Multiple measurement average type time-to-digital converter, system and signal processing method thereof

By propagating multi-edge square wave signals in the delay chain, combining multi-edge and multi-period measurement methods, the resource consumption and dead time problems of high-precision time digital converters in the prior art are solved, and efficient picosecond-level measurement accuracy is achieved and low resource consumption.

CN119045302APending Publication Date: 2024-11-29UNIV OF SCI & TECH OF CHINA
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202411455336.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-18
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

When existing high-precision time-digital converters reach picosecond measurement accuracy, they face the contradiction between measurement accuracy, dead time and resource consumption, especially the hardware resource consumption of multi-chain measurement average TDC and multi-edge measurement average TDC. and coding complexity problems.

Method used

Design a multi-time measurement average time digital converter, using square wave signal generation module, delay chain module and data processing by propagating square wave signals with multiple edges in the delay chain, combining multi-edge and multi-period measurement methods The module determines the rising edge of the signal to be measured by sampling latch sequence value processing, and combines the advantages of multi-edge measurement and multi-period measurement to reduce resource consumption and dead time.

Benefits of technology

It realizes that while achieving the same measurement accuracy, reduce resource consumption and dead time, reduce delay chain length and system clock cycle, solves the contradiction between measurement accuracy and resource consumption, and improves the efficiency and accuracy of the time-digital converter.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119045302A_ABST
    Figure CN119045302A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-measurement average type time-to-digital converter, a system and a signal processing method thereof, which can be applied to the technical fields of laser radar ranging, nuclear medicine, nuclear physics experiments and the like relating to high-precision time interval measurement. When receiving the rising edge of the signal to be measured, the multi-measurement average type time-to-digital converter can continuously generate a square wave signal, and the square wave signal enters the delay chain to be propagated once the square wave signal is generated, so that the tap value of the delay unit in the delay chain jumps, and the signal to be measured is obtained. The sampling latch sequence value corresponding to each system clock rising edge moment is processed so as to determine the rising edge moments of the signals to be measured, and then high-precision time interval information is obtained by measuring the time difference between the rising edge moments of two adjacent signals to be measured. According to the circuit and the signal processing mode, the advantages of multi-edge measurement and multi-period measurement can be fused, the time difference of rising edges of two signals to be measured is measured, and the contradiction between the measurement precision and dead time and resource consumption is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of high-precision time interval measurement, and in particular to a multiple-measurement average time-to-digital converter, system and signal processing method thereof based on FPGA (Field-Programmable Gate Array). Background Art

[0002] Time-to-Digital Converter (TDC) can convert time interval information into high-resolution digital signals, and has broad application prospects in aerospace, deep space exploration, geological mapping, medical imaging, radar scanning and other fields. The currently widely used high-precision time-to-digital converter generally adopts the method of deploying a delay chain (TDL) in FPGA, which completes the timing measurement by measuring the time interval between the rising edges of two signals.

[0003] like Figure 1 and Figure 2As shown, the basic structure of TDL type TDC consists of two parts: one is a system clock cycle counter running at the system clock frequency, which counts up by one at each time the rising edge of the system clock arrives, and outputs the counting result in the next system clock cycle when the signal to be measured arrives. This counting result multiplied by the system clock cycle is the time of the system clock rising edge, which is called the coarse time. The second is a delay chain formed by cascading multiple delay units. The delay chain is responsible for directly receiving the signal to be tested. When the signal to be tested arrives, the signal to be tested directly enters the delay chain for propagation until the next rising edge of the system clock arrives. The rising edge of the system clock triggers the D flip-flop sequence to sample and latch the tap values ​​of each delay unit in the delay chain. Since the signal to be tested requires a fixed delay time to pass through each delay unit, and the tap value of the delay unit will jump when the signal to be tested passes through each delay unit, the delay time of the signal to be tested can be calculated by observing the sampling and latching sequence value of the delay chain, that is, the sampling and latching sequence value of the delay chain uniquely and definitely corresponds to the delay time of a signal to be tested, and this delay time is the time interval between the rising edge of the signal to be tested and the rising edge of the next system clock. Since the delay time of the delay chain is much smaller than the system clock cycle, the resolution of the delay time measured by the delay chain is much higher than the resolution of the coarse time measured by the system clock cycle counter, so this delay time is called fine time. By subtracting the fine time measured by the delay chain from the coarse time measured by the system clock cycle counter, a digital timing information with a higher resolution of the rising edge of a test signal can be obtained. The same method can be used to obtain digital timing information with a higher resolution of the rising edge of another test signal. Since the rising edge of the test signal with an earlier arrival time represents the starting time of the time interval measurement, and the rising edge of the test signal with a later arrival time represents the ending time of the time interval measurement, the required timing measurement result with a higher clock resolution can be obtained by subtracting the digital timing information of the rising edges of the two test signals measured by the time-to-digital converter.

[0004] In practical applications, considering the impact of development cycle, development cost, application difficulty, etc., TDL type TDC is often deployed using FPGA, which is implemented by using the field compilable logic resources in FPGA. However, due to factors such as economic cost and development cycle, the hardware process in FPGA is often made of mature process rather than advanced process, resulting in large device delay. Therefore, when deploying the delay chain, it leads to a large system clock frequency, a long system clock cycle, a large delay of each delay unit in the delay chain, a large delay difference between delay units, and poor system consistency. Therefore, in order to achieve picosecond timing accuracy, it is necessary to adopt a method of multiple measurement averaging. The method of multiple measurement averaging can prevent the results of multiple measurements from falling entirely in the delay unit with huge delay, that is, the measurement results falling in the delay unit with smaller delay can effectively alleviate the measurement inaccuracy caused by the delay unit with huge delay, and significantly reduce the measurement error.

[0005] There are three types of TDC currently using the multiple measurement averaging method: multi-chain method TDC, multi-edge measurement averaging TDC (Wave Union A method TDC), and multi-cycle measurement averaging TDC (Wave Union B method TDC).

[0006] The structure of multi-chain measurement average TDC is as follows Figure 3 As shown, its working principle is the same as that of TDL TDC. It introduces the signal to be measured into multiple delay chains at the same time for propagation, obtains multiple fine time measurement results, and thereby obtains multiple timing measurement results. The timing measurement results are averaged to obtain a timing measurement result with higher accuracy. In current practice, the RMS (root mean square) resolution of the timing measurement result can reach the picosecond level. The number of multiple measurements of the multi-chain measurement averaging TDC is consistent with the number of delay chains. In order to complete the purpose of multiple measurements, a large number of delay chains are required, which significantly increases the consumption of hardware resources.

[0007] The structure of multi-edge measurement average TDC is as follows Figure 4As shown, its working principle is to use the signal to be measured to trigger a signal sequence with multiple edges (also called edge change) to propagate in a delay chain. Since each edge has the same propagation time in the delay chain, the purpose of multiple measurements can be achieved by measuring the propagation positions of these edges in the delay chain respectively. Similar to the multi-chain measurement average TDC, the edge falling in the delay unit with a shorter delay time can reduce the timing inaccuracy caused by the edge falling in the delay unit with a larger delay time, so that the measurement accuracy of picosecond level can be achieved. Since the increase in the number of multiple measurements requires an increase in the number of edges, the signal sequence with multiple edges has a certain width, and in order to enable the circuit sequence to identify each edge and encode its specific delay position, the required delay chain length will increase to accommodate the signal sequence with multiple edges to propagate in it, which will significantly increase the consumption of hardware resources and significantly increase the difficulty and complexity of coding.

[0008] The structure of multi-cycle measurement average TDC is as follows Figure 5 As shown, the signal to be tested is introduced into a delay chain for propagation, and a return channel is set to introduce the signal to be tested into the delay chain for propagation multiple times in the next few system clock cycles, thereby completing the delay timing. At the arrival time of the next few system clock rising edges, the tap values ​​of each delay unit in the delay chain are sampled and latched to obtain the propagation position of the signal to be tested at that moment, and the position uniquely corresponds to a fine time. Each timing result is increased by a fixed time, which is a system clock cycle. Since there is a certain difference between this time and the delay time of the signal to be tested passing through a whole delay chain and entering the delay chain again through the return channel, the rising edge of the signal to be tested will fall in different delay units each time the sampling and latching occurs, so the measurement result falling in the delay unit with a smaller delay time can be used to reduce the measurement error and achieve picosecond measurement accuracy. Since the number of measurements in this way is consistent with the number of system clock cycles used for repeated measurements, this way will result in a longer dead time for the measurement, making it unsuitable for application scenarios with a high case rate.

[0009] In order to resolve the contradiction between measurement accuracy, dead time and resource consumption, it is necessary to invent and design a time-to-digital converter with multi-edge and multi-cycle measurement capabilities. Summary of the invention

[0010] The purpose of the present application is to provide a multiple measurement averaging time-to-digital converter, system and signal processing method thereof, which can resolve the contradiction between measurement accuracy and dead time and resource consumption.

[0011] To achieve the above objectives, this application provides the following solutions:

[0012] In a first aspect, the present application provides a multiple measurement average time-to-digital converter, the multiple measurement average time-to-digital converter comprising:

[0013] A square wave signal generating module is used to continuously generate a square wave signal when receiving a rising edge of a signal to be tested, until the Ath system clock rising edge moment after the rising edge of the signal to be tested arrives; the square wave signal is a signal with multiple edges, and the edge is a rising edge or a falling edge; the square wave signal is used to propagate in a delay chain in a delay chain module, so that the tap value of a delay unit in the delay chain jumps; A is a positive integer greater than 1, and the square wave signal in each system clock cycle has multiple edges;

[0014] The delay chain module is connected to the square wave signal generating module, and is used to receive the square wave signal, and to sample and latch the tap value of each delay unit in the delay chain at each system clock rising edge, so as to obtain a sampling and latching sequence value corresponding to each system clock rising edge; the sampling and latching sequence value is used to characterize the propagation position of each edge in the square wave signal in the delay chain at each system clock rising edge;

[0015] The data processing module is connected to the delay chain module and is used to process the sampling latch sequence value corresponding to each system clock rising edge moment to determine the rising edge moment of the signal to be measured.

[0016] In a second aspect, the present application provides a signal processing method of the above-mentioned multiple measurement averaging type time-to-digital converter, and the signal processing method of the multiple measurement averaging type time-to-digital converter comprises:

[0017] Get the sampling latch sequence value corresponding to each system clock rising edge;

[0018] The sampling latch sequence value corresponding to each rising edge of the system clock is processed to determine the rising edge moment of the signal to be measured.

[0019] In a third aspect, the present application provides a time interval measurement system, the time interval measurement system comprising: a time interval measurement module and a plurality of the above-mentioned multiple measurement averaging type time-to-digital converters;

[0020] When a multiple measurement average type time digital converter is included, the multiple measurement average type time digital converter is used to respectively determine the time when the rising edge of the signal to be measured starts and the time when the rising edge of the signal to be measured stops; the time interval measurement module is connected to the multiple measurement average type time digital converter, and is used to calculate the difference between the time when the rising edge of the signal to be measured starts and the time when the rising edge of the signal to be measured stops, so as to obtain the second time interval;

[0021] When two multiple measurement averaging type time digital converters are included, they are respectively recorded as the first time digital converter and the second time digital converter; the first time digital converter is used to determine the moment of starting the rising edge of the signal to be measured; the second time digital converter is used to determine the moment of stopping the rising edge of the signal to be measured; the time interval measurement module is connected to the first time digital converter and the second time digital converter respectively, and is used to calculate the difference between the moment of starting the rising edge of the signal to be measured and the moment of stopping the rising edge of the signal to be measured, so as to obtain the second time interval.

[0022] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0023] The present application provides a multiple measurement average time digital converter, system and signal processing method thereof, wherein a square wave signal generating module is used to continuously generate a square wave signal upon receiving a rising edge of a signal to be measured, until the Ath system clock rising edge moment after the rising edge of the signal to be measured arrives, the square wave signal is a signal with multiple edges, which propagates in a delay chain in a delay chain module, so that the tap value of a delay unit in the delay chain jumps; the delay chain module is used to sample and latch the tap value of each delay unit in the delay chain at each system clock rising edge moment, and obtain a sampling latch sequence value corresponding to each system clock rising edge moment; the data processing module is used to process the sampling latch sequence value corresponding to each system clock rising edge moment, and determine the moment of the rising edge of the signal to be measured. The present application generates a square wave signal with multiple edges in multiple system clock cycles by setting a square wave signal generating module. The square wave signal along the edge can integrate the advantages of multi-edge measurement averaging TDC and multi-cycle measurement averaging TDC. Compared with multi-edge measurement averaging TDC, under the premise of achieving the same number of measurements, since multiple system clock cycles are used, there is no need to arrange too many edges, and resource consumption is less. Compared with multi-cycle measurement averaging TDC, under the premise of achieving the same number of measurements, since each system clock cycle has multiple edges, there is no need for too many system clock cycles, and dead time is less. At the same time, multiple measurements are combined to improve measurement accuracy. That is, the present application integrates multi-edge and multi-cycle measurement capabilities to ensure that the delay chain length accommodating multiple edges is shorter, and solves the problems of long delay chain and high resource consumption caused by multi-edge measurement averaging TDC. At the same time, it can significantly reduce the system clock cycle required for measurement and reduce the dead time of the system, thereby resolving the contradiction between measurement accuracy and dead time and resource consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 This is a schematic diagram of the structure of a common TDL type TDC provided in this application.

[0026] Figure 2 A timing diagram of a common TDL type TDC provided in this application.

[0027] Figure 3 A schematic diagram of the structure of a multi-chain measurement averaging type TDC provided in this application.

[0028] Figure 4 A schematic diagram of the structure of a multi-edge measurement averaging type TDC provided in this application.

[0029] Figure 5 A schematic diagram of the structure of a multi-cycle measurement averaging type TDC provided in this application.

[0030] Figure 6 This is a schematic diagram of the structure of a multiple measurement averaging time-to-digital converter provided in Example 1 of the present application.

[0031] Figure 7 This is a schematic diagram of the delay chain input timing provided in Example 1 of the present application.

[0032] Figure 8 A schematic diagram of the sampling latch sequence values ​​provided in Example 1 of the present application.

[0033] Fig. 9 This is a schematic diagram of the principle of bubble error provided in Example 1 of the present application.

[0034] Fig.10 This is a schematic diagram of the principle of the code density method provided in Example 1 of the present application.

[0035] Fig.11 A schematic diagram of the corresponding relationship between the code value stored in the first sampling latch and the delay time provided in Example 1 of the present application.

[0036] Fig.12 A schematic diagram of the corresponding relationship between the code value stored in the second sampling latch and the delay time provided in Example 1 of the present application.

[0037] Fig.13 A schematic diagram of the corresponding relationship between the code value stored in the four-time sampling latch and the delay time provided in Example 1 of the present application.

[0038] Fig.14 A schematic diagram of the code value processing provided in Example 1 of the present application. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0040] Example 1

[0041] like Figure 6 As shown, this embodiment provides a multiple measurement average type time-to-digital converter, and the multiple measurement average type time-to-digital converter includes:

[0042] The square wave signal generating module is used to continuously generate a square wave signal when receiving the rising edge of the signal to be tested, until the Ath system clock rising edge moment after the rising edge of the signal to be tested arrives. The square wave signal is a signal with multiple edges, where the edge is a rising edge or a falling edge. The square wave signal is used to propagate in the delay chain in the delay chain module, so that the tap value of the delay unit in the delay chain jumps. A is a positive integer greater than 1, and the square wave signal of each system clock cycle has multiple edges.

[0043] The delay chain module is connected to the square wave signal generating module, and is used to receive the square wave signal, and to sample and latch the tap value of each delay unit in the delay chain at each rising edge of the system clock, and obtain the sampling and latching sequence value corresponding to each rising edge of the system clock. The sampling and latching sequence value is used to characterize the propagation position of each edge in the square wave signal in the delay chain at each rising edge of the system clock.

[0044] The data processing module is connected to the delay chain module and is used to process the sampling latch sequence value corresponding to each system clock rising edge moment to determine the rising edge moment of the signal to be measured.

[0045] This embodiment only needs to measure fewer but more than one edge position in each system clock cycle, and repeatedly measures fewer but more than one edge position in fewer but more than one system clock cycle, which can ensure that the delay chain length accommodating multiple edges is short, solve the problem of longer delay chain caused by the multi-edge multiple measurement method, and reduce resource consumption. In addition, in several adjacent system clock cycles, multiple edges are measured in each system clock cycle, so when performing multiple measurements of a fixed number of times, the system clock cycle required for measurement can be significantly reduced, and multiple measurements of a fixed number of times can be completed, reducing the dead time of the system, thereby proposing a multi-edge multi-cycle multiple measurement average type time digital converter circuit structure, which can be named Wave Union D-type TDC, to solve the contradiction between measurement accuracy and dead time and resource consumption.

[0046] In this embodiment, the square wave signal generating module specifically includes:

[0047] The first D flip-flop is used for continuously outputting a high level signal when receiving a rising edge of the signal to be tested, until the Ath system clock rising edge moment after the rising edge of the signal to be tested arrives.

[0048] The lookup table is respectively connected to the Q port of the first D flip-flop and the tap port of the Bth delay unit of the delay chain, and is used to output a high level when the first D flip-flop outputs a high level and the Bth delay unit outputs a low level, and to output a low level when the first D flip-flop outputs a high level and the Bth delay unit outputs a high level, so as to generate a square wave signal.

[0049] Among them, the CLK port of the first D flip-flop is connected to the signal to be tested, the D port of the first D flip-flop is connected to a high-level signal, the Q port of the first D flip-flop is connected to an input port of the lookup table, and the Clr port of the first D flip-flop is connected to the reset module. The reset module is used to reset the first D flip-flop at the Ath system clock rising edge moment after the rising edge of the signal to be tested arrives, so that the first D flip-flop outputs a low-level signal.

[0050] In this embodiment, the function of the first D flip-flop is to control the pulse width of the signal to be tested to a required length, which is determined by A. When the rising edge of the signal to be tested arrives, it immediately enters the CLK port of the first D flip-flop, and the first D flip-flop is immediately triggered, and the high-level signal of the D port is sent to the Q port, then the Q port jumps from a low level to a high level, and continues to output a high-level signal until the Ath system clock rising edge after the rising edge of the signal to be tested arrives. The reset module pulls up the Clr port of the first D flip-flop, resets the first D flip-flop, resets the Q port level, and returns to a low level, thereby completing the control of the pulse width of the signal to be tested.

[0051] In this embodiment, the function of the lookup table is to continuously generate a square wave signal having multiple edges, and the start time of the square wave signal (ie, the time when the first rising edge is generated) is aligned with the rising edge of the signal to be measured.

[0052] The functions of the lookup table are shown in Table 1:

[0053] Table 1 Lookup table function table

[0054]

[0055] like Figure 7 As shown, when the rising edge of the signal to be tested does not arrive, the input of the input port 1 of the lookup table is low level 0, then the lookup table outputs low level 0, and the input of the input port 2 of the lookup table is also low level 0. When the rising edge of the signal to be tested arrives, the input of input port 1 is high level 1, and the input of input port 2 is still low level 0 due to the propagation delay of the delay unit, then the lookup table outputs high level 1, and the output high level immediately enters the delay chain and propagates, and reaches the return signal point (i.e., the tap port of the Bth delay unit) after propagating for a period of time. On the one hand, this high level continues to propagate forward along the delay chain, and on the other hand, it propagates to the input port 2 of the lookup table through the return path, so that the input of input port 2 is high level 1, then the lookup table immediately outputs low level 0, and the low level output by the lookup table enters the delay chain and propagates. Similar to the high level, it changes the level of the input port 2 of the lookup table through the return path, and again makes the lookup table output a high level, and so on, so that a square wave signal with multiple edges propagates in the delay chain until the first D flip-flop receives the reset signal, so that the Q port is pulled low, and when the input of the input port 1 of the lookup table is low level 0, the lookup table continues to output a low level to propagate in the delay chain, completing the reset of the delay chain.

[0056] In this embodiment, the reset module includes:

[0057] The signal detector to be tested is connected to the tap ports of several delay units in the delay chain, and is used to detect whether the rising edge of the signal to be tested has arrived according to the tap values ​​of the several delay units and trigger the square wave signal to propagate in the delay chain to obtain a judgment result. The several delay units include the first delay unit.

[0058] The reset counter is connected to the system clock signal and the output port of the signal detector to be tested respectively, and is used to start counting the system clock cycle of the system clock signal to obtain a first count value when the judgment result is yes.

[0059] The reset logic is connected to the output port of the reset counter and the Clr port of the first D flip-flop respectively, and is used to determine whether the first count value is equal to A, and reset the first D flip-flop when the first count value is equal to A.

[0060] The function of the signal detector to be tested is to monitor a certain bit, part or all of the sequence value given by the D flip-flop sequence. When the value of this part shows a specific feature, it means that the signal to be tested has entered the FPGA and has triggered the square wave signal to propagate in the delay chain through the previous series of circuit structures. Then the signal detector to be tested will send a signal to the reset counter, and the reset counter will start counting. The reset counter works under the system clock. After starting to count, it will add one at each rising edge of the system clock, indicating that several system clock cycles have passed. When the preset number A is reached, the reset counter sends a signal to the reset logic, and the reset logic sends a reset signal to the Clr port of the first D flip-flop to reset the first D flip-flop, so that the first D flip-flop outputs a low level.

[0061] This embodiment immediately releases a clock-shaped waveform signal with multiple edges and extremely short time intervals (about 100ps) between adjacent edges to propagate in the delay chain upon receiving the rising edge of the signal to be tested. This square wave signal lasts for several system clock cycles, and the number of system clock cycles can be adjusted as needed.

[0062] In this embodiment, the delay chain module specifically includes:

[0063] The D flip-flop sequence includes a plurality of second D flip-flops, each of which corresponds to a delay unit in a delay chain, the CLK port of the second D flip-flop is connected to a system clock signal, the D port of the second D flip-flop is connected to a tap port of the delay unit, and the second D flip-flop is used to sample and latch the tap value of the D port and output it to the Q port at the rising edge of the system clock, so as to sample the output of the Q port of each second D flip-flop to obtain a sampled latch sequence value.

[0064] The function of the delay chain and the D flip-flop sequence is to use the system clock cycle as the sampling cycle, so that the D flip-flop sequence samples and latches the tap value of each delay unit in the delay chain. The sampling latch sequence value obtained by sampling and latching is used to characterize the propagation position of the square wave signal with multiple edges in the delay chain at the sampling moment, and this position can characterize the delay time of the signal to be measured.

[0065] When the rising edge of the signal to be measured arrives, a square wave signal with multiple edges begins to propagate in the delay chain until the next rising edge of the system clock arrives. The rising edge of the system clock drives the D flip-flop sequence to sample and latch the tap values ​​of each delay unit in the entire delay chain. The sampling latch sequence value obtained by the sampling latch is as follows: Figure 8As shown in the figure, the square wave signal in the delay chain propagates from left to right. The rightmost edge is generated at the same time as the rising edge of the signal to be measured. Once it is generated, it enters the delay chain for propagation. Similarly, once other edges are generated, they also enter the delay chain for propagation until the rising edge of the system clock arrives and is sampled and latched by the D flip-flop sequence, as shown in the figure. Figure 8 As shown in the first sampling latch sequence value in, the sampling latch sequence value obtained by sampling latch will be transmitted to the subsequent circuit structure for processing. The sampling latch of the delay chain by the D flip-flop sequence does not affect the propagation of the square wave signal in the delay chain. When the next system clock rising edge (i.e., the second system clock rising edge) arrives, the D flip-flop sequence will sample and latch the delay chain again, as shown in Figure 8 As shown in the second sample latch sequence value in . Since the total delay time of the delay chain is only slightly longer than one system clock cycle, the front end of the square wave waveform will propagate out of the delay chain, that is, Figure 8 As shown by the dotted line in , this part of the propagated signal will not be sampled and latched. Similarly, the third sampling latch sequence value, the fourth sampling latch sequence value and the fifth sampling latch sequence value are obtained. In this embodiment, A can be set to 5. When the fifth system clock rising edge is reached, the reset module applies a signal to the Clr port to reset the first D flip-flop, and the square wave signal is no longer generated. Therefore, no new edge signal will enter the delay chain at this time. As long as the last edge signal is propagated, the delay chain will return to the full low level state, and the sampling latch sequence value obtained by the D flip-flop sequence sampling and latching will be all 0, as shown in FIG. Figure 8 As shown in the sixth sampling latch sequence value in , since all edge signals in the delay chain have been propagated out at this time, the sampling latch sequence values ​​are all 0, which will be identified and removed in the subsequent encoding process.

[0066] In this embodiment, the data processing module specifically includes:

[0067] The encoder is connected to the Q port of each second D flip-flop in the D flip-flop sequence, and is used to encode the sampling latch sequence value corresponding to each system clock rising edge moment, and obtain the code value corresponding to each system clock rising edge moment. The code value is the binary code of the sum of the propagation positions of each edge in the square wave signal in the delay chain at each system clock rising edge moment.

[0068] The calibrator is connected to the encoder and is used to take the code value corresponding to the first system clock rising edge moment as input, and obtain the measurement result of the first measurement based on the correspondence between the first measurement code value and the delay time interval obtained by measuring using the code density method; take the code value corresponding to the a-th system clock rising edge moment as input, and obtain the measurement result of the a-th measurement based on the correspondence between the a-th measurement code value and the delay time interval when the first measurement code value obtained by measuring using the code density method is C; and determine the first time interval between the rising edge moment of the signal to be measured and the rising edge moment of the first system clock based on the measurement result of the A-th measurement. a=2, 3, ..., A, C are the code values ​​corresponding to the first system clock rising edge moment, and the measurement result is the delay time interval, which includes the upper limit and lower limit of the first time interval between the rising edge moment of the signal to be measured and the rising edge moment of the first system clock.

[0069] The system clock cycle counter is connected to the system clock signal and is used to output a second count value in the first system clock cycle. The second count value is obtained by counting the system clock cycle of the system clock signal.

[0070] The time calculator is used to calculate the product of the second count value and the system clock cycle, and calculate the difference between the product and the first time interval to obtain the time of the rising edge of the signal to be measured.

[0071] The role of the encoder is to identify and encode the edges of the sampled latch sequence values ​​obtained by sampling and latching the D flip-flop sequence, that is, first identify the edges in the sampled latch sequence values ​​(the places where 0 to 1 and 1 to 0 jump), then get the propagation positions of these edges, and then add up all the propagation positions of these edges to get the sum of the edge propagation positions, and finally perform binary encoding on the sum of the edge propagation positions to get the code value. Among them, the propagation position of the edge refers to the number of numbers on the left side of the edge (regardless of 1 or 0, it is a number), which is the edge position.

[0072] During the encoding process, you will face bubble problems, such as Fig. 9 As shown. It can be seen that the sampling latch sequence value obtained by the D flip-flop sequence sampling latch has the situation that 1 runs into 0 and 0 runs into 1 at the edge position. The area where these 0s and 1s are mixed is the bubble area, and the width of this bubble area (the number of digits contained) is called the bubble depth. The bubble depth in the FPGA used in this embodiment is 6, so in this embodiment, the sampling latch sequence value obtained by the D flip-flop sequence sampling latch is divided into 8 sub-sequences. In these 8 sub-sequences, the number of edges is consistent with the number of edges in the original sequence. The propagation positions of the edges in the 8 sub-sequences are added together, which is the propagation position of the edges in the original sequence. Then, the sum of the edge propagation positions in the 8 sub-sequences is the sum of the edge propagation positions in the original sequence.

[0073] In this embodiment, the code density method is used to pre-calibrate the correspondence between multiple measurement code values ​​and delay time intervals, and the measurement results of A measurements can be determined based on the code value corresponding to each system clock rising edge moment.

[0074] After obtaining the measurement results of A measurements, determine the median of the measurement results of each measurement (i.e., the delay time interval), and calculate the average of all medians to obtain the first time interval between the rising edge moment of the signal to be measured and the rising edge moment of the first system clock; or, select the maximum value of the lower limit of the measurement result of each measurement as the target lower limit, select the minimum value of the upper limit of the measurement result of each measurement as the target upper limit, determine the median of the target lower limit and the target upper limit, and obtain the first time interval between the rising edge moment of the signal to be measured and the rising edge moment of the first system clock.

[0075] The second count value is the number of system clock cycles from the initial system clock cycle of the system clock signal to the system clock cycle corresponding to the first system clock rising edge moment. The product of the second count value and the system clock cycle is the coarse time. The first time interval between the rising edge moment of the signal to be measured and the first system clock rising edge moment is the fine time. The time of the rising edge of the signal to be measured can be obtained by calculating the difference between the coarse time and the fine time. When it is desired to solve the second time interval between the rising edges of two signals to be measured, the above method can be used to determine the time of each rising edge of the signal to be measured, and further subtract them to obtain the second time interval between the rising edges of the two signals to be measured.

[0076] In this embodiment, when the rising edge of the signal to be measured arrives, the square wave signal is triggered to enter the delay chain and propagate until the next rising edge of the system clock arrives. The tap values ​​of each delay unit in the delay chain are sampled and latched, and the fine time is calculated based on the sampled latch sequence values ​​of the delay chain obtained at multiple system clock rising edge times. The measurement accuracy of the fine time can be improved through multiple measurements.

[0077] The corresponding relationship between the encoded code value and the delay time can be measured by the code density method. The working principle of the code density method is as follows: Fig.10 As shown in the figure, the signal to be measured is set to a period that is unrelated to the system clock period and measured in the same delay chain. After measuring enough times (for example, 100,000+), it can be seen as Fig.10The signal to be tested shown arrives evenly within a system clock cycle. Each time it arrives, it is sampled and latched when it propagates to the next system clock rising edge, and falls into different delay units. The code value falls into the number of delay units. Since the arrival time is uniform in time, the number of delay units it falls into is proportional to the delay time of the delay unit. When the ratio of the delay time of each delay unit is known, and the sum of the delay time of all delay units is known, which is a system clock cycle, the delay time of each delay unit can be known. Then, if another signal to be tested falls into a delay unit, the delay time interval is known, that is, one code value corresponds to one delay time interval. The delay time interval includes a lower limit and an upper limit, the lower limit is the minimum delay time, and the upper limit is the maximum delay time. The maximum delay time of the signal to be tested will not exceed the sum of the delay time of the delay unit and all the previous delay units, and the minimum will not be less than the sum of the delay time of all the delay units before this delay unit. In this embodiment, the median of the delay time interval is taken as the measured delay time.

[0078] like Fig.11 As shown, at the time of the first sampling and latching, as the number of edges increases, the delay time of each code value becomes shorter and shorter, which is reflected in the graph as the slope becomes smaller and smaller. The shorter the delay time corresponding to the code value, the smaller the uncertainty during measurement, and the more precise the measurement result. The code value obtained by the first sampling and latching can give the maximum and minimum values ​​of the delay time (fine time) of a signal to be measured. The code width is known, that is, the maximum value minus the minimum value. The code width is relatively wide, about the order of 20ps, but this embodiment hopes to reach 5ps, or even 3ps. Therefore, another sampling and latching is performed at the rising edge of the second system clock, as shown in FIG. Fig.12 As shown in the figure, the code value and the delay time do not correspond one to one, and the code value and the delay time do not have a linear relationship. The corresponding relationship between the code value and the delay time of the subsequent third and fourth sampling latches is basically similar to that of the second sampling latch, with only some minor shifts. Fig.13 , only the edge position and code value of the sampling latch sequence value obtained by the first sampling latch has a positive linear relationship with the time difference between the arrival time of the rising edge of the signal to be measured and the time of the next system clock rising edge (i.e., fine time). In other words, the larger the edge position and code value obtained by the first sampling latch, the longer the propagation time of the signal to be measured in the delay chain. Because of this linear relationship, we can know the corresponding relationship between the edge position and code value and the delay time. However, the edge position and code value obtained by the second, third, fourth, and fifth sampling latches are no longer linearly related to the delay time. Therefore, we need to find a corresponding relationship. Fig.13It can be seen that if the code value obtained by the first sampling latch is divided according to the corresponding relationship between the code value and the delay time, then the delay interval corresponding to each code value obtained by the first sampling latch is further divided at the second, third, and fourth sampling latches, such as Fig.14 .

[0079] Fig.14 In case 1, during the code density test, the code values ​​obtained by the first sampling latch are the same, for example, they are all 6. In case 1, the code values ​​obtained by the second, third, and fourth sampling latches will be different, but the obtained code values ​​are continuous, and the larger these continuous code values ​​are, the larger the corresponding delay time is. The code density method can be used to determine the code width corresponding to the code values ​​obtained by these subsequent sampling latches. Specifically, the interval corresponding to 6 is divided again to determine the corresponding relationship between the code values ​​obtained by the second, third, and fourth sampling latches and the sub-intervals obtained by dividing the delay time interval of 6. Then, the uncertainty can be reduced through multi-cycle measurement.

[0080] For case 2, some measurements will obtain some discontinuous code values, which is very easy to distinguish because it is discontinuous (for example, 258, 259, 223, 224, 198), while case 1 is continuous (for example, 229, 230, 231, 232, 233, 234). For case 2, the reason for the cliff-like drop in code value is that the last edge moves out of the sequence in each subsequence, and the code value decreases by 37 (subsequence length) for each subsequence moved out. Therefore, this embodiment fills in the edge that moves out of the sequence. For example, for 258, 259, 222, 223, 196, it is filled in as 258, 259, 223+37=260, 224+37=261, 198+37+37=262, and the number of fills is an integer multiple of 37. The result after filling is to form a continuous sequence. At this time, these code values ​​increase with the increase of delay time, which makes it return to situation 1. The code density method can be used to determine the code width of each newly obtained code value, thereby improving the accuracy of measurement.

[0081] This embodiment can reduce the circuit scale and the amount of FPGA hardware resources used. The low amount of resources can achieve multi-channel deployment, increase the number of channels of the delay chain in the FPGA, reduce costs, and improve integration. Combining the advantages of multi-edge measurement and multi-cycle measurement, while improving the measurement accuracy, it ensures that the circuit dead time is significantly lower than the multi-cycle measurement method, and ensures that the resource consumption of the chain circuit is significantly lower than the multi-edge measurement method, thereby achieving the effect of particularly saving resources and significantly reducing dead time.

[0082] Example 2

[0083] This embodiment provides a signal processing method of the multiple measurement averaging time-to-digital converter described in Embodiment 1, and the signal processing method of the multiple measurement averaging time-to-digital converter includes:

[0084] Get the sampling latch sequence value corresponding to each system clock rising edge.

[0085] The sampling latch sequence value corresponding to each rising edge of the system clock is processed to determine the rising edge moment of the signal to be measured.

[0086] The sampling latch sequence value corresponding to each system clock rising edge is processed to determine the rising edge moment of the signal to be measured, specifically including:

[0087] The sampling latch sequence value corresponding to each system clock rising edge moment is encoded respectively to obtain the code value corresponding to each system clock rising edge moment.

[0088] Taking the code value corresponding to the first system clock rising edge moment as input, based on the correspondence between the first measurement code value and the delay time interval measured by using the code density method, obtaining the measurement result of the first measurement; taking the code value corresponding to the a-th system clock rising edge moment as input, based on the correspondence between the a-th measurement code value and the delay time interval when the first measurement code value measured by using the code density method is C, obtaining the measurement result of the a-th measurement; based on the measurement results of the A-th measurement, determining the first time interval between the rising edge moment of the signal to be measured and the first system clock rising edge moment; a=2, 3, ..., A, A is the total number of system clock rising edge moments, and C is the code value corresponding to the first system clock rising edge moment.

[0089] The product of the second count value outputted in the first system clock cycle and the system clock cycle is calculated, and the difference between the product and the first time interval is calculated to obtain the time of the rising edge of the signal to be measured.

[0090] When the code density method is used to measure and obtain the correspondence between the a-th measured code value and the delay time interval when the first measured code value is C, when the first measured code value is C, if the extreme difference of multiple code values ​​obtained by the a-th measurement is greater than a preset value, a number of subsequence lengths are added to the target code value, so that the difference between the target code value after the number of subsequence lengths are added and the maximum value among the multiple code values ​​is less than the preset value, wherein the target code value is a code value whose difference with the maximum value among the multiple code values ​​is greater than the preset value.

[0091] Determining a first time interval between a rising edge of the signal to be measured and a rising edge of the first system clock based on the measurement results of the A measurements specifically includes:

[0092] The median value of each measurement result is determined, and the average value of all medians is calculated to obtain a first time interval between a rising edge of the signal to be measured and a rising edge of the first system clock.

[0093] Alternatively, the maximum value of the lower limit of the measurement result of each measurement is selected as the target lower limit, and the minimum value of the upper limit of the measurement result of each measurement is selected as the target upper limit, and the median of the target lower limit and the target upper limit is determined to obtain the first time interval between the rising edge moment of the signal to be measured and the rising edge moment of the first system clock.

[0094] Example 3

[0095] An embodiment of the present application provides a time interval measurement system, which includes: a time interval measurement module and several multiple measurement averaging time-to-digital converters described in Embodiment 1.

[0096] When a multiple-measurement averaging time-to-digital converter is included, the multiple-measurement averaging time-to-digital converter is used to respectively determine the moment when the rising edge of the signal to be measured starts and the moment when the rising edge of the signal to be measured stops; the time interval measurement module is connected to the multiple-measurement averaging time-to-digital converter, and is used to calculate the difference between the moment when the rising edge of the signal to be measured starts and the moment when the rising edge of the signal to be measured stops, so as to obtain a second time interval.

[0097] When two multiple measurement averaging type time digital converters are included, they are respectively recorded as the first time digital converter and the second time digital converter; the first time digital converter is used to determine the moment of starting the rising edge of the signal to be measured; the second time digital converter is used to determine the moment of stopping the rising edge of the signal to be measured; the time interval measurement module is connected to the first time digital converter and the second time digital converter respectively, and is used to calculate the difference between the moment of starting the rising edge of the signal to be measured and the moment of stopping the rising edge of the signal to be measured, so as to obtain the second time interval.

[0098] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A multiple measurement averaging type time-to-digital converter, characterized in that: The multiple measurement averaging type time-to-digital converter comprises: A square wave signal generating module is used to continuously generate a square wave signal when receiving a rising edge of a signal to be tested, until the Ath system clock rising edge moment after the rising edge of the signal to be tested arrives; the square wave signal is a signal with multiple edges, and the edge is a rising edge or a falling edge; the square wave signal is used to propagate in a delay chain in a delay chain module, so that the tap value of a delay unit in the delay chain jumps; A is a positive integer greater than 1, and the square wave signal in each system clock cycle has multiple edges; The delay chain module is connected to the square wave signal generating module, and is used to receive the square wave signal, and to sample and latch the tap value of each delay unit in the delay chain at each system clock rising edge, so as to obtain a sampling and latching sequence value corresponding to each system clock rising edge; the sampling and latching sequence value is used to characterize the propagation position of each edge in the square wave signal in the delay chain at each system clock rising edge; The data processing module is connected to the delay chain module and is used to process the sampling latch sequence value corresponding to each system clock rising edge moment to determine the rising edge moment of the signal to be measured.

2. The multiple measurement averaging type time-to-digital converter according to claim 1, characterized in that: The square wave signal generating module specifically comprises: The first D flip-flop is used to continuously output a high-level signal when receiving a rising edge of the signal to be tested until the Ath system clock rising edge after the rising edge of the signal to be tested arrives; The lookup table is respectively connected to the Q port of the first D flip-flop and the tap port of the Bth delay unit of the delay chain, and is used to output a high level when the first D flip-flop outputs a high level and the Bth delay unit outputs a low level, and to output a low level when the first D flip-flop outputs a high level and the Bth delay unit outputs a high level, so as to generate a square wave signal.

3. The multiple measurement averaging type time-to-digital converter according to claim 2, characterized in that: The CLK port of the first D flip-flop is connected to the signal to be tested, the D port of the first D flip-flop is connected to the high-level signal, the Q port of the first D flip-flop is connected to an input port of the lookup table, and the Clr port of the first D flip-flop is connected to the reset module; the reset module is used to reset the first D flip-flop at the Ath system clock rising edge time after the rising edge of the signal to be tested arrives, so that the first D flip-flop outputs a low-level signal; Wherein, the reset module includes: The signal detector to be tested is connected to the tap ports of several delay units in the delay chain, and is used to detect whether the rising edge of the signal to be tested has arrived according to the tap values ​​of several delay units and trigger the square wave signal to propagate in the delay chain to obtain a judgment result; A reset counter is connected to the output port of the system clock signal and the signal to be tested detector respectively, and is used to start counting the system clock cycle of the system clock signal to obtain a first count value when the judgment result is yes; The reset logic is connected to the output port of the reset counter and the Clr port of the first D flip-flop respectively, and is used to determine whether the first count value is equal to A, and reset the first D flip-flop when the first count value is equal to A.

4. The multiple measurement averaging type time-to-digital converter according to claim 1, characterized in that: The delay chain module specifically includes: The D flip-flop sequence includes a plurality of second D flip-flops, each of which corresponds to a delay unit in a delay chain, the CLK port of the second D flip-flop is connected to a system clock signal, and the D port of the second D flip-flop is connected to a tap port of the delay unit; the second D flip-flop is used to sample and latch the tap value of the D port and output it to the Q port at the rising edge of the system clock.

5. The multiple measurement averaging type time-to-digital converter according to claim 4, characterized in that: The data processing module specifically includes: An encoder connected to the Q port of each second D flip-flop in the D flip-flop sequence, for respectively encoding the sampling latch sequence value corresponding to each system clock rising edge moment to obtain a code value corresponding to each system clock rising edge moment; the code value is a binary code of the sum of the propagation positions of each edge in the square wave signal in the delay chain at each system clock rising edge moment; A calibrator is connected to the encoder and is used to take the code value corresponding to the first system clock rising edge moment as input, and obtain the measurement result of the first measurement based on the correspondence between the first measurement code value and the delay time interval obtained by measuring using a code density method; take the code value corresponding to the a-th system clock rising edge moment as input, and obtain the measurement result of the a-th measurement based on the correspondence between the a-th measurement code value and the delay time interval when the first measurement code value obtained by measuring using a code density method is C; determine the first time interval between the rising edge moment of the signal to be measured and the rising edge moment of the first system clock based on the measurement result of the A-th measurement; a=2, 3, ..., A; C is the code value corresponding to the first system clock rising edge moment; the measurement result is the delay time interval, and the delay time interval includes the upper limit and lower limit of the first time interval between the rising edge moment of the signal to be measured and the rising edge moment of the first system clock; A system clock cycle counter, connected to the system clock signal, for outputting a second count value during a first system clock cycle; the second count value is obtained by counting the system clock cycles of the system clock signal; The time calculator is used to calculate the product of the second count value and the system clock cycle, and calculate the difference between the product and the first time interval to obtain the time of the rising edge of the signal to be measured.

6. A signal processing method for a multiple measurement averaging time-to-digital converter according to any one of claims 1 to 5, characterized in that: The signal processing method of the multiple measurement averaging type time-to-digital converter comprises: Get the sampling latch sequence value corresponding to each system clock rising edge; The sampling latch sequence value corresponding to each rising edge of the system clock is processed to determine the rising edge moment of the signal to be measured.

7. The signal processing method of the multiple measurement averaging type time-to-digital converter according to claim 6, characterized in that: The sampling latch sequence value corresponding to each system clock rising edge is processed to determine the rising edge moment of the signal to be measured, specifically including: Encode the sampling latch sequence value corresponding to each system clock rising edge moment respectively to obtain the code value corresponding to each system clock rising edge moment; Taking the code value corresponding to the first system clock rising edge moment as input, based on the correspondence between the first measurement code value and the delay time interval measured by the code density method, obtaining the measurement result of the first measurement; taking the code value corresponding to the a-th system clock rising edge moment as input, based on the correspondence between the a-th measurement code value and the delay time interval when the first measurement code value measured by the code density method is C, obtaining the measurement result of the a-th measurement; based on the measurement result of the A-th measurement, determining the first time interval between the rising edge moment of the signal to be measured and the first system clock rising edge moment; a=2, 3, ..., A, A is the total number of system clock rising edge moments, and C is the code value corresponding to the first system clock rising edge moment; The product of the second count value outputted in the first system clock cycle and the system clock cycle is calculated, and the difference between the product and the first time interval is calculated to obtain the time of the rising edge of the signal to be measured.

8. The signal processing method of the multiple measurement averaging type time-to-digital converter according to claim 7, characterized in that: When the code density method is used to measure and obtain the correspondence between the a-th measured code value and the delay time interval when the first measured code value is C, when the first measured code value is C, if the extreme difference of multiple code values ​​obtained by the a-th measurement is greater than a preset value, a number of subsequence lengths are added to the target code value, so that the difference between the target code value after the number of subsequence lengths are added and the maximum value among the multiple code values ​​is less than the preset value; wherein the target code value is a code value whose difference with the maximum value among the multiple code values ​​is greater than the preset value.

9. The signal processing method of the multiple measurement averaging type time-to-digital converter according to claim 7, characterized in that: Determining a first time interval between a rising edge of the signal to be measured and a rising edge of the first system clock based on the measurement results of the A measurements specifically includes: Determine the median value of the measurement result of each measurement, and calculate the average value of all medians to obtain a first time interval between the rising edge of the signal to be measured and the rising edge of the first system clock; Alternatively, the maximum value of the lower limit of the measurement result of each measurement is selected as the target lower limit, and the minimum value of the upper limit of the measurement result of each measurement is selected as the target upper limit, and the median of the target lower limit and the target upper limit is determined to obtain the first time interval between the rising edge moment of the signal to be measured and the rising edge moment of the first system clock.

10. A time interval measurement system, characterized in that: The time interval measurement system comprises: a time interval measurement module and a plurality of multiple measurement averaging type time-to-digital converters according to any one of claims 1 to 5; When a multiple measurement average type time digital converter is included, the multiple measurement average type time digital converter is used to respectively determine the time when the rising edge of the signal to be measured starts and the time when the rising edge of the signal to be measured stops; the time interval measurement module is connected to the multiple measurement average type time digital converter, and is used to calculate the difference between the time when the rising edge of the signal to be measured starts and the time when the rising edge of the signal to be measured stops, so as to obtain the second time interval; When two multiple measurement averaging type time digital converters are included, they are respectively recorded as the first time digital converter and the second time digital converter; the first time digital converter is used to determine the moment of starting the rising edge of the signal to be measured; the second time digital converter is used to determine the moment of stopping the rising edge of the signal to be measured; the time interval measurement module is connected to the first time digital converter and the second time digital converter respectively, and is used to calculate the difference between the moment of starting the rising edge of the signal to be measured and the moment of stopping the rising edge of the signal to be measured, so as to obtain the second time interval.

Citation Information

Cited By

  • Anti-crosstalk ignition fault arc detection device and method

    CN120559409A

  • A crosstalk-resistant arc detection device and method for fire faults

    CN120559409B

  • High-precision timing system and method based on dual-channel time measurement

    CN121680025A

  • Carry chain calibration circuit

    CN122509096A