0085] One existing dead-zone-free time-to-digital converter (TDC) scheme employs a dual-path time amplifier architecture. Its core idea is to introduce a fixed delay module, allowing two parallel time amplification paths to operate in different time regions. This ensures that when one amplifier is in its dead zone, the other can still function normally, thus achieving dead-zone-free measurement. This scheme uses two complete time amplifiers and two delay links, compensating for process variations by adjusting the output capacitance difference through an offset calibration circuit. While this solves the small-signal dead-zone problem of the time amplifier, the dual-path architecture results in a large chip area and high power consumption, and it cannot fundamentally address the inherent defects of traditional vernier TDC architectures.
[0086] Figure 11A schematic diagram of the time amplifier in Comparative Example 1 is given, including the control circuit and the amplification circuit. Figure 12 The circuit diagram of the control circuit in Comparative Example 1 is given. Figure 13 The circuit diagram of the amplifier circuit in Comparative Example 1 is given. Figure 14 The graph shows the change in the output time difference and the slope of the output time difference from the input of the time amplifier provided in Comparative Example 1, from -200ps to 200ps.
[0087] Compared to the dual-path time amplifier architecture used in Comparative Example 1, the vernier dual-chain structure TDC of this invention has significant advantages in terms of technical implementation and performance:
[0088] This invention adopts a two-level architecture of "coarse measurement-fine measurement". Only a single time amplifier is needed to achieve the fine measurement function through time division multiplexing, while Comparative Example 1 requires two complete time amplifiers and two independent delay paths, resulting in a significant increase in chip area.
[0089] In terms of power consumption, the single-time amplifier architecture of this invention avoids the problem of static and dynamic power consumption increasing exponentially due to the dual-path structure in Comparative Example 1, and the measured power consumption is reduced by about 75%, which is of great significance for low-power application scenarios.
[0090] In terms of calibration mechanism, this invention achieves a delay cell matching degree of less than 10ps through a vernier double-chain structure and MOS parameter adjustment technology, without the need for an additional offset calibration circuit; while Comparative Example 1 must rely on a complex offset calibration circuit and additional calibration time, which increases system complexity and startup delay.
[0091] Regarding the solution to the dead zone problem: Comparative Example 1 only solved the dead zone problem of the time amplifier itself, but still could not overcome the inherent dead zone caused by the mismatch of delay units in the traditional vernier TDC architecture; while this invention, through the innovative "coarse measurement-fine measurement" architecture, fundamentally and completely eliminates the measurement dead zone, realizes seamless measurement across the entire range, and improves the system resolution to 3.0ps, which is significantly better than the performance of Comparative Example 1.
[0092] Experiments show that this invention achieves a coarse measurement at the 10 ps level by adjusting the width-to-length ratio of the MOS transistor to make the delay difference of the dual-chain unit ≤10 ps; after locating the residual time difference node by the data selector, a single time amplifier module performs time-division multiplexing to complete a fine measurement at the 3.0 ps level, and finally integrates the output results. This invention completely eliminates the measurement dead zone, reduces the chip area by more than 80%, reduces power consumption by 75%, and has a nonlinearity error of <0.5%. It can be modularly expanded to 8-32 bits; it has stable performance under operating conditions of -40℃ to 85℃, and is suitable for scenarios such as lidar (distance accuracy <0.45 mm) and quantum computing. It can be implemented using 180 nm CMOS technology, with good compatibility and controllable cost.
[0093] The above embodiments are merely preferred embodiments of the present invention and should not be considered as limiting the scope of the present invention. All equivalent variations and improvements made within the scope of the present invention should still fall within the patent coverage of the present invention.
Claims
1. A time-to-digital converter based on a vernier double-chain structure, characterized in that, It includes a signal holding circuit, a vernier double-chain coarse measurement module, a data selector, a time amplifier module, and a digital encoding unit; The signal holding circuit is used to sample and hold the START and STOP input signals, and output a stable sampled signal to the vernier double-chain coarse measurement module. The vernier dual-chain coarse measurement module includes a main delay chain and a vernier delay chain. Each delay chain consists of a series-connected inverter delay unit and a D flip-flop. The START signal is input to the main delay chain, and the STOP signal is input to the vernier delay chain. By comparing the output states of the corresponding D flip-flops in the two chains, the coarse measurement result of the time difference between the START and STOP signals is obtained. The data selector is used to select the phase difference node that is closest to the time difference between the main delay chain and the vernier delay chain based on the coarse measurement result output by the D flip-flop of the vernier double-chain coarse measurement module, and output the residual time difference signal to the time amplifier module. The time amplifier module includes a transmission gate, an RC network, and a current source, which is used to perform fine measurement on the residual time difference after coarse measurement of the data selector output, and output a voltage signal that is linearly related to the residual time difference. The digital encoding unit is used to receive the coarse measurement data output by the vernier double-chain coarse measurement module and the fine measurement data output by the time amplifier module, integrate and encode the two types of data, and output the final digital time measurement result.
2. The time-to-digital converter based on a vernier double-chain structure according to claim 1, characterized in that, The inverter delay unit consists of two inverters connected in series to ensure that the logic state remains unchanged after the signal passes through the delay unit. In the main delay chain and the vernier delay chain, the parameters of the second-stage inverters in the delay unit are completely identical, while the width-to-length ratio of the MOS transistors in the first-stage inverters differs. By adjusting the width-to-length ratio W / L of the MOS transistors, the unit delay difference is adjusted so that the unit delay time difference between the main delay chain and the vernier delay chain is controlled within 10 ps.
3. The time-to-digital converter based on a vernier double-chain structure according to claim 1, characterized in that, The data selector selects the closest phase difference node based on the coarse digital code output by the D flip-flop and outputs it to the time amplifier module for fine measurement.
4. The time-to-digital converter based on a vernier double-chain structure according to claim 1, characterized in that, The time amplifier module includes two parallel transmission gates, which are used to control the charging switch and discharging circuit of the RC network, respectively. The resistor R in the RC network is 20-30Ω, the capacitor C is 400-500fF, and the current source provides a 10mA bias current. The time amplifier module can provide a linear time-to-voltage conversion gain of 11 mV / ps in the 0-10 ps input range, with a nonlinearity error of <0.5%.
5. A time-to-digital converter based on a vernier double-chain structure according to claim 1, characterized in that, The time constant of the RC network is matched with the coarse measurement resolution of the vernier double-chain coarse measurement module to ensure that the fine measurement range covers the residual time difference after the coarse measurement.
6. A time-to-digital converter based on a vernier double-chain structure according to claim 1, characterized in that, With only one time amplifier module configured, the phase difference nodes selected by the data selector are measured in sequence through time-division multiplexing, thereby reducing chip area and power consumption.
7. A time-to-digital converter based on a vernier double-chain structure according to claim 1, characterized in that, The modular design uses a 4-bit TDC as the basic functional unit, and multiple 4-bit TDC units are cascaded to achieve higher bit extensions. Two inverters with high fan-out capability are connected in series between two adjacent 4-bit TDC units to reshape the signal, eliminate timing jitter and delay errors accumulated in the previous module, and realize integral linearity error reset.
8. A time-to-digital converter based on a vernier double-chain structure according to claim 7, characterized in that, 8-32 bit TDC expansion is achieved by cascading 2-8 TDC units; the modules are connected via front-stage output → strong-drive inverter → rear-stage input, and the strong-drive inverter is used to reset the cascaded accumulated integral linearity error.
9. A time-to-digital converter based on a vernier double-chain structure according to claim 8, characterized in that, The fan-out capability of the strong drive inverter is ≥8, which can reshape the previous stage signal and ensure signal integrity; the cascaded modules share a set of time amplifier modules and digital encoding units, and the timing synchronization of each unit is achieved through the cascade controller.
