Noise-corrected jitter measurement apparatus and method

Abstract

The test and measurement instrument has an input for receiving a test waveform from a device under test (DUT) having a plurality of input level transitions, a selector configured to individually extract only those portions of the test waveform that match two or more predetermined patterns of input level transitions of the test waveform, a noise compensation unit configured to individually determine and remove, for each extracted portion of the waveform, a component of a jitter measurement due to random noise of the test and measurement instrument receiving the test waveform, a summer configured to generate a composite distribution of timing measurements from the extracted portions of the test waveform with the noise components removed, and a jitter processor configured to determine a first noise compensated jitter measurement for the DUT from the composite distribution. A method for determining the noise compensated jitter measurement is also disclosed.

Description

[Technical Field]

[0001] This disclosure relates to systems and methods related to test and measurement systems, and more particularly to a test and measurement device capable of accurately quantifying the jitter component of an input signal. [Background technology]

[0002] Many modern electronic devices and communication systems use serialized streams of digital data bits to transfer digital information from transmitter to receiver over a channel. Measuring the quality of the transmitted or received signal is of great interest to users in order to predict the error rate. In particular, jitter analysis refers to the process of measuring jitter, which is the temporal displacement of the rising or falling waveform edges from their ideal position, and then analyzing this jitter to identify individual subcomponents, for the purpose of predicting bit error rates and developing and debugging electronic circuits. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] U.S. Patent No. 8594169 [Patent Document 2] U.S. Patent No. 9,294,237 [Patent Document 3] U.S. Patent No. 8891602 [Patent Document 4] U.S. Patent No. 7516030 [Overview of the project] [Problems that the invention aims to solve]

[0004] One problem in analyzing input signal jitter is that the measuring device introduces noise as part of the input signal processing before the jitter analysis of the input signal itself can be performed. This noise affects the apparent position of the waveform transitions that determine the jitter. Therefore, analyzing the jitter of the processed input signal does not isolate the device under test (DUT) because the jitter from the DUT and the noise from the measuring device are combined. Furthermore, it is difficult to remove the noise from the measuring device because the noise from the measuring device biases the jitter measured at each transition, and the bias itself correlates with the slew rate of specific edges of a particular symbol sequence in the input signal. In short, because the noise from the measuring device affects each transition differently, it is not possible to simply subtract the noise from the combined noise to analyze the jitter of the input signal.

[0005] Embodiments of the disclosed technology address these and other shortcomings of the prior art.

[0006] The aspects, features, and effects of embodiments of the disclosed technology will become apparent from the following description of embodiments with reference to the accompanying drawings. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 is a block diagram model of jitter in a conventional test and measurement system that includes a device under test (DUT) and a measuring device. [Figure 2] Figure 2 is a block diagram of an example of a test measurement apparatus including noise-corrected jitter measurement according to an embodiment of the present invention. [Figure 3A] Figure 3A shows an example flowchart illustrating the operation when performing noise-corrected jitter measurement according to the embodiment. [Figure 3B] Figure 3B shows an example flowchart illustrating the operation when performing noise-corrected jitter measurement according to the embodiment. [Figure 4] Figure 4 is a block diagram example showing an example of elements or operations that may be present in the jitter processor of a noise-corrected jitter measuring device according to an embodiment of the present invention. [Figure 5A] Figure 5A shows the combination of a conventional Gaussian distribution and a pair of Dirac functions. [Figure 5B] Figure 5B shows the combination of a conventional Gaussian distribution and a pair of Dirac functions. [Figure 5C] Figure 5C shows the combination of a conventional Gaussian distribution and a pair of Dirac functions. [Figure 6A] Figure 6A shows the result of fitting a set of dual Dirac parameters that model the distribution behavior of a set of measurement data using a q-scale plot, as implemented in an embodiment of the present invention. [Figure 6B] Figure 6B shows the result of fitting a set of dual Dirac parameters that model the distribution behavior of a set of measurement data using a q-scale plot, as implemented in an embodiment of the present invention. [Figure 7] Figure 7 is an exemplary block diagram showing additional exemplary elements or operations that may be present in the jitter processor of a noise-corrected jitter measuring device according to an embodiment of the present invention. [Modes for carrying out the invention]

[0008] As mentioned above, conventional jitter analysis methods cannot compensate for noise in the measuring instrument when measuring the device under test. There are several equivalent methods for representing jitter values. One method uses measured values ​​Jrms and JNu. Jrms is defined as the root mean square value of all uncorrelated jitter in the waveform. JNu (where N is an integer) is (1 * 10⁻¹⁰ -N The lowest observed value of ) / 2 is excluded, and further (1*10 -N Excluding the highest observed value of ) / 2, 1 - 1 * 10 -Nis defined as uncorrelated jitter values in a range where only the center of the observed value is retained. Uncorrelated jitter is defined as jitter remaining after subtracting deterministic jitter correlated with the data pattern. This deterministic jitter associated with the data pattern is sometimes called Data Dependent Jitter (DDJ). Jrms and JNu appear in Ethernet (registered trademark) (IEEE 802.3) and other high-speed serial standards, and have been used internally by PCI-SIG (Peripheral Component Interface Special Interest Group) for the purpose of investigating and identifying jitter. In the development of the PCIe Gen6 standard, the development group particularly used the jitter measurement JNu (N=6, that is, J6u) and explored several methods for correcting instrument noise. However, the PCIe standard is not clear about how to perform such measurements. Embodiments of the present invention provide noise-corrected jitter measurement that enables such measurements, and in fact, users can very well control the accuracy of noise-corrected jitter measurement for general testing of data standards and devices.

[0009] Jitter compliance values such as those used in PCIe can be expressed by Jrms and JNu introduced above.

[0010] [Formula 1] T TX-UTJ =J6u

[0011] [Formula] here, Q6=4.8916

[0012] [Formula]

[0013] When performing measurements for PCIe Gen6 compliance purposes, the DUT transmits a defined periodic pattern of 52 PAM4 (quadriple pulse amplitude modulation) symbols with 48 transitions. For various reasons, such as the physical device under development not yet being configured to generate the exact 52-symbol pattern specified by the PCIe standard, it may be beneficial to perform these or similar measurements on a different pattern or any sequence of symbols without repeating patterns. According to embodiments of the present invention, a user can define any pattern or sequence of patterns as described herein and perform measurements on them. By using embodiments of the present invention, a user can quickly perform desired jitter measurements for PCIe Gen6 and various other data communication standards.

[0014] In the following explanation, “noise” refers to unintended or unexpected fluctuations in signal parameters. When the parameter is time (absolute or delta value), the more specific term “jitter” is generally used. When the parameter is voltage, the terms “voltage noise,” “vertical noise,” or simply “noise” are used, the latter of which will be clear from the context. “Voltage” is used to describe a dependent parameter measured by an instrument such as an oscilloscope, although the actual parameter may be current, light magnitude, or other values, depending on the type of probe used.

[0015] RJ (random jitter) is the standard deviation of Gaussian random timing fluctuations. Similarly, RN, which represents random noise, is the standard deviation of Gaussian random voltage fluctuations. Standard deviation is sometimes represented by "sigma" or σ.

[0016] The terms "transition" and "edge" are used interchangeably and refer to the amplitude transition of a signal from one nominal level to another different level in a sequence of symbols, which occurs at a non-zero time interval and therefore at a non-infinite slew rate.

[0017] FIG. 1 is a block diagram model of jitter and noise in a conventional test and measurement system including a device under test (DUT) 10 and a measurement device 40 such as an oscilloscope. In this model of FIG. 1, RJi represents the standard deviation of the inherent jitter of DUT 10, which refers to random modulation of a clock signal source unrelated to the result of the amplitude of the test signal generated by the DUT. Pure data symbols from a PAM-N generator 12 are temporally modulated by a D flip-flop 16, then H u (t) 18 converts the symbols into signal transitions with finite rise time, which is the unit step response of the transmitter of DUT 10. Finally, after the inherent voltage noise RNi of DUT 10 is linearly added to the signal, the signal appears at TP1, which is the output terminal of DUT 10. Since the data generator 12 generates any number of N (N=2, 3, 4, etc.) pulse amplitude modulation symbols, it is labeled PAM-N.

[0018] The test signal is received by the measurement device 40 or an attached test probe, where RN s represents additional noise of the measurement device 40. RN s is assumed to be a stationary random variable uncorrelated with other variables. Both the signal source RJi and RNi are positively part of the values of Jrms and J6u whose characteristics are evaluated, but the value of RN s is generated by the measurement device 40, so this is not the case. RN s is the single largest cause of measurement imperfection imposed by the measurement device 40. Therefore, the influence of RN s needs to be removed before calculating and reporting Jrms and J6u.

[0019] Embodiments of the present invention include the use of a new data analysis process to eliminate the influence based on inherent random noise of the measurement device itself when measuring noise or jitter of a DUT.

[0020] When two uncorrelated Gaussian variables X and Y are superimposed, a new Gaussian variable with a calculated standard deviation is obtained, as shown in equation 4 below.

[0021]

number

[0022] Gaussian random voltage noise with standard deviation RN has a slew rate SR k When applied to the waveform edge k, the standard deviation of the resulting jitter (defined as the variation in the time at which the edge intersects with a selected fixed voltage reference value) is given by Equation 5.

[0023] [Number 5] RJ (v),k =RN / SR k

[0024] Combining equations 4 and 5, the total jitter observed at edge k is characterized as shown in equation 6.

[0025]

number

[0026] By rearranging equation 6, we obtain the equation for the noise-corrected jitter of edge k, as shown in equation 7.

[0027]

number

[0028] If only the cumulative time interval error (TIE) of multiple edges with the same slew rate in a test signal needs to be measured, and the sigma value of RN is known from the measuring instrument, then the corrected random jitter of the DUT can be calculated using Equation 7. However, many test signals contain edges with different slew rates. Therefore, the random jitter of a DUT (including PAM-N generating devices where N is greater than 2) that transmits test signals with different slew rates cannot be calculated directly from Equation 7.

[0029] In PCIe Gen6 patterns, major PAM4 edge formats, such as the 0-to-3 edge, appear four times for every 52 predefined symbol repetitions. Technically, each of these four edges is unique because its preceding and succeeding symbol sequences differ from the other three 0-to-3 transitions. Furthermore, each of the four 0-to-3 edges, or any other edge under test, may have a different slew rate. In fact, other edge formats, such as the 0-to-1 edge in an input test signal, are likely to have significantly different slew rates than the 0-to-3 edge.

[0030] Embodiments of the present invention provide the ability to classify or define edges having different characteristics. For example, a template consisting of M integers can represent M consecutively transmitted symbols. If PAM4 symbols are represented using the alphabet {0, 1, 2, 3}, then a two-symbol template

[0003] represents all 0-to-3 transitions. A four-symbol template [0 0 0 3] may represent a narrower subset of 0-to-3 transitions, i.e., 0-to-3 transitions following at least three adjacent sets of 0 symbols. Generally, longer templates result in higher pattern specificity, thus narrowing the range of slew rates represented by the subset.

[0031] By combining such specificity in the template definition with the slew rate of the measured jitter and the correction for the effects of voltage noise, embodiments of the present invention can correct voltage noise with higher accuracy than any other known method.

[0032] Figure 2 is a block diagram of an exemplary test measurement apparatus 200, such as an oscilloscope, for carrying out embodiments of the disclosed technology disclosed herein. The apparatus 200 has one or more inputs 220, which may be any electrical signal transmission medium and may function as a test interface. The signal inputs 220 receive data from the device under test 210. As described above, the DUT 210 may transmit a PAM-N signal. Clock recovery 240 and time interval error (TIE) 250 are two processes that may be performed on the input signal. The clock recovery process 240 determines the clock period and the exact clock edge position corresponding to the input signal, while the TIE process 250 calculates the time interval error for each edge in the input waveform. The test signal may be stored in acquisition memory 222 either before or after the clock recovery 240 and TIE process 250 are performed.

[0033] The noise-compensated jitter analyzer 270 may have multiple subcomponents or subprocesses, as described with reference to Figures 4 and 7 below. The jitter analyzer 270 may be implemented as any processing circuit, such as a programmed dedicated or general-purpose computer processor, an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a field-programmable gate array (FPGA), or a combination thereof. In some embodiments, the jitter analyzer 270 may be configured to execute instructions from memory 262 or other memory, and may perform any method or associated steps indicated by such instructions.

[0034] The measuring device 200 has one or more processors 260. For the sake of simplicity of illustration, only one processor 260 is shown in Figure 2, but as those skilled in the art will understand, a combination of multiple processors of various types may be used instead of a single processor 260. One or more processors 260 operate in conjunction with memory 262, which may store instructions for controlling one or more processors, data related to the measurement of the measuring device 200 or the general operation of the measuring device 200, or other data. Memory 262 may be implemented as a processor cache, random access memory (RAM), read-only memory (ROM), solid-state memory, hard disk drive, or any other memory format. And although shown as separate memories 222 and 262, one, both, or other memories may reside in various locations within the measuring device 200. Any data or instructions stored in memory 222, 262, or any other location may be accessible by any component within the measuring device 200.

[0035] The user input unit and interface 280 are coupled to or integrated into the measuring device 200. The user input unit 280 may have a keyboard, mouse, touchscreen or other optional operating device that the user can use to interactively operate the measuring device 200. The display / output unit 290 may be a touchscreen display that accepts user input and provides output from the device. Alternatively, the display / output unit 290 may be an output-only display. The display / output unit 290 may be a digital screen, computer monitor or any other monitor for displaying test results, timestamps, noise levels, jitter, jitter data or other results to the user, as described herein. The display / output unit 290 may also have one or more data outputs that may or may not correlate with the visual display. The data output from the display / output unit 290 can be sent to a data network such as a local area network, which may be coupled to a host computer for viewing the data. The display / output unit 290 may also transmit data to a remote network, such as a cloud network, accessible via the Internet by a host computer. Although the components of the test and measurement device 200 are depicted as being integrated with the test and measurement device 200, it will be understood by those skilled in the art that any of these components may be located outside the test and measurement device 200 and may be coupled to the test and measurement device 200 in any conventional manner, such as by wired or wireless communication media or other mechanisms.

[0036] The measuring device 200 may generally have one or more measuring units, which are not shown. Such measuring units may include any components that can measure the characteristics of the signal received via the input unit 220 (e.g., voltage, amperage, amplitude, etc.). The measuring device 200 may also have conditioning circuits, analog-to-digital converters, and other circuits.

[0037] Figures 3A and 3B together show exemplary flowcharts illustrating the steps in flow 300 for performing noise-corrected jitter measurement according to the embodiment. The processes described in Figures 3A and 3B or elsewhere in this application may be performed by a measuring device such as the apparatus 200 in Figure 2.

[0038] Process 300 begins in step 302 by forming a set of K templates, in which case each individual template k of the set of K templates defines a target class in multiple transitions within a signal stream, such as a test signal sent from the DUT 210 in Figure 2. These templates are typically formed to define a set of mutually exclusive transitions, and typically almost all possible transitions from the PAM-N generator are represented by one of these templates. These templates may be predefined or user-defined. Figure 4 is a block diagram of some components or processes of the jitter processor 400, which may be one embodiment of the noise-corrected jitter analyzer 270 in Figure 2. In other words, the jitter processor 400 shows components, functions or processes that may be present in the noise-corrected jitter analyzer 270.

[0039] Referring to Figure 4, the template referenced in step 302 of Figure 3A may be received from the user or may be predefined and stored in the template library 408. 2. In the example of a symbol template, a specific transition of test data can be identified. For example, template [0→3] is an example of a sample stream of input signals "1-0-2- 0-3This matches the underlined transition "-3-0-2". Other templates may include additional symbols, such as [0,0,3], [1,0,3], [2,0,3], and [3,0,3]. Some templates may use wildcards. Templates can be of any length, allowing users to gain greater specificity when analyzing input signals. Templates can also be characterized in various ways other than by the received symbols, such as slew rate.

[0040] Next, process 300 enters a loop that is executed for each of the K transitions defined in process 302. Firstly, process 304 identifies and isolates transitions in the input signal that match the current template k.

[0041] Next, for each transition that matches the input signal template, the average slew rate is determined in step 306. In step 308, TIE observations for all transitions that match the current template are accumulated, and then in step 310, a histogram is created from the TIE observations.

[0042] Referring to Figure 4, each of the steps described above may be performed by a specific component or process of the jitter processor 400. While these are generally referred to as components in this application, those skilled in the art will understand that the individual functions described with reference to the jitter processor 400 may be implemented in a complex system operating on one or more processors.

[0043] Steps 304, 306, and 308 may be performed by corresponding components within the jitter processor 400. A component of the template or pattern identification unit 420 may perform step 304 by selecting the current template k from a group of K defined templates for analysis of the waveform under test. The slew rate measurement in step 306 may be performed by the slew rate measurement component 430, and the TIE value based on step 308 may be accumulated in the TIE accumulator 440. The histogram mentioned in step 310 may be generated in the histogram generation unit 450.

[0044] The histogram analyzer / processor 460 may be used to perform an analysis on the generated histogram. For example, in step 312 (Figure 3A), the histogram analyzer / processor 460 may calculate the mean of the histogram generated in step 310. Then, in step 314, this calculated mean is removed to generate a new histogram (which is a zero-mean histogram).

[0045] Next, in step 316, the histogram analyzer / processor 460 determines the Gaussian component and the non-Gaussian component of the histogram generated in step 314. The Gaussian and non-Gaussian processors are shown in Figure 4 as processors 462 and 464. The Gaussian component is characterized by the standard deviation, and the non-Gaussian component is characterized by the peak-to-peak portion of the dual-dirac model corresponding to the histogram.

[0046] Several examples of curve analysis are shown in Figures 5A, 5B, and 5C. These figures show three examples of what happens when a Gaussian distribution 502 is combined with or convolved with a pair of Dirac functions 504, 514, and 524 spaced apart by different amounts of "DJdd". In these three examples, the same Gauss-sigma 502 is used, but because the DJdd values ​​are different, three different distributions 506, 516, and 526 are ultimately obtained. These convolution operations are sometimes called dual-Dirac models.

[0047] When performing the analysis, the Gaussian processor 462 and the non-Gaussian processor 464 may fit the histogram of the data, such as the histogram generated in step 314, to a certain curve. Figures 6A and 6B show an example of the process of fitting data 602 to curve 612 based on an example of data histogram. These figures show how to form a histogram for a given distribution of heterogeneous discrete data samples 602 to generate curve 612, and how to find a set of dual Dirac parameters {σRJ, DJdd} that gives a distribution that best matches the behavior of the measured data 602, i.e., models the behavior of the measured data 602. Figure 6A shows the data 602 and curve 612 on a linear scale, while Figure 6B shows them on a logarithmic scale. The logarithmic scale representation in Figure 6B provides a better representation of the "tails" region, i.e., the agreement of the measured data 602 with respect to curve 612 at the ends of the curve furthest from the center. Those skilled in the art will understand that while q-scale analysis is one known approach for constructing such a model, other approaches may also be used.

[0048] Referring again to Figure 3A, in step 318, the Gaussian processor 462 and the non-Gaussian processor 464, in conjunction with the dual Dirac processor 470, generate RJdd sigma and DJdd values ​​based on the histogram generated in step 314. Based on these and other values ​​observed in the measuring device 200, step 320 uses equation 5 above to determine the effective random jitter caused by noise introduced by the measuring device itself. This random jitter caused by the noise of the measuring device is determined only for the portion of the input waveform that matches each template base, i.e., the description k of the current template. In the subsequent loop, process 300 similarly determines the random jitter for input data that matches other templates. The component of jitter caused by noise of the measuring device may be determined by or in conjunction with the measuring device noise component 480 of the jitter processor 400 in Figure 4.

[0049] In step 322, the random jitter obtained in step 320 is removed from RJdd using the above formula 7, and a corrected RJdd is generated. This corrected RJdd is RJdd comp It is called [this]. Next, the jitter processor 400 uses the RJdd obtained in step 322. comp This is used to synthesize the dual Dirac probability distribution and the DJdd value obtained in step 312 with high resolution. The tails of jitter represented by this dual Dirac model are not necessarily limited to the probabilities or populations present in the actual jitter observations.

[0050] Next, steps 304-324 of flow 300 are repeated for the next template k in the series of K templates described with reference to step 302 above. Referring to Figure 4, the components of the jitter processor 400 that perform the multiple steps of flow 300 are shown as a group of components 410, with some parts of the jitter processor 400 operating on a subset of data defined by a template or class definition, and another component 414 of the jitter processor 400 operating on the sum (total) of data generated by the component group 410.

[0051] Referring to Figures 3B and 7, further steps and components used to generate noise-corrected jitter according to embodiments of the present invention will now be described. Flow 300 from Figure 3A continues to step 326 shown in Figure 3B. Step 326 combines the composite probability distribution functions generated in step 324 for each of the templates k in the set of K templates. In other words, in step 326, the probability function generated in step 324 for k=1 is added to the probability function generated in step 324 for k=2, and this process is repeated for all templates k in the set of K templates. This combination may be performed by a summation component 710 (Figure 7), which is part of the jitter processor 400 in Figure 4. Specifically, the summation process 416, as referred to in Figure 4, receives outputs from all K distributions or histograms from component group 410. The combined component 710, in process 328, creates a composite distribution that is no longer in the dual-Dirac form, but the composite distribution includes tails, i.e., the region of the distribution furthest from the center. The tails of the composite distribution are reasonably extrapolated far beyond the actual population of data analyzed in flow 300 in Figure 3A.

[0052] Based on the combined distribution with a tail, shown as 720, the distribution analyzer 722 generates an rms value in step 330, which is generated as Jrms. The combined distribution without a tail is created in step 332 by removing the tail portion of the distribution created for 720, and is shown as 730. The process of removing the tail portion of the distribution function is described above. Based on the tailless distribution 730, the distribution analyzer 732 generates a range of minimum and maximum jitter values ​​in step 334, which is recorded as J6u.

[0053] Once the Jrms and JNu parameters are generated, various equivalent jitter parameters can be generated using formulas 1-3. Other equivalent parameters may also be generated using other formulas, depending on the specific parameters to be reported.

[0054] To improve the accuracy of the jitter measurement described in this application, the steps in Figures 3A and 3B may be repeated using other sample populations from the DUT to increase the number of each of the K histogram populations.

[0055] According to embodiments of the present invention, since templates can be generated flexibly, the user's measuring device has a wide range of control over precisely specifying which data transitions from the DUT210 to use in a particular analysis.

[0056] Embodiments of the disclosed technology can operate on a specially programmed general-purpose computer, including specially created hardware, firmware, digital signal processors, or processors that operate according to programmed instructions. The terms “controller” or “processor” in this application mean microprocessors, microcomputers, ASICs, and dedicated hardware controllers, etc. Embodiments of the disclosed technology can be implemented by one or more computers (including monitoring modules) or other devices, using computer-readable data such as program modules and computer-executable instructions. Generally, program modules include routines, programs, objects, components, data structures, etc., which, when executed by a processor in a computer or other device, perform specific tasks or implement specific abstract data formats. Computer-executable instructions may be stored on computer-readable storage media such as hard disks, optical disks, removable storage media, solid-state memory, RAM, etc. As will be understood by those skilled in the art, the functions of the program modules may be combined or distributed as needed in various embodiments. Furthermore, these functions can be embodied in whole or in part in firmware or hardware equivalents such as integrated circuits or field-programmable gate arrays (FPGAs). One or more aspects of the disclosed technology can be more effectively implemented using specific data structures, such data structures are considered to be within the scope of computer-executable instructions and computer-usable data described herein.

[0057] The disclosed embodiments may, in some cases, be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried or stored in one or more computer-readable media that can be read and executed by one or more processors. Such instructions may be referred to as computer program products. The computer-readable media described herein means any medium accessible by a computing device. For example, but not limited to, computer-readable media may include computer storage media and communication media.

[0058] Computer storage media means any medium that can be used to store computer-readable information. Examples of computer storage media include, but are not limited to, random-access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory and other memory technologies, compact disc read-only memory (CD-ROM), DVD (Digital Video Disc) and other optical disc storage devices, magnetic cassettes, magnetic tapes, magnetic disk storage devices and other magnetic storage devices, and any other volatile or non-volatile removable or non-removable media implemented by any technology. Computer storage media exclude signals themselves and temporary forms of signal transmission.

[0059] A communication medium means any medium that can be used for the communication of computer-readable information. Examples of communication mediums, though not limited to them, include coaxial cables, fiber optic cables, air, or any other medium suitable for the communication of electrical, optical, radio frequency (RF), infrared, sound, or other forms of signals. Examples

[0060] The following examples are provided that are useful for understanding the technology disclosed herein. These embodiments may include one or more of the examples described below, or any combination thereof.

[0061] Embodiment 1 is a test measurement device comprising: an input unit for receiving a test waveform having multiple input level transitions from a device under test (DUT); a selector configured to individually extract only portions of the test waveform that match two or more predetermined patterns of input signal level transitions of the test waveform; a noise correction unit configured to individually determine and remove components of jitter measurements caused by random noise of the test measurement device receiving the test waveform for each of the extracted portions of the waveform; an aggregation unit configured to generate a composite distribution of timing measurements from which noise components have been removed from the extracted portions of the test waveform; and a jitter processor configured to determine a first noise-corrected jitter measurement value of the DUT from the composite distribution.

[0062] Example 2 is a test measurement apparatus according to Example 1, wherein the summing unit is configured to generate a central composite distribution by excluding the tail of the composite distribution, and the jitter processor is further configured to determine the second noise-corrected jitter measurement value of the DUT from the central composite distribution.

[0063] Example 3 is a test measurement apparatus according to Example 2, wherein the first noise-corrected jitter measurement is Jrms and the second noise-corrected jitter measurement is JNu.

[0064] Example 4 is a test measurement apparatus according to any of the preceding embodiments, wherein the slew rate of transitions that match the first predetermined pattern of the test waveform is different from the slew rate of transitions that match the second predetermined pattern of the test waveform.

[0065] Example 5 is a test measurement apparatus according to any of the preceding embodiments, wherein the noise correction unit is configured to fit a histogram of a set of timing data of an extracted portion of a test waveform that matches one of a predetermined patterns to a certain curve.

[0066] Example 6 is a test measurement apparatus according to Example 5, wherein the noise correction unit is configured to fit the histogram of the timing data set to the curve based on the characteristics of the q scale.

[0067] Example 7 is a test measurement apparatus according to Example 6, in which the curve shown above is a probability distribution corresponding to the dual Dirac model.

[0068] Embodiment 8 is a method for determining a jitter measurement value of an input signal, in which noise caused by a measuring device configured to receive an input signal has been corrected. The method of this embodiment comprises: receiving the input signal from a device under test (DUT) with the measuring device; generating a test waveform from the received input signal; selecting two or more portions of the test waveform that match two or more predetermined patterns of the input signal; individually determining the jitter component caused by measuring device noise for each of the two or more selected portions and individually synthesizing histograms of timing measurements corrected for measuring device noise; summing the noise-corrected histograms of the timing measurements of the two or more selected portions to generate a composite probability distribution; and determining a jitter measurement value of the DUT corrected for noise from the composite probability distribution.

[0069] Example 9 is a method according to Example 8, wherein the process for individually determining the jitter components caused by measuring device noise includes, for each of the two or more selected portions of the test waveform, a process for generating an unprocessed (raw) histogram of the timing error of the selected portion of the test waveform; a process for removing the average value of the timing error from the unprocessed histogram to generate a zero-mean histogram; and a process for performing a dual Dirac analysis to fit the zero-mean histogram to determine the Gaussian and non-Gaussian components of the zero-mean histogram.

[0070] Example 10 is the method according to Example 9, further comprising the steps of: deriving dual Dirac deterministic jitter from the zero-mean histogram for each of the two or more selected portions of the test waveform; and deriving random jitter from the zero-mean histogram.

[0071] Example 11 is a method according to Example 10, further comprising: a process for determining, for each of the two or more selected portions of the test waveform, a random jitter component added to the random noise of the measuring device due to the influence of the average slew rate of the selected portion of the test waveform; and a process for removing the added random jitter component to generate a corrected random jitter component for the selected portion of the test waveform.

[0072] Example 12 is the method according to Example 11, further comprising the process of synthesizing a corrected dual-Dirac model of the selected portion of the test waveform by convolving the corrected random jitter component with the dual-Dirac deterministic jitter from the zero-mean histogram for each of the two or more selected portions of the test waveform.

[0073] Example 13 is a method according to any of the methods of the above examples, wherein the process of summing noise-corrected histograms of timing measurements comprises a process of performing a weighted summation of noise-corrected histograms of the two or more selected portions, at least in part on the relative number of input signal transitions in each of the two or more selected portions.

[0074] Example 14 is a method according to any of the above examples, further comprising a process of generating a median-sum distribution by excluding the tails of the composite probability distribution, and a process of determining a noise-corrected jitter value from the median-sum distribution.

[0075] Example 15 is a test measurement apparatus comprising an input unit for receiving an input signal from a device under test (DUT) and creating a test waveform from the received input signal, and one or more processors, wherein the one or more processors are configured to execute a program that causes the one or more processors to perform the following in order to determine the jitter in the input signal after noise caused by the device that receives the input signal has been corrected: receiving the input signal with the receiving device; generating a test waveform from the received input signal; selecting two or more portions of the test waveform that match two or more predetermined patterns of the input signal, respectively; individually determining the jitter component caused by the noise of the measuring device for each of the two or more selected portions and individually synthesizing histograms of the timing measurements corrected for the noise of the measuring device; summing the noise-corrected histograms of the timing measurements of the two or more selected portions to generate a composite probability distribution; and determining the noise-corrected jitter measurement value of the DUT from the composite probability distribution.

[0076] Example 16 is a test measurement apparatus according to Example 15, wherein the process for individually determining the jitter component caused by the measurement apparatus noise includes, for each of the two or more selected portions of the test waveform, a process for generating an unprocessed histogram of the timing error of the first portion of the test waveform; a process for removing the average value of the timing error from the unprocessed histogram to generate a zero-mean histogram; and a process for performing a dual Dirac analysis to fit the zero-mean histogram to determine the Gaussian and non-Gaussian components of the zero-mean histogram.

[0077] Example 17 is a test measurement apparatus according to Example 16, wherein one or more processors are further configured to perform the process of deriving dual Dirac deterministic jitter from the zero-mean histogram and the process of deriving random jitter from the zero-mean histogram.

[0078] Example 18 is a test measurement apparatus according to Example 17, wherein one or more processors are further configured to perform the following processes for each of the two or more selected portions of the test waveform: determining a random jitter component added to the random noise of the receiving device due to the influence of the average slew rate of the selected portion of the test waveform; and removing the added random jitter component to generate a corrected random jitter component for the selected portion of the test waveform.

[0079] Example 19 is a test measurement apparatus according to any of Examples 15 to 18, wherein one or more processors are further configured to synthesize a corrected dual-Dirac model of the selected portion of the test waveform by convolving the corrected random jitter component with the dual-Dirac deterministic jitter from the zero-mean histogram for each of the two or more selected portions of the test waveform.

[0080] Example 20 is a test measurement apparatus according to Example 19, wherein one or more processors are further configured to perform a process of summing noise-corrected histograms of timing measurements by performing a weighted summation process of noise-corrected histograms of timing measurements of the two or more selected parts, at least in part based on the relative number of input signal transitions in each of the two or more selected parts.

[0081] Example 21 is a test measurement device according to any of Examples 15 to 20, wherein one or more processors are further configured to perform the process of generating a median-sum distribution by excluding the tails of the composite probability distribution, and the process of determining a noise-corrected jitter value from the median-sum distribution.

[0082] The disclosed versions of this object have many effects that are described or will be obvious to those skilled in the art. Nevertheless, not all of these effects or features are required in all versions of the disclosed apparatus, system, or method.

[0083] In addition, the description of this application refers to certain features. It should be understood that the disclosures herein include all possible combinations of these particular features. Where a particular feature is disclosed in relation to a particular aspect or example, that feature may, to the extent possible, also be used in relation to other aspects and examples.

[0084] Furthermore, when this application refers to a method having two or more defined steps or processes, these defined steps or processes may be performed in any order or simultaneously, as long as the circumstances do not rule out such possibilities.

[0085] For the sake of explanation, specific embodiments of the present invention have been illustrated and described, but it should be understood that various modifications are possible without deviating from the gist and scope of the present invention. Therefore, the present invention should not be limited to anything other than the appended claims.

Claims

1. A test and measurement device, An input section for receiving a test waveform having multiple input level transitions from the device under test (DUT), A selector configured to individually extract only the portions of the test waveform that match two or more predetermined patterns of input signal level transitions of the test waveform, A noise correction unit is configured to individually determine and remove, for each extracted portion of the above test waveform, the component of the jitter measurement value caused by random noise in the test measurement device that receives the above test waveform. The summing unit is configured to generate a composite distribution of timing measurement values ​​obtained by removing noise components from the extracted portion of the above test waveform, A jitter processor configured to determine the first noise-corrected jitter measurement value of the DUT from the above composite distribution, A test and measurement device equipped with [the following features].

2. The test measurement apparatus according to claim 1, wherein the summing unit is configured to generate a central composite distribution by excluding the tail of the composite distribution, and the jitter processor is further configured to determine a second noise-corrected jitter measurement value of the DUT from the central composite distribution.

3. A method for determining a jitter measurement value of an input signal, after correcting for noise caused by a measuring device configured to receive an input signal, The process involves receiving the above input signal from the device under test (DUT) using the above measuring device, The process of generating a test waveform from the received input signal, A process of selecting two or more portions of the test waveform that match two or more predetermined patterns of the input signal, respectively. For each of the two or more selected parts above, the jitter component caused by the measuring device noise is individually determined, and the histograms of the timing measurements corrected for the measuring device noise are individually synthesized. The process involves summing the noise-corrected histograms of the timing measurements of two or more selected portions to generate a composite probability distribution. A process to determine the noise-corrected jitter measurement value of the DUT from the above composite probability distribution, and A method for determining jitter measurement values ​​that includes the following.

4. The process of individually determining the jitter component caused by noise from the measuring device is performed for each of the two or more selected portions of the test waveform. The process involves generating an unprocessed histogram of timing errors in the selected portion of the above test waveform, The process involves removing the average value of the timing error from the above unprocessed histogram to generate a zero-mean histogram. The process involves performing a dual Dirac analysis to fit the zero-mean histogram to determine the Gaussian and non-Gaussian components of the zero-mean histogram, and A method for determining jitter measurement values ​​according to claim 3.

5. For each of the two or more selected portions of the above test waveform, The process of deriving the dual Dirac deterministic jitter from the zero-mean histogram above, The process of deriving random jitter from the zero-mean histogram above The method for determining jitter measurement value according to claim 4, further comprising the above.

6. A test and measurement device, An input section for receiving an input signal from the device under test (DUT) and creating a test waveform from the received input signal, One or more processors and The system comprises one or more processors that, in order to determine the jitter in the input signal from which noise caused by the device receiving the input signal has been corrected, The above receiving device performs the process of receiving the above input signal, The process of generating a test waveform from the received input signal, A process of selecting two or more portions of the test waveform that match two or more predetermined patterns of the input signal, respectively. For each of the two or more selected portions mentioned above, the jitter component caused by the measuring device noise is determined individually, and the histograms of the timing measurements corrected for the measuring device noise are synthesized individually. A process to generate a composite probability distribution by summing the noise-corrected histograms of the timing measurements of two or more selected parts above, A process to determine the noise-corrected jitter measurement value of the DUT from the above composite probability distribution, and A test and measurement device configured to execute a program that causes one or more of the above-mentioned processors to perform the above-mentioned task.

7. The process of individually determining the jitter component caused by noise from the measuring device is performed for each of the two or more selected portions of the test waveform. The process involves generating an unprocessed histogram of the timing errors in the first part of the above test waveform, The process involves removing the average value of the timing error from the above unprocessed histogram to generate a zero-mean histogram. The process involves performing a dual Dirac analysis to fit the zero-mean histogram to obtain the Gaussian and non-Gaussian components of the zero-mean histogram, and A test and measurement apparatus according to claim 6, having the following features.

Images