STATISTICAL VALUATION DEVICE, STATISTICAL VALUATION METHOD, AND STATISTICAL VALUATION PROGRAM

Abstract

Statistical value calculation device (100) that calculates a variation in the propagation delay occurring at a measurement object (400) that transmits data, comprising: a unit of variation calculation (110) to calculate a fixed delay that varies each time the object of measurement (400) is activated, based on a multitude of runtime delays measured each time the object of measurement (400) is activated, and to calculate a variation of the fixed delay based on a multitude of fixed delays calculated over a multitude of activations, wherein The variation calculation unit (110) calculates a variable delay that varies during continuous operation on the object of measurement (400), calculates a variation of the variable delay based on a multitude of run-time delays measured on the object of measurement (400) in continuous operation, and calculates the variation of the run-time delay based on the variation of the fixed delay and the variation of the variable delay.

Description

Technical field

[0001] The present disclosure relates to a statistical value calculation device, a statistical value calculation method, and a statistical value calculation program. Background on the state of the art

[0002] In a factory automation (FA) network, time synchronization errors must be prevented across the entire network. A primary cause of time synchronization errors is variation in propagation delay between communication stations and channels. This propagation delay variation is the difference between the maximum and minimum propagation delay values. Therefore, it is desirable to predict this propagation delay variation within an FA network.

[0003] Patent literature 1 discloses a method that repeatedly measures the propagation delay of a test object in continuous operation and estimates the variation of the propagation delay from the distribution of several measured propagation delays. Conventional techniques, including patent literature 1, typically measure the propagation delay of a test object in continuous operation.

[0004] US2010 / 0030516A1 discloses a statistical analysis of propagation delays for integrated circuits, in which a maximum deviation of the propagation delay and a basic propagation delay of the circuit are determined and a propagation delay variance is calculated from these. Reference list patent literature

[0005] Patent Literature 1: JP 2014-022918 A Summary of the invention: Technical problem

[0006] In a FA network, the variation in propagation delay at a measurement object is considered as a variation in fixed delay and a variation in variable delay. A fixed delay is a delay whose value changes with each activation of the measurement object. A variable delay is a delay whose value varies within a constant range, independent of when the measurement object is activated.

[0007] The problem with conventional techniques is that they do not take into account the variation of the fixed delay when evaluating the variation of the propagation delay.

[0008] The present disclosure aims to take into account the variation of the fixed delay when evaluating the variation of the propagation delay at the object under test in an FA network. Solution to the problem

[0009] A statistical value calculation device according to the present disclosure comprises: If a model that specifies that a variation in propagation delay consists of a variation in variable delay and a variation in fixed delay is assumed to be a propagation delay model that specifies the variation in propagation delay, where propagation delay is a delay that occurs at a device under test that transmits data in a wired network, variable delay is a delay that varies while the device under test is in continuous operation, and fixed delay is a delay that varies each time the device under test is activated; and if the activation of the device under test is performed multiple times, and each activation of the device under test results in a plurality of propagation delays being measured at the device under test in continuous operation, the unit of variation calculation to measure the fixed delay based on the multitude of propagation delays measured on the object under test during continuous operation, as an object-fixed delay at each activation of the object under test, and to calculate the variation of the fixed delay based on the multitude of measured multiple object-fixed delays. Advantageous effects of the invention

[0010] According to the present disclosure, the variation calculation unit measures the fixed delay based on several propagation delays measured during continuous operation of the device under test, and calculates the variation of the fixed delay based on several measured fixed delays. Here, the device under test can be a device as a component of a function analysis (FA) network. Furthermore, the calculated variation of the fixed delay can be taken into account when evaluating the variation of the propagation delay. Thus, according to the present disclosure, the variation of the fixed delay can be considered when evaluating the variation of the propagation delay at the device under test in a function analysis (FA) network. Brief description of the drawings Fig. Figure 1 shows a configuration example of a time delay measurement system 90 according to embodiment 1. Fig. Figure 2 is a representation showing runtime delays according to embodiment 1. Fig. Figure 3 shows an example of the hardware configuration of a statistical value calculation device 100 according to embodiment 1. Fig. Figure 4 is a flowchart showing the operation of the transit-time delay measurement system 90 according to embodiment 1. Fig. Figure 5 shows an example of the hardware configuration of the statistical value calculation device 100 according to a modification of embodiment 1. Fig. Figure 6 is a flowchart showing the operation of the transit-time delay measurement system 90 according to embodiment 2. Fig. Figure 7 is a flowchart showing the operation of the transit-time delay measurement system 90 according to a modification of embodiment 2. Fig. Figure 8 is a flowchart showing the operation of the transit-time delay measurement system 90 according to embodiment 3. Fig. Figure 9 is a flowchart showing the operation of the transit-time delay measurement system 90 according to embodiment 4. Fig. Figure 10 shows a configuration example of the time delay measurement system 90 according to embodiment 6. Description of embodiments

[0011] In the embodiments and drawings, the same or corresponding elements are designated with the same reference numerals. Descriptions of elements with the same reference numerals are omitted or simplified where appropriate. The arrows in the drawings primarily indicate a data flow or a processing flow. Furthermore, a "unit" can be understood as a "circuit," "step," "method," "process," or "circuit element," depending on the context. Design 1.

[0012] The present embodiment is described in detail below with reference to the drawings. *** Configuration description ***

[0013] Fig. Figure 1 shows a configuration example of a time-delay measurement system 90 according to the present embodiment. The time-delay measurement system 90 comprises a statistical value calculation device 100, a delay measurement device 200, a frame transmitter 300, a measurement object 400, and a frame receiver 500, as shown in Figure 1. Fig. Figure 1 shows the statistical value calculation device 100 and the delay measuring device 200, which can also be configured as a single unit. The frame transmitter 300, the measurement object 400, and the frame receiver 500 can be components of a factory automation (FA) network.

[0014] The frame transmitter 300 and the device under test 400, as well as the device under test 400 and the frame receiver 500, are connected to each other via a wired network. The wired network is a network that conforms to the Ethernet standard (registered trademark), as a specific example.

[0015] The delay measuring device 200 communicates the difference between a reception time at the frame receiver 500 and a transmission time at the frame transmitter 300 as the propagation delay to the statistical value calculation device 100. The delay measuring device 200 measures the propagation delay based on the difference between the time at which the frame transmitter 300 sends a frame of interest and the time at which the frame receiver 500 receives the frame of interest.

[0016] The frame transmitter 300 repeatedly sends frames to the frame receiver 500 in the wired network and informs the delay measuring device 200 of the transmission times of the frames.

[0017] The device under test 400 is a relay for forwarding data in the wired network and transmits frames between the frame sender 300 and the frame receiver 500. Propagation delays occur at the device under test 400.

[0018] The frame receiver 500 receives the frames transmitted by the frame transmitter 300 in the wired network and reports the reception times of the frames to the delay measuring device 200.

[0019] Propagation delay is a delay that occurs during communication over a wired network. Propagation delay is used, for example, to correct time synchronization errors in a fixed access network. In the Fig. In the time-delay measurement system 90 shown in Figure 1, the time delay is the difference between the time at which the frame sender 300 sends a frame of interest and the time at which the frame receiver 500 receives the frame of interest. The time delay is measured by calculating the difference between the time at which the frame sender 300 sends the frame of interest and the time at which the frame receiver 500 receives the frame of interest, after it has passed through the object of measurement 400, using the delay measuring device 200, as shown in a specific example. The term "time delay" can refer to data that specify a time delay.

[0020] In conventional methods, the value of the propagation delay variation is predicted by measuring the propagation delays when the device under test repeatedly transmits frames (400) in continuous operation, and the propagation delay distribution is estimated based on the measurement results. Accordingly, the fixed delay cannot be measured using conventional techniques.

[0021] The present embodiment uses a propagation delay model to describe the variation of the propagation delay. This propagation delay model specifies that the variation in propagation delay consists of the variation in variable delay and the variation in fixed delay. A variable delay is a delay that varies while the device under test 400 is in continuous operation. A fixed delay is a delay that varies each time the device under test 400 is activated.

[0022] The statistical value calculation device 100 comprises a variation calculation unit 110 and a communication unit 120, as shown in Fig. 1 shown.

[0023] If the activation of the object 400 is performed multiple times, and several propagation delays are measured at the object 400 during continuous operation with each activation, the variation calculation unit 110 measures the fixed delay based on several propagation delays measured at the object 400 during continuous operation as the object-fixed delay for each activation of the object 400. The variation calculation unit 110 then calculates the variation of the fixed delay based on the measured multiple object-fixed delays. The variation calculation unit 110 can perform a process to calculate an average of all multiple propagation delays measured at the object 400 during continuous operation as a process to measure the fixed delay.The variation calculation unit 110 also measures the variable delay based on several propagation delays measured during continuous operation on the test object 400, and calculates the variation of the variable delay based on the result of the variable delay measurement. The variation calculation unit 110 calculates the variation in propagation delay based on the calculated variation in the fixed delay and the calculated variation in the variable delay.

[0024] The processing performed by the variation calculation unit 110 is described in more detail below. The variation calculation unit 110 receives multiple propagation delays from the delay measuring device 200 via the communication unit 120 when the object under test 400 is in continuous operation and performs a delay measurement process based on the received propagation delays. The delay measurement process includes processing to measure the distribution of the propagation delays as a distribution of variable delays based on multiple propagation delays and processing to calculate the mean of all multiple propagation delays as a fixed delay. The mean can be calculated based on any definition.The Variation Computing Unit 110 can perform a process to calculate the maximum and minimum values ​​of the variable delay as a process to measure the distribution of variable delays. In this process, the Variation Computing Unit 110 updates the maximum value of the variable delay if the maximum value calculated from several received runtime delays is greater than the maximum value of the variable delay during the execution of the delay measurement process, and updates the minimum value of the variable delay if the minimum value calculated from several received runtime delays is less than the minimum value of the variable delay during the execution of the delay measurement process. The variable delay can be negative.The variation calculation unit 110 can also perform the delay measurement process using run-time delays, which are stored, for example, in a database containing the run-time delays measured by the delay measuring device 200.

[0025] To measure the distribution of fixed delays, the object 400 is reactivated multiple times. The variation calculation unit 110 performs the delay measurement process each time the object 400 is reactivated. When the object 400 is reactivated, at least one of the following devices can also be reactivated in addition to the object 400 itself: the delay measuring device 200, the frame transmitter 300, the object 400, and the frame receiver 500.

[0026] The Variation Computing Unit 110 then measures the distribution of fixed delays based on several fixed delays calculated by repeatedly executing the delay measurement process. The Variation Computing Unit 110 then determines the sum of the variation obtained from the distribution of variable delays and the variation obtained from the distribution of fixed delays as the variation of the propagation delay. The Variation Computing Unit 110 can also perform processing to calculate each of the maximum and minimum values ​​of the fixed delay as a process to measure the distribution of fixed delays, just as it does when measuring the distribution of variable delays.

[0027] Here, the variation in runtime delay is represented by Formula 1. In Formula 1, "Variation in variable delay" refers to the variation resulting from the distribution of variable delays. "Variation in fixed delay" indicates the variation determined from the distribution of fixed delays.

[0028] "Variation in variable delay" is the maximum value of a range of variable delay, or a value estimated to be the maximum value of the range of variable delay, as specific examples. The range of variable delay is the difference between the minimum and maximum values ​​of the variable delay.

[0029] "Variation in fixed delay" refers to the range of observed fixed delays, or a value estimated as the range of fixed delay, as specific examples. The range of fixed delay is the difference between the minimum and maximum values ​​of the fixed delay.

[0030] The minimum value of the variable delay, the maximum value of the variable delay, the minimum value of the fixed delay, and the maximum value of the fixed delay can each be an actually observed value, a value estimated from actually observed values, or a value theoretically calculated on the basis of the properties of the object of measurement 400, a specification, or the like. (Variation in runtime delay)=(Variation in variable delay)+(Variation in fixed delay)

[0031] Fig. Figure 2 is a table illustrating the variation in propagation delay using a concrete example of the propagation delay measurement results. This example shows the measurement results for the case where the object under test, 400, is activated six times. The table in Fig. Column 2, "Maximum [ns]", shows the maximum propagation delay values ​​in units of ns when the device under test (400) is in continuous operation, and column 2, "Minimum [ns]", shows the minimum propagation delay values ​​in units of ns when the device under test (400) is in continuous operation. Column 3, "Mean [ns]", shows the mean of all observed propagation delays (i.e., the fixed delay) in ns, and column 4, "Maximum - Minimum [ns]", shows the differences between the values ​​in column 4 and column 5 (maximum - minimum [ns]) (i.e., the range of the variable delay) in ns.

[0032] Since the maximum fixed delay value is 640 ns and the minimum fixed delay value is 620 ns in this example, the variation of the fixed delay is 20 ns. Furthermore, since the maximum variable delay range is 35 ns, the variation of the variable delay is also 35 ns. Therefore, in this example, the variation calculation unit 110 receives a value of 55 ns (= 35 ns + 20 ns) from Formula 1 as the variation of the propagation delay.

[0033] In this example, 655 ns in (A) is the largest propagation delay and 605 ns in (B) is the smallest propagation delay. Since the variation of the actually observed propagation delays is 50 ns (= 655 ns - 605 ns), the variation calculation unit 110 appears to be able to calculate a suitable variation of the propagation delay according to Formula 1.

[0034] The communication unit 120 communicates with the delay measuring device 200.

[0035] Fig. Figure 3 shows an example of the hardware configuration of the statistical value calculation device 100 according to the present embodiment. The statistical value calculation device 100 consists of one computer. The statistical value calculation device 100 can include more than one computer.

[0036] As shown in this illustration, the statistical value calculation device 100 is a computer comprising hardware components such as a processor 11, a main memory 12, an auxiliary storage device 13, an input / output interface (IF) 14, and a communication device 15. These hardware components are interconnected via signal lines 19.

[0037] The processor 11 is an IC (integrated circuit) that performs arithmetic operations and controls the hardware within the computer. The processor 11 can be, for example, a CPU (central processing unit), a DSP (digital signal processor), or a GPU (graphics processing unit).

[0038] The statistical value calculation device 100 can also contain multiple processors that replace processor 11. The multiple processors share the roles of processor 11.

[0039] The main memory 12 is typically a volatile storage device, and a specific example is RAM (Random Access Memory). The main memory 12 is also referred to as a primary storage device or main working memory. The data stored in the main memory 12 is stored in the auxiliary storage device 13 as needed.

[0040] The auxiliary storage device 13 is typically a non-volatile storage device and is, for example, a ROM (Read Only Memory), an HDD (Hard Disk Drive), or flash memory. Data stored in the auxiliary storage device 13 is loaded into the main memory 12 as needed.

[0041] The main memory 12 and the auxiliary storage device 13 can be configured as a single unit.

[0042] Input / Output IF 14 is a connector to which input and output devices are connected. A USB (Universal Serial Bus) connector is a concrete example of an input / output device. Specific examples of input devices are a keyboard and a mouse. A display is a concrete example of an output device.

[0043] The communication device 15 is a receiver and a transmitter. The communication device 15 is, for example, a communication chip or a network interface card (NIC).

[0044] The components of the statistical value calculation device 100 can use the input / output IF 14 and the communication device 15 in a suitable manner when communicating with other devices or the like.

[0045] The auxiliary storage device 13 stores a statistical value calculation program. The statistical value calculation program is a program that causes a computer to execute the functions of the components provided by the statistical value calculation device 100. The statistical value calculation program is loaded into the main memory 12 and is executed by the processor 11. The functions of the components contained in the statistical value calculation device 100 are implemented by software.

[0046] Data for use in the execution of the statistical value calculation program, data resulting from the execution of the statistical value calculation program, and the like are stored in a storage device as needed. The components of the statistical value calculation device 100 utilize the storage device as required. The storage device comprises at least one of the following elements: the main memory 12, the auxiliary storage device 13, a register in the processor 11, and a cache memory in the processor 11. The storage device may be independent of the computer.

[0047] The functions of the main memory 12 and the auxiliary storage device 13 can be implemented by another storage device.

[0048] The statistical calculation program can be stored on a computer-readable, non-volatile recording medium. This non-volatile recording medium could be, for example, an optical disc or flash memory. The statistical calculation program can be provided as a software product. ***Description of Functioning***

[0049] An operating procedure of the statistical value calculation device 100 corresponds to the statistical value calculation procedure. A program for activating the functions of the statistical value calculation device 100 corresponds to the statistical value calculation program.

[0050] Fig. Figure 4 is a flowchart illustrating an example of the operation of the propagation delay measurement system 90. The functionality of the propagation delay measurement system 90 is explained using the following: Fig. 4 described. (Step S101)

[0051] The variation calculation unit 110 measures the distribution of propagation delays at the test object 400 during continuous operation as a distribution of variable delays and calculates a mean value M1 as the fixed delay. The mean value M1 is the average of all propagation delays at the test object 400 during continuous operation and represents the fixed delay. (Step S102)

[0052] If the distribution of the fixed delays can be determined because the number of mean values ​​M1 calculated in step S101 is sufficient, step S104 is executed. Otherwise, step S103 is executed. (Step S103)

[0053] The measuring object 400 is reactivated. (Step S104)

[0054] The variation calculation unit 110 calculates the variation of the run-time delay according to formula 1. That is, the variation calculation unit 110 determines the sum of the variation of the variable delay obtained from the distribution of the measured variable delays and the variation of the fixed delay obtained from the resulting distribution of the fixed delays as the variation of the run-time delay. ***Description of the effects of embodiment 1***

[0055] As described above, according to the present embodiment, the variation in transit time delay that may occur on the object 400 can be calculated more accurately by calculating the variation in transit time delay based on the variation in variable delay and the variation in fixed delay. ***Other configurations***<Modifikation 1>

[0056] Fig. Figure 5 shows an example of the hardware configuration of the statistical value calculation device 100 according to this modification.

[0057] The statistical value calculation device 100 comprises a processing circuit 18 in place of the processor 11, the processor 11 and the main memory 12, the processor 11 and the auxiliary storage device 13, or the processor 11, the main memory 12 and the auxiliary storage device 13.

[0058] The processing circuit 18 is a piece of hardware that implements at least some of the components contained in the statistical value calculation device 100.

[0059] The processing circuit 18 can be dedicated hardware or can be a processor that executes programs stored in the main memory 12.

[0060] If the processing circuit 18 is dedicated hardware, the processing circuit 18 can be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (application-specific integrated circuit), an FPGA (field-programmable gate array), or a combination thereof.

[0061] The statistical value calculation device 100 can contain several processing circuits to replace the processing circuit 18. The multiple processing circuits share the tasks of the processing circuit 18.

[0062] In the statistical value calculation device 100, some of the functions can be implemented by dedicated hardware and the remaining functions can be implemented by software or firmware.

[0063] The processing circuit 18 is implemented, for example, by hardware, software, firmware or a combination thereof.

[0064] The processor 11, the memory 12, the auxiliary storage device 13, and the processing circuit 18 are collectively referred to as the "processing circuit." That is, the functions of the functional components of the statistical value calculation device 100 are implemented by the processing circuit.

[0065] The statistical value calculation devices 100 according to other embodiments may have a similar configuration to this modification. Design 2.

[0066] Differences from the embodiment described above are explained below, mainly with reference to drawings.

[0067] In the present embodiment, the variation of the variable delay and the deviation of the fixed delay are calculated from the maximum and minimum values ​​of the results of a predetermined number of measurements of the transit time delays. *** Configuration description ***

[0068] The configuration of the transit-time delay measurement system 90 according to the present embodiment is similar to the configuration of the transit-time delay measurement system 90 according to embodiment 1.

[0069] The variation calculation unit 110 according to the present embodiment calculates the maximum and minimum values ​​of the fixed deceleration based on the results of the fixed deceleration measurements and calculates the variation of the fixed deceleration based on the calculated maximum and minimum values ​​of the fixed deceleration. The variation calculation unit 110 also calculates the maximum and minimum values ​​of the variable deceleration based on the results of the variable deceleration measurements and calculates the variation of the variable deceleration based on the calculated maximum and minimum values ​​of the variable deceleration.

[0070] The object 400, according to the present embodiment, is reactivated for a first number of measurements. During continuous operation, a second number of measurements are taken on the object 400 to determine the runtime delay.

[0071] Specifically, if the runtime delay is reported N1 times during the continuous operation of the measured object 400 (where N1 is a natural number), the computational analysis unit 110 records the mean M1 of all reported runtime delays, as well as a maximum value MAX1 and a minimum value MIN1 calculated from the reported runtime delays. The measured object 400 is then reactivated. After this cycle has been repeated N2 times (where N2 is a natural number), the computational analysis unit 110 records a maximum value MAX2 and a minimum value MIN2 from the recorded mean values ​​M1, which serve as samples. Based on these results, the variation of the runtime delay is calculated using formula 2. Here, N1 and N2 each denote the number of measurements and can be defined arbitrarily. The value N1 represents the number of the first measurements. The value N2 represents the number of the second measurements.The values ​​of N1 and N2 can differ from each other. The mean value M1 represents the mean of all propagation delays observed during continuous operation of the device under test 400. The maximum value MAX1 represents the maximum of all variable delays observed during continuous operation of the device under test 400. The minimum value MIN1 represents the minimum of all variable delays observed during continuous operation of the device under test 400. The maximum value MAX2 represents the maximum of all calculated mean values ​​M1. The minimum value MIN2 represents the minimum of all calculated mean values ​​M1. (Maximum value MAX2 - Minimum value MIN2) represents the variation of the fixed delay. (Maximum value MAX1 - Minimum value MIN1) represents the variation of the variable delay. (Variation in runtime delay) = Maximum value MAX2 − Minimum value MIN2 + Maximum value MAX1 − Minimum value MIN1 ***Description of Functioning***

[0072] Fig. Figure 6 is a flowchart illustrating an example of the operation of the propagation delay measurement system 90. The functionality of the propagation delay measurement system 90 is explained using the following: Fig. 6 described. (Step S111)

[0073] If the variation calculation unit 110 has received runtime delays after the activation of the measurement object 400 N1, step S112 is executed. Otherwise, step S111 is executed again. (Step S112)

[0074] The variation calculation unit 110 calculates the mean value M1, the maximum value MAX1 and the minimum value MIN1 from the received N1 runtime delays and stores the calculated mean values ​​M1, maximum values ​​MAX1 and minimum values ​​MIN1. (Step S113)

[0075] If the variation calculation unit has recorded 110 N2 means M1, step S114 is executed. Otherwise, step S103 is executed. (Step S114)

[0076] The variation calculation unit 110 calculates the maximum value MAX2 and the minimum value MIN2 based on the recorded N2 means M1 and stores the calculated maximum value MAX2 and the minimum value MIN2.

[0077] The variation calculation unit 110 then calculates the variation of the runtime delay according to formula 2 with the recorded maximum value MAX2, the minimum value MIN2, the maximum value MAX1 and the minimum value MIN1. ***Description of the effects of embodiment 2***

[0078] As described above, according to the present embodiment even a simple statistical value calculation device 100, which is not able to calculate standard deviations, can calculate the variation of the transit time delay on the object 400 being measured.

[0079] Furthermore, by simply reactivating the object of measurement 400, the fixed delays of the measurement errors become constant; that is, the fixed delays of the measurement errors contained in the maximum value MAX2 and the minimum value MIN2 are equal. Thus, the fixed delays of the measurement errors can be eliminated by determining the difference between the maximum value MAX2 and the minimum value MIN2, which allows for a more accurate estimation of the variation in the propagation delay. A measurement error is an error contained in the propagation delay and originates in a measuring instrument. A measuring instrument is a device for measuring propagation delays, consisting, as a concrete example, of the delay measuring device 200, the frame transmitter 300, and the frame receiver 500. ***Other configurations***<Modifikation 2>

[0080] In this modification, each of the first number of measurements and the second number of measurements is a number defined as a function of a target measurement error in the variation of the runtime delay.

[0081] If N measurements are obtained for each of N1 and N2, the standard deviation of the samples consisting of the obtained measurements is ((N-1) / N)^0.5 times the population standard deviation, and the difference between the maximum and minimum values ​​is also ((N-1) / N)^0.5 times the population difference. Here, the population standard deviation represents a theoretical value. Thus, this modification defines the value of N1 and N2, respectively, as a value of N that satisfies Formula 3 when a target error for the variation in runtime delay is ±M% (where M ≥ 0). If a significant digit i (where i is a natural number) is required, the value of N1 and N2 is each set to a value of N that satisfies Formula 4. ((N−1) / N)0.5≥1−M / 100 ((N−1) / N)0.5>1−0.5*10(1−i)

[0082] Fig. Figure 7 is a flowchart illustrating an example of the operation of the propagation delay measurement system 90 according to this modification. The functionality of the propagation delay measurement system 90 is explained using the following: Fig. 7 described. (Step S121)

[0083] If M% is entered as the target error for the runtime delay variation, the variation calculation unit 110 sets the value of N1 and N2 each to a value of N that satisfies formula 3.

[0084] According to this modification, a variation of the runtime delay is calculated that satisfies the target error M%. Design 3.

[0085] Differences from the embodiment described above are explained below, mainly with reference to drawings.

[0086] The present embodiment is an application example of embodiment 2. *** Configuration description ***

[0087] The configuration of the transit-time delay measurement system 90 according to the present embodiment is similar to the configuration of the transit-time delay measurement system 90 according to embodiment 1.

[0088] The variation calculation unit 110, according to the present embodiment, calculates a first standard deviation based on the result of the variable delay measurement when the device 400 is in continuous operation. When the first standard deviation converges with a first accuracy, the device 400 is reactivated. The variation calculation unit 110 also calculates a second standard deviation based on the result of the fixed delay measurement, and when the calculated second standard deviation converges with a second accuracy, it calculates the variation of the fixed delay. The first accuracy and the second accuracy can each be defined arbitrarily.

[0089] Specifically, the computational analysis unit 110 measures a standard deviation S1 at each measurement of the variable delay and a standard deviation S2 at each measurement of the fixed delay. Standard deviation S1 represents the variation of the variable delay and is equal to the first standard deviation. Standard deviation S2 represents the variation of the fixed delay and is equal to the second standard deviation. The computational analysis unit 110 determines that the respective errors of standard deviation S1 and standard deviation S2 with respect to the population standard deviation can be ignored at a point where standard deviation S1 and standard deviation S2 converge with the required accuracy, and terminates the measurement regardless of the values ​​of N1 and N2. Here, standard deviation S1 is the standard deviation of the variable delay.The standard deviation S2 is the standard deviation of the fixed delay. The required accuracies can be defined arbitrarily. The required accuracy according to standard deviation S1 and the required accuracy according to standard deviation S2 can differ from each other. ***Description of Functioning***

[0090] Fig. Figure 8 is a flowchart illustrating an example of the operation of the propagation delay measurement system 90 according to the present embodiment. The operation of the propagation delay measurement system 90 is explained by reference to Fig. 8 described. (Step S131)

[0091] The variation calculation unit 110 calculates the standard deviation S1 using all runtime delays received after the activation of the measurement object 400. (Step S132)

[0092] If the standard deviation S1 converges with the required accuracy, step S112 is executed. Otherwise, step S131 is executed. (Step S133)

[0093] The variation calculation unit 110 calculates the standard deviation S2 using all recorded means M1. (Step S134)

[0094] If the standard deviation S2 converges with the required accuracy, step S114 is executed. Otherwise, step S103 is executed. ***Description of the effects of embodiment 3***

[0095] According to the present embodiment, it is expected that a variation of the runtime delay will be calculated that satisfies a predetermined target accuracy. Design 4.

[0096] Differences from the embodiment described above are explained below, mainly with reference to drawings.

[0097] The present embodiment is an example of calculating the variation of the propagation delay based on a theoretical standard deviation when the distribution of propagation delays can be predicted. *** Configuration description ***

[0098] The configuration of the transit-time delay measurement system 90 according to the present embodiment is similar to the configuration of the transit-time delay measurement system 90 according to embodiment 1.

[0099] According to the present embodiment, the variation calculation unit 110 calculates a third standard deviation based on a result of the variable delay measurement, a fourth standard deviation based on a result of the fixed delay measurement, and the variation in runtime delay based on the calculated third and fourth standard deviations. In this process, the variation calculation unit 110 calculates the variation in runtime delay based on the third standard deviation and a distribution that, as a specific example, is likely to be followed by the variable delay. The variation calculation unit 110 also calculates the variation in runtime delay based on the fourth standard deviation and a distribution that the fixed delay is likely to follow.The distribution likely to be followed by each of the variable delays and the fixed delay is a uniform distribution or a normal distribution, to give specific examples. The standard deviation S1 according to the present embodiment corresponds to the third standard deviation. The standard deviation S2 according to the present embodiment corresponds to the fourth standard deviation.

[0100] In particular, the variation calculation unit 110 can calculate the variation in propagation delay using the measured standard deviation S1 and the standard deviation S2, provided the propagation delay distribution can be predicted. A concrete example: In a synchronization circuit with a common clock, much of the variation in propagation delay results from the transition between different clock domains. Therefore, the propagation delay distribution is expected to be a uniform distribution or a superposition of uniform distributions.

[0101] Here is the standard deviation of a uniform distribution as given in Formula 5. If the standard deviations S1 and S2 have each converged with the required accuracy, it is assumed that they are sufficiently close to their theoretical values. Since both S1 and S2 are sufficiently close to their theoretical values, Formula 5 yields the variation of the variable delay and the variation of the fixed delay in the population, as shown in Formula 6. Because a uniform distribution is such that the distribution extends symmetrically on both sides of the mean, the maximum and minimum values ​​of the runtime delay are given in Formula 7.Therefore, if the distribution of variable delays and the distribution of fixed delays are both uniform, the variation in runtime delay is as shown in formula 8. Here, the mean M2 denotes the mean value in the distribution of runtime delays.

[0102] The variation calculation unit 110 can also calculate the maximum and minimum values ​​of the runtime delay according to formula 9. Here, Max(...) denotes the maximum value in the parentheses and Min(...) the minimum value in the parentheses. (Standard deviation of a uniform distribution) = ((Maximum value) − (Minimum value)) / 2√3 (Maximum value of the variable delay)−(Minimum value of the variable delay)=Standard deviation S1*2√3 (Maximum value of the fixed delay) − (Minimum value of the fixed delay) = Standard deviation S2 * 2√3 (Maximum value of the runtime delay) = Mean M2 + Standard deviation S1*√3 + Standard deviation S2*√3 (Minimum value of the runtime delay) = Mean M2 + Standard deviation S1*√3 + Standard deviation S2*√3 (Variation in runtime delay) = (Maximum value of runtime delay) − (Minimum value of runtime delay) = (Standard deviation S1 + Standard deviation S2) × 2√3 (Maximum value of the runtime delay) = Average M2 + Max(Maximum value MAX1 - Average M1) + Max(Maximum value MAX2 - Average M2) = Maximum value MAX2 + Max(Maximum value MAX1 - Average M1) (Minimum value of the runtime delay) = Average M2 + Min(Minimum value MIN1 - Average M1) + Min(Minimum value MIN2 - Average M2) = Minimum value MIN2 + Min(Minimum value MIN1 - Average M1) ***Description of Functioning***

[0103] Fig. Figure 9 is a flowchart illustrating an example of the operation of the propagation delay measurement system 90 according to the present embodiment. The operation of the propagation delay measurement system 90 is explained by means of Fig. 9 described. (Step S141)

[0104] The variation calculation unit 110 records the calculated standard deviation S1. (Step S142)

[0105] The variation calculation unit 110 calculates the variation of the runtime delay according to formula 8 using the calculated standard deviation S1 and the standard deviation S2. ***Description of the effects of embodiment 4***

[0106] According to the present embodiment, the variation of the propagation delay is expected to be calculated with higher accuracy than when calculating the variation of the propagation delay using maximum and minimum values ​​based on measurement results. ***Other configurations***<Modifikation 3>

[0107] It can be assumed that in embodiment 4, only either the fixed delay or the variable delay follows a uniform distribution. Assuming that only the fixed delay follows a uniform distribution, the maximum and minimum values ​​of the propagation delay are each given as in formula 10. Assuming that only the variable delay follows a uniform distribution, the maximum and minimum values ​​of the propagation delay are each given as in formula 11.

[0108] The variation calculation unit 110 according to this modification calculates the variation of the runtime delay according to formula 10 or formula 11. (Maximum value of the runtime delay) = Mean M2 + Max(Maximum value MAX1 + Mean M1) + Standard deviation S2*√3 (Minimum value of the runtime delay) = Mean M2 + Min(Minimum value MIN1 + Mean M1) − Standard deviation S2*√3 (Maximum value of the runtime delay) = Mean M2 + Standard deviation S1*√3 + Max(Maximum value MAX2 - Mean M2) = Maximum value MAX2 + Standard deviation S1*√3 (Minimum value of the runtime delay) = Mean M2 + Standard deviation S1*√3 + Min(Minimum value MIN2 - Mean M2) = Minimum value MIN2 + Standard deviation S1*√3 Design 5.

[0109] Differences from the embodiment described above are explained below, mainly with reference to drawings.

[0110] The present embodiment is an example of calculating the variation of the propagation delay based on theoretical standard deviations when the propagation delay distribution cannot be predicted. *** Configuration description ***

[0111] The configuration of the transit-time delay measurement system 90 according to the present embodiment is similar to the configuration of the transit-time delay measurement system 90 according to embodiment 1.

[0112] The variation calculation unit 110 according to the present embodiment assumes that the distribution likely to be followed by the fixed delay is a normal distribution, and assumes that the distribution likely to be followed by the variable delay is also a normal distribution. In particular, if the distribution of the runtime delays cannot be predicted, the variation calculation unit 110, as a concrete example, assumes a normal distribution as the distribution of the runtime delays and calculates the variation of the runtime delay using a normal distribution table.

[0113] Here, a situation must be created in which variable delays equal to or less than the minimum value MIN1 and variable delays equal to or greater than the maximum value MAX1 are unlikely to occur in the communications carried out by the device under test 400 during its lifetime. That is, it is required that the product of the number of communications carried out by the device under test 400 during its lifetime and the sum of the probability of a variable delay equal to or less than the minimum value MIN1 occurring in a communication and the probability of a variable delay equal to or greater than the maximum value MAX1 occurring in a communication is 0.5 or less, as shown in Formula 12-1.Here, due to the symmetry of a normal distribution, the probability of a variable delay occurring equal to or greater than the maximum value MAX1 is equal to the probability of a variable delay occurring equal to or less than the minimum value MIN1. Therefore, Formula 12-1 is essentially equivalent to Formula 12-2. Furthermore, rearranging Formula 12-2 yields Formula 13. The Calculation of Variation Unit 110 calculates the minimum value MIN1 by substituting the mean M1 and the standard deviation S1 into a cumulative distribution function that satisfies Formula 13 and solving it. The Calculation of Variation Unit 110 then calculates the maximum value MAX1 according to Formula 17. If both the variable delay and the fixed delay follow a normal distribution, Formula 16 applies. Rearranging Formula 16 yields Formula 17.

[0114] For the fixed delay, a situation must also be created in which fixed delays equal to or less than the minimum value MIN2 and fixed delays equal to or greater than the maximum value MAX2 are unlikely to occur, as is the case with the variable delay. Thus, Formula 14 must be satisfied. Furthermore, rearranging Formula 14 yields Formula 15. The Calculation of Variation Unit 110 calculates the minimum value MIN2 by substituting the mean M3 and the standard deviation S2 into a cumulative distribution function that satisfies Formula 15 and solving it. The Calculation of Variation Unit 110 then calculates the maximum value MAX2 according to Formula 17. Here, the mean M3 denotes the mean of all fixed delays.

[0115] The values ​​to be inserted into the cumulative distribution functions can be calculated in any way. {(Probability that a variable delay occurs which is equal to or less than the minimum value MIN1) + (Probability that a variable delay occurs which is equal to or greater than the maximum value MAX1)} * (Number of communications carried out by the object of measurement during its lifetime) ≤ 0.5 (Probability of a variable delay occurring that is equal to or less than the minimum value MIN1)*(the number of communications carried out by the object 400 during its lifetime)*2≤0.5 (Probability of a variable delay occurring that is equal to or less than the minimum value MIN1) ≤ 0.5 / (Number of communications carried out by the object 400 during its lifetime) / 2 (Probability that a fixed delay occurs which is equal to or less than the minimum value MIN2)*(the number of activations of the measured object 400 during its lifespan)*2≤0.5 (Probability that a fixed delay occurs which is equal to or less than the minimum value MIN2)≤0.5(number of activations of the measured object 400 during its lifespan) / 2 Average M1 = (Maximum value MAX1 + Minimum value MIN1) / 2 Average M3 = (Maximum value MAX2 + Minimum value MIN2) / 2 Maximum value MAX1 = Average value M1 − Minimum value MIN1 Maximum value MAX2 = 2 ∗ Average value M3 − Minimum value MIN2

[0116] After calculating the maximum and minimum values ​​corresponding to the variable delay and the maximum and minimum values ​​corresponding to the fixed delay as described above, the variation calculation unit 110 calculates the variation of the runtime delay according to formula 18. (Variation of runtime delay) = (Difference between the maximum and minimum values ​​corresponding to the variable delay) + (Difference between the maximum and minimum values ​​corresponding to the fixed delay) ***Description of Functioning***

[0117] The operation of the transit-time delay measurement system 90 according to the present embodiment is similar to the operation of the transit-time delay measurement system 90 according to embodiment 4. Differences between the present embodiment and embodiment 4 are explained below. (Step S142)

[0118] The variation calculation unit 110 calculates the variation of the runtime delay according to formula 18. ***Description of the effects of embodiment 5***

[0119] According to the present embodiment, the variation of the propagation delay is expected to be calculated with higher accuracy than when calculating the variation of the propagation delay using maximum and minimum values ​​based on measurement results. Design 6.

[0120] Differences from the embodiment described above are explained below, mainly with reference to drawings. *** Configuration description ***

[0121] The present embodiment uses the transit-time delay measurement system 90 according to embodiment 1 and a transit-time delay measurement system 91 as a transit-time delay measurement system. The term "transit-time delay measurement system" is also a generic term for the transit-time delay measurement system 90 and the transit-time delay measurement system 91.

[0122] The time-delay measurement system 91 comprises the statistical value calculation device 100, the delay measurement device 200, the frame transmitter 300 and the frame receiver 500, as shown in Fig. 10 shown.

[0123] The propagation delay measurement system 90 is also referred to as the first system. The propagation delay measurement system 91 is also referred to as the second system.

[0124] The propagation delay model used to describe the variation in the propagation delay of the first system is one that states that the variation in the propagation delay of the first system consists of a variation in a first variable delay and a variation in a first fixed delay. The first variable delay is a delay that varies when all components of the first system are in continuous operation. The first fixed delay is a delay that varies each time all components of the first system are activated.

[0125] The propagation delay model used to describe the variation in the propagation delay of the second system is one that specifies that the variation in the propagation delay in the second system consists of a variation in a second variable delay and a variation in a second fixed delay. The second variable delay is a delay that varies when all components of the second system are in continuous operation. The second fixed delay is a delay that varies each time all components of the second system are activated.

[0126] If the activation of all components of the first system is performed multiple times, and multiple propagation delays are measured in the first system during each activation of all components of the first system in continuous operation, the variation computation unit 110, according to the present embodiment, measures the first fixed delay based on multiple propagation delays measured in the first system during each activation of all components of the first system. The variation computation unit 110 also measures the first variable delay based on multiple propagation delays measured in the first system in continuous operation.The variation calculation unit 110 then calculates a first fixed standard deviation as a standard deviation based on the result of the measurement of the first fixed delay and calculates a first variable standard deviation as a standard deviation based on the result of the measurement of the first variable delay.

[0127] If the activation of all components of the second system is performed multiple times, and multiple runtime delays are measured in the second system during each activation of all components of the second system in continuous operation, the variation computational unit 110 measures the second fixed delay based on multiple runtime delays measured in the second system during each activation of all components of the second system. The variation computational unit 110 also measures the second variable delay based on multiple runtime delays measured in the second system in continuous operation.The variation calculation unit 110 then calculates a second fixed standard deviation as a standard deviation based on the result of the measurement of the second fixed delay and calculates a second variable standard deviation as a standard deviation based on the result of the measurement of the second variable delay.

[0128] The variation calculation unit 110 then calculates the third standard deviation based on the first variable standard deviation and the second variable standard deviation, and calculates the fourth standard deviation based on the first fixed standard deviation and the second fixed standard deviation.

[0129] In the present embodiment, all components of the propagation delay measurement system are reactivated for measuring fixed delays.

[0130] In the present embodiment, to calculate the variation of the transit-time delay at the object 400 while eliminating the measurement errors of the measuring instrument, the standard deviation is measured in each of the transit-time delay measurement systems 90 and 91. The variation of the transit-time delay at the object 400 is then calculated using the additivity of the distribution. In particular, the variation calculation unit 110 calculates the standard deviations according to formula 19 and uses the calculated standard deviations to calculate the variation of the transit-time delay at the object 400. Here, "a configuration with the object 400" corresponds to transit-time delay measurement system 90, and "a configuration without the object 400" corresponds to transit-time delay measurement system 91.

[0131] The variation calculation unit 110 can also calculate the mean of the delays according to formula 20 and use the calculated mean to calculate the variation of the runtime delay on the measurement object 400. (Standard deviation) = {(Standard deviation in a configuration with the object being measured 400)² − (Standard deviation in a configuration without the object being measured 400)²} 0.5 (Mean) = (Mean in the configuration with the measurement object 400) − (Mean in the configuration without the measurement object 400) ***Description of Functioning***

[0132] The operation of the transit-time delay measurement system 90 according to the present embodiment is similar to the operation of the transit-time delay measurement system 90 according to embodiment 4. Differences between the present embodiment and embodiment 4 are explained below. (Step S103)

[0133] All components of the propagation delay measurement system are reactivated. (Step S142)

[0134] This step is performed when the standard deviation S1 and the standard deviation S2 in the time-delay measurement system 90 have converged with the required accuracies, and also when the standard deviation S1 and the standard deviation S2 in the time-delay measurement system 91 have converged with the required accuracies.

[0135] The variation calculation unit 110 calculates the standard deviation S1 and the standard deviation S2 corresponding to the measurement object 400 according to formula 19 using the standard deviation S1 and the standard deviation S2 in the transit-time delay measurement system 90 and the standard deviation S1 and the standard deviation S2 in the transit-time delay measurement system 91. The variation calculation unit 110 then calculates the variation of the transit-time delay on the measurement object 400 using the calculated standard deviation S1 and the standard deviation S2 corresponding to the measurement object 400 in the diagrams shown in embodiment 4 or embodiment 5. ***Description of the effects of embodiment 6***

[0136] According to the present embodiment, the variation of the propagation delay is expected to be calculated with higher accuracy than when calculating the variation of the propagation delay using maximum and minimum values ​​based on measurement results. *** Other versions ***

[0137] Any combination of the embodiments described above, any modification of a component in the embodiments, or any omission of a component in the embodiments is possible.

[0138] The embodiments are not limited to those specified in embodiments 1 to 6; various modifications can be made as needed. The procedures described with flowcharts and the like can be modified as required. Reference symbol list

[0139] 11: Processor, 12: Main memory, 13: Auxiliary storage device, 14: Input / output interface, 15: Communication device, 18: Processing circuit, 19: Signal line, 90, 91: Time-delay measurement system, 100: Statistical calculation device, 110: Variation calculation unit, 120: Communication unit, 200: Delay measurement device, 300: Frame transmitter, 400: Object under test, 500: Frame receiver

Claims

Statistical value calculation device (100) that calculates a variation in the propagation delay occurring at a measurement object (400) transmitting data, comprising: a variation calculation unit (110) for calculating a fixed delay that varies each time the measurement object (400) is activated, based on a plurality of propagation delays measured at each activation of the measurement object (400), and for calculating a variation in the fixed delay based on a plurality of fixed delays calculated at a plurality of activations, wherein the variation calculation unit (110) calculates a variable delay that varies during continuous operation at the measurement object (400).a variation of the variable delay is calculated based on a multitude of transit-time delays measured on the object (400) in continuous operation, and the variation of the transit-time delay is calculated based on the variation of the fixed delay and the variation of the variable delay. Statistical value calculation device (100) according to claim 1, wherein the variation calculation unit (110) determines a mean value of a plurality of run-time delays measured on the object (400) during continuous operation as the fixed delay. Statistical value calculation device (100) according to claim 1, wherein the variation calculation unit (110) calculates a minimum value and a maximum value of the fixed delay based on the calculation result of the fixed delay, calculates the variation of the fixed delay based on the minimum value and the maximum value of the fixed delay, calculates a minimum value and a maximum value of the variable delay based on a calculation result of the variable delay, and calculates the variation of the variable delay based on the minimum value and the maximum value of the variable delay. Statistical value calculation device (100) according to one of claims 1 to 3, wherein in continuous operation the object being measured (400) is reactivated for a first number of measurements, and the run-time delay for a second number of measurements on the object being measured (400) is measured. Statistical value calculation device (100) according to claim 4, wherein each of the first number of measurements and the second number of measurements is a number which is defined as a function of a target measurement error in the variation of the runtime delay. Statistical value calculation device (100) according to claim 1, wherein the variation calculation unit (110) calculates a standard deviation as a first standard deviation based on a result of the calculation of the variable delay when the object being measured (400) is in continuous operation, when the first standard deviation has converged with a first accuracy, the object being measured (400) is reactivated and the variation calculation unit (110) calculates a standard deviation as a second standard deviation based on a result of the calculation of the fixed delay, and when the calculated second standard deviation has converged with a second accuracy, it calculates the variation of the run-time delay using maximum and minimum values ​​of the fixed delay and the variable delay. Statistical value calculation device (100) according to claim 6, wherein the variation calculation unit (110) calculates a standard deviation as the third standard deviation based on a result of the calculation of the variable delay, calculates a standard deviation as the fourth standard deviation based on a result of the calculation of the fixed delay, and calculates the variation of the run-time delay based on the calculated third standard deviation and the calculated fourth standard deviation. Statistical value calculation device (100) according to claim 7, wherein the variation calculation unit (110) calculates the variation of the run-time delay based on the third standard deviation and a distribution which the variable delay is likely to follow, and calculates the variation of the run-time delay based on the fourth standard deviation and a distribution which is likely to be followed by the fixed delay. Statistical value calculation device (100) according to claim 8, wherein the variation calculation unit (110) assumes that the distribution which the fixed delay is likely to follow is a normal distribution, and that the distribution which the variable delay is likely to follow is a normal distribution. Time delay measurement system (90, 91) comprising two statistical value calculation devices (100) according to claim 7, wherein the variation calculation unit (110) calculates the third standard deviation on the basis of two first variable standard deviations calculated by the two statistical value calculation devices (100) and calculates the fourth standard deviation on the basis of the two second standard deviations calculated by the two statistical value calculation devices (100). Statistical value calculation method, which calculates a variation in the propagation delay that occurs at a measurement object (400) that transmits data, comprising: by a computer, calculating a fixed delay that varies each time the measurement object (400) is activated, based on a plurality of propagation delays measured at each activation of the measurement object (400); and by the computer, calculating a variation in the fixed delay based on a plurality of fixed delays calculated at a plurality of activations, wherein the computer calculates a variable delay that varies during continuous operation of the measurement object (400); and a variation in the variable delay based on a plurality of propagation delays measured at the measurement object (400) in continuous operation.and the variation in runtime delay is calculated based on the variation in fixed delay and the variation in variable delay. A statistical value calculation program that calculates a variation in the propagation delay occurring at a measurement object (400), which forwards data, and which causes a computer, which is a statistical value calculation device (100), to: perform a variation calculation process, wherein the fixed delay, which varies each time the measurement object (400) is activated, is calculated based on a plurality of propagation delays measured at each activation of the measurement object (400), and a variation in the fixed delay is calculated based on a plurality of fixed delays calculated at a plurality of activations, wherein the variation calculation process calculates a variable delay that varies during continuous operation of the measurement object (400).a variation of the variable delay is calculated based on a multitude of transit-time delays measured on the object (400) in continuous operation, and the variation of the transit-time delay is calculated based on the variation of the fixed delay and the variation of the variable delay.

Images