Test measurement system and signal generation method

Abstract

This replicates signals from a device under test with specific environmental and channel conditions. [Solution] The test measurement system 10 includes a plurality of nodes 12, each having one or more sensors configured to receive signals; one or more generative adversarial network (GAN) models; one or more signal generators configured to generate and transmit signals; and one or more processors 16. The system is configured to execute a program that causes one or more processors to perform the following processes: receive a request for a signal having a signal profile from any of the plurality of nodes 12, send the signal profile and the request to one or more GAN models, receive a signal profile that matches the signal profile, send the matching signal profile to one or more signal generators, and generate and transmit a matching signal.

Description

Technical Field

[0001] The present disclosure relates to a test measurement system, and particularly to artificial intelligence (AI) for generating waveforms for test measurement systems and environments.

Background Art

[0002] A waveform is a graph of a signal over time and has various functions in the testing of electronic circuits. A device under test (DUT) generates a signal that forms a waveform, enabling analysis of the performance of the DUT. The DUT may generate a signal in response to a signal being applied to the DUT. Some test devices, such as arbitrary waveform generators and arbitrary function generators, generate and apply signals.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Non-Patent Documents

[0004]

Non-Patent Document 1

Non-Patent Document 2

[0005] In many cases, updating the software built into these waveform generation products is necessary to maintain their ability to generate various types of waveforms with specific characteristics. Furthermore, some characteristics of the generated signals may not take into account the effects of unknown channels or unknown environments on the signal.

[0006] The ability to reproduce signals from the DUT and generate signals for specific environments and channel conditions provides a level of flexibility and precision previously unattainable. [Means for solving the problem]

[0007] Embodiments of the present invention include a test and measurement system comprising a network of multiple nodes, one or more central servers, and a database. Some embodiments may include a large language model (LLM) interface to the database. Some embodiments have multiple nodes, each consisting of multiple test and measurement devices. Many, if not all, of the nodes have the processing power to operate a generative adversarial network (GAN). These processing powered nodes can also be configured to utilize federated learning. [Brief explanation of the drawing]

[0008] [Figure 1]Figure 1 shows a diagram of an artificial intelligence-based system for test measurement. [Figure 2] Figure 2 shows a flowchart of an embodiment of the closed-loop optimization test and measurement system. [Figure 3] Figure 3 shows an embodiment of a sensor node including a test measurement device. [Figure 4] Figure 4 shows an embodiment of a node during reception processing. [Figure 5] Figure 5 shows an embodiment of a node during the transmission process. [Figure 6] Figure 6 shows a flowchart of an embodiment of a method for generating a matching signal. [Figure 7] Figure 7 shows an embodiment of an artificial intelligence-based system for testing and measurement under specific environmental conditions. [Modes for carrying out the invention]

[0009] Figure 1 shows an embodiment of a test measurement system. The test measurement system 10 includes a network of multiple nodes, such as node 12, one or more servers, such as server 14, and one or more databases, such as database 22. Server 14 is called a “centralized server,” which means that server 14 communicates with many, if not all, nodes 12, not in the sense of its physical location. Server 14 and the nodes 12 that make up the test measurement system may have one or more processors 16, memory 24, ports 18 for server 14 or node 12 to communicate with the device under test (DUT), and a machine learning system 20. The machine learning system 20 may have a processor separate from the main processor, which is programmed to run the machine learning process, and both are represented by processor 16.

[0010] In this explanation, the term "artificial intelligence (AI)" refers to machine learning, which means neural networks and other machine learning architectures that have undergone supervised or unsupervised learning to produce output, as well as "generative AI." Generative AI may have a large-scale language model (LLM) and may include a transformer. Generative AI may also refer to GANs, which are more likely to operate without a transformer.

[0011] A GAN typically consists of two neural networks. One network, i.e., a generative network or generator, produces an output that mimics the input. The other network, a discriminative network or discriminator, evaluates the output and determines whether it truly mimics the input. The two networks continue to "compete" with each other until the output mimics the input very faithfully, and the discriminator fails to distinguish the differences that are usually determined by error handling, and the discriminator "loses." In this application, the terms "GAN model" or "GAN" refer to a pair of neural networks. Multiple GAN models, i.e., multiple pairs of neural networks, may exist in AI20.

[0012] System 10 in Figure 1 employs federative learning, and a basic GAN model or "bootstrap" model is present in a large part, if not the entire system. Some of the nodes 12 in System 10 may consist of forwarding nodes or other types of nodes that do not manipulate inputs or outputs. These nodes handle data distribution (routing) and other tasks that do not involve receiving or transmitting signals from outside the system. In this description, nodes that receive signals are referred to as receiving nodes, and nodes that send signals outside the system are referred to as transmitting nodes. Many nodes may be both receiving and transmitting nodes.

[0013] For ease of understanding, as an example, node 12 of system 10 may be configured as a sensor node that receives signals. Since the waveform occurs based on the signal over time, the terms "signal" and "waveform" are used synonymously. The signal may be a signal from the DUT or may be composed of "threat" signals such as the input radar signal. Node 12 receives the signal and generates a response. This response may include replicating the signal using a GAN and deriving environmental and channel characteristics based on this replication process.

[0014] Figure 2 shows an embodiment of each element of the AI component within AI component 20 and their interactions. As described above, AI component 20 may exist on server 14 within system 10 of FIG. 1 and on a plurality of nodes. This process has a receiving side 30 and a transmitting side 40. On the receiving side 30, the input signal data received by the sensor becomes learning data 34, which may, in some cases, be combined with verification data 32. Verification data 32 may be received by direct sampling of the DUT environment (referred to as online in this application) or by file upload (referred to as offline in this application). Learning data 34 and context information 36 serve as inputs to GAN learning discriminator 42. Context information 36 includes the environment in which the receiving node exists, details of the node's characteristics (processing power, role, etc.), and characteristics of the received signal (derived from the DUT, external signal source, etc.). GAN learning discriminator 42 receives parameters 39 extracted from learning data 34, context information 36, and verification data 32. GAN learning discriminator 42 sets what form the replicated signal of the received signal should be.

[0015] The data coming from the receiver 30 passes directly through the parameter extraction module 38 and is stored in the waveform library 44. The input data 32 and 34 may undergo digital additions or encoding, such as watermarking. The waveform library 44 may also contain context information 36. The GAN generator 48 may receive the received waveform from the waveform library 44 if the replication / transmission process is performed sequentially with the received signal. If a response to the received signal is not required, the system may rely on encoding to save the signal for later use and to find it systematically and quickly.

[0016] If a response occurs, the GAN generator 48 may use a saved waveform or, in the case of a continuous response, the actual response waveform. Based on the input received signal, the GAN generator 48 generates a candidate signal. The candidate signal generated by the GAN generator 48 may be a duplicate of the received signal, a duplicate of the received signal with some characteristics altered, or an inverse signal used to "cancel" the received signal in electronic warfare. The characteristics of the signal are also called the signal profile. The generated signal is then compared in the comparator 47 with verification data 32 of the received signal that has passed through the parameter extraction process 38 without modification.

[0017] This comparison process may use error measures such as mean squared error (MSE) or other error measures. The optimization function applied to comparator 47 controls the error measurement. The optimization function generally finds the best solution and usually shows the maximum or minimum value. The optimization function evaluates candidate solutions and defines an objective function to find the input that yields the most desirable output. In this case, this objective function minimizes the error between the "real waveform" and the synthesized waveform. If the error measurement of the synthesized waveform is too large, the "real" waveform based on the received signal may be selected. If the real waveform is insufficient because the desired waveform is an inverse or some kind of distortion of the real waveform, the data may be mixed between the real waveform and the synthesized waveform until the desired performance metric is met.

[0018] The error measurement values include, but are not limited to, mean squared error (MSE), mean absolute error, root mean square error, logarithmic mean squared error, adjusted logarithmic mean squared error, median absolute deviation, etc. This process can evaluate the accuracy of the GAN model using any of these error measurement values. The generated, i.e., "synthesized," signal also undergoes filtering by filter 49 to adjust the error detected in the comparison process. The synthesized signal without filtering or the filtered signal is then selected by selector 46 according to the error measurement value and the output. For example, if the MSE is greater than the target, selector 46 selects the filtered signal. Otherwise, it selects the unfiltered synthesized signal.

[0019] Next, if the generated signal requires transmission, the output signal may be transmitted. Alternatively, it is also possible to apply the output signal to the DUT. The application of the signal to the DUT is performed when the DUT is under test, originally has not received the signal, and there is a test need for the DUT to have a waveform with specific characteristics.

[0020] When the selector 46 selects the synthesized signal from the GAN generator 48, the synthesized signal may be stored in the waveform library 44 together with the relevant error measurement value. The error measurement value may be included in the waveform used as the selection criterion for the waveform library 44.

[0021] Figure 3 shows an example of a node that may exist in the network shown in Figure 1, which has a test measurement device. In one embodiment, the node may use a GAN to replicate signals within the device. The test measurement device may consist of a “composite” device, meaning that the device consists of two different devices. In the embodiment of Figure 3, the composite device consists of a mixed-signal oscilloscope (MSO) 54 and an arbitrary waveform generator (AWG) 58. The term “AWG” as used in this application includes both arbitrary waveform generators and arbitrary function generators.

[0022] DUT50 is connected to MSO54 through one or more channels 52. In this embodiment, channel 52 may include some of the multiple channels of MSO54 or additional wiring connected to DUT50. However, the overall characteristics of channel 52 may be unknown. DUT50 generates a signal, which reaches MSO54 via channel 52. MSO54 receives the signal on channel 52 and copies the signal onto channel 53. The signal input to the first channel 52 undergoes signal processing and signal adjustment in MSO54. This adjustment process includes analog-to-digital conversion, signal adjustment, and timing adjustment. The signal from the first channel 52 is filtered by filter 56 and then duplicated by GAN57. GAN57 iterates through each other until the discriminator determines that the generator has produced a duplicate signal that is indistinguishable from the original. The iteration of GAN57 generates a signal for the "calibrated" channel 59. AWG58 receives the signal from GAN57 and generates another (identical) signal on channel 59. The signal on channel 59 is compared with the "original" signal, and the results may be used to adjust filter 56 to remove the difference or modify the original signal as needed. The "original" signal at this time is input to channel 52, goes through MSO to channel 53, and is sent to comparator 55. The results of this comparison can also be used to generate the characteristics of channel 52 based on the calibration process. This channel characterization and the original signal may be stored in waveform library 22 in Figure 1 or waveform library 44 in Figure 2.

[0023] As described above, any node in a system capable of running a GAN can participate in federative learning. In federative learning, each node may broadcast updates to its bootstrap model as the GAN model on that node is tuned and modified through the discriminator process. Other nodes in the system behave similarly, and as a result, all nodes running GANs in the system that are capable of updating themselves will be in parity with respect to their models. In some embodiments, various federative learning policies may be implemented by different nodes, such as accepting only selected updates and rejecting all others, or conversely, rejecting only certain updates and accepting all others.

[0024] Furthermore, updated nodes send their update information to the central or main server to update the models residing on the server. In the system, when a new node starts service, the update process on the server allows that node to receive the latest updated version of the model. The node also receives the bootstrap model and then updates, allowing it to filter out updates that are irrelevant to it.

[0025] Figures 4 and 5 show more generalized versions of the node for both the receiving process in Figure 4 and the transmitting process in Figure 5. In the receiving process in Figure 4, the sensor node 60 receives the input signal and further captures the characteristics of the DUT, the environment under test (EUT), and the channel under test that received the signal, and provides these to the model on the node. The signal is adjusted by the signal conditioner 62, and the GAN 64 operates so that the duplicated signal is as close as possible to the original signal. Next, result 66 is generated so that it can be analyzed by those who need it, and the obtained duplicated signals are stored in the waveform database 68, which includes inverse signals and variant signals from the original signal. The waveform database 68 may be configured as a library, as described below. The measurement and metering module 70 enables the sensor node 60 to measure the input signal, decompose the signal components and characteristics as needed, and monitor the performance of GAN64. Depending on the measurement results, the measurement and metering module may interact with other modules as needed, for example, by communicating with sensors, which is done if the information was not fully received or if there is a problem with the signal. If signal adjustment is necessary due to the nature of the signal, module 70 may also communicate with the signal adjustment module. When the results of module 70 are changed while GAN64 is running, the results need to be updated, so module 70 and module 66 may exchange results until they are ready to be saved.

[0026] Figure 5 illustrates the transmission process. When a request for a signal is received in response to an input signal or for testing a DUT, EUT, or channel, the node or central server in Figure 1 accesses the database / library 68 and provides the desired version of the signal via the results buffer module 66. The signal is then provided to the GAN 64 along with the required specific characteristics, and the GAN 64 operates to generate the desired signal. The signal undergoes adjustments, such as digital-to-analog conversion, in the signal adjustment unit 62 and is supplied to the sensor node 60, which the AWG on this sensor node uses to generate and transmit the signal.

[0027] As mentioned above, changes and updates to the GAN model may be distributed to other nodes as part of the federative learning environment.

[0028] With the development of the waveform database and the increase in the amount of data, the waveform database 22 can also be used as a library for use with large-scale language models. Returning to Figure 1, the large-scale language model (LLM) interface 26 allows the user to input prompts to generate (develop) specific waveforms as needed. The prompts are based on the test environment, and the characteristics of that test environment and the waveform specified by the user can be converted into an LLM-type vector. The LLM vector is then used to access the waveform library to find a matching signal or a signal close to a matching signal, and based on this signal, the GAN model can operate to generate the desired waveform.

[0029] This disclosure includes concepts related to generating waveforms using a learned large waveform model (LWM). This embodiment involves training a model using artificial intelligence or machine learning based on uploaded data applicable to the target experiment. The generated waveforms are based on recognition of the context (background and situational information) of the transmitted test environment. The interface used for querying can be various interfaces, not just the keyboard.

[0030] The distributed learning that enables this embodiment is based on the principles of distributed learning, which include, but are not limited to, associative learning, and has an application programming interface (API) that enables direct conversion, loading, or waveform sequencing for delivering physical streaming data interacting with communication media such as air or water.

[0031] This disclosed technology solves the problem of rapidly creating prototypes of stimulus waveforms for a device under test (DUT) and provides a unified format for performing agnostic waveform transformation, which is used to transmit signals to the real world. Furthermore, users benefit from other participants who have permitted them to share their learned waveforms, obtained from shared measurement data and uploaded to the platform, in order to further train the relevant LWM. The LWM may be public or private, and may be specific to a single industry and its participants.

[0032] Dynamically prototyping waveforms directly using memory via the LLM interface is beneficial for learning optimal waveforms based on the context (background and circumstances) provided through the LLM interface. Based on the provided context data, the learned waveforms may be tagged and grouped according to the customer's unique DUT environment. This functionality does not exist at the time of this disclosure.

[0033] In one embodiment, the user interface to the test measurement system consists of a large-scale language model interface. In embodiments using the large-scale language model interface, metadata is tagged to the provided waveforms, allowing the library to be updated with new models.

[0034] In this way, the test measurement system can use machine learning to receive and generate signals and waveforms as needed, while ensuring accuracy and matching to the desired signal profile.

[0035] As an example of the application of this system, Figures 6 and 7 show a flowchart of an embodiment of a method for generating a matching signal from a received signal, and a diagram of nodes in a specific embodiment of a threat environment. For example, sensor nodes 90, 92, or 94 in Figure 7 may receive a test signal representing a threat signal from an enemy in electronic warfare (see Patent Documents 3 and 4). The system receives the received signal and determines a response. Depending on the capabilities of the node (e.g., sensor nodes 90, 92, 94), if the node does not have the capability to respond, the response may be made from a central server 14 (one of the nodes).

[0036] In any of the above situations, the node generates a request for a signal having a signal profile at 80 in Figure 6, which may include, but is not limited to, information about the signal, the environment in which the node is operating, metadata information related to the signal, possibly a target for the matching signal to be transmitted, and the captured information described above. This request is sent at 82 to one or more GAN models as described above. These GAN models may be GAN models from the node to the central server, and models on adjacent nodes with the capability to do so, and all of these nodes are referred to in this application as the requester. The requester then receives a matching signal profile from one or more GAN models at 84. The requester sends the matching signal profile to one or more signal generators at 86. As an optional process, at 88, the signal generators generate a signal and send it to the node, so that the node may send a signal that matches the received signal as needed.

[0037] One form of response is a countermeasure signal for electronic warfare, which interferes with the received signal. Another form may include a radar signal transmitted to pinpoint the location of the signal source. The received signal may undergo adjustment processing to generate two signals, both of which are in some form duplicates of the received signal. The system generates the duplicated signals based on the underlying signal and access to a waveform library / database, and these generated signals are transmitted by an AWG within the system. Note that the above description is not limited to the specific embodiment used in this application, but merely constitutes an example.

[0038] 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 type expressions. Computer-executable instructions may be stored on computer-readable storage media such as hard disks, optical disks, removable storage media, solid-state memory, and RAM. 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.

[0039] 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.

[0040] 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.

[0041] A communication medium means any medium that can be used to transmit 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 transmitting electrical, optical, radio frequency (RF), infrared, sound, or other types of signals. Examples

[0042] 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.

[0043] Example 1 is a test measurement system, A plurality of nodes, each having one or more nodes with sensors configured to receive signals, One or more Generative Adversarial Network (GAN) models, One or more signal generators configured to generate and transmit signals, One or more processors and Equipped with, The one or more processors The process of receiving a request for a signal with a signal profile from any of the above multiple nodes, The process of sending the above signal profile and the above request to one of the GAN models among the one or more above, The process involves receiving a matching signal profile that matches the above signal profile, The process involves sending the above matching signal profile to one of the one or more signal generators, and generating and transmitting a matching signal. It is configured to execute a program that causes one or more of the above processors to perform the task.

[0044] Example 2 is a test measurement system of Example 1, wherein the signal profile includes the environment in which any of the nodes among the plurality of nodes operates, the metadata of the request, and the potential target of the matching signal.

[0045] Example 3 is a test measurement system according to either Example 1 or 2, wherein one or more processors are further configured to execute a program that causes one or more processors to perform the process of saving the matching signal to a database.

[0046] Embodiment 4 is the test measurement system of Embodiment 3, wherein one or more processors are configured to execute a program that causes the one or more processors to perform the following processes: accessing the database to determine whether the signal profile exists in the database, and sending the signal profile from the database to one or more GAN models.

[0047] Example 5 is a test measurement system according to any of Examples 1 to 4, wherein one or more processors are further configured to execute a program that causes one or more processors to perform a process that operates one or more GAN models to generate the matching signal profile.

[0048] Example 6 is a test measurement system according to any of Examples 1 to 5, wherein the one or more processors configured to execute a program that causes the one or more processors to perform the process of operating the one or more GAN models include one or more processors on the node that sent the request among the multiple nodes.

[0049] Example 7 is a test measurement system according to any of Examples 1 to 6, wherein the one or more processors configured to execute a program that causes the one or more processors to perform the process of operating the one or more GAN models include one or more processors on a central server within the test measurement system.

[0050] Example 8 is a test measurement system of Example 7, wherein one or more processors on the central server are further configured to execute a program that causes one or more processors to perform the process of distributing the updated GAN model obtained from the request to the multiple nodes.

[0051] Example 9 is a test measurement system according to any of Examples 1 to 8, wherein one or more of the sensors receive an input signal from the device under test (DUT).

[0052] Example 10 is a test measurement system of Example 9, further comprising one or more sensors, and the test measurement device, A port on the channel of the test and measurement device that receives the input signal from the above DUT, One or more analog-to-digital converters (ADCs) for converting the above input signal from the above DUT into a digital representation of the above input signal, One or more processors and It has, The one or more processors A process for evaluating the characteristics of the above input signal, The process involves using the local GAN ​​model present in the above-mentioned test measurement device to match its characteristics and duplicate the above-mentioned input signal as a duplicated signal. The process of sending updates to the above local GAN ​​model to the central server It is configured to execute a program that causes one or more of the above processors to perform the task.

[0053] Example 11 is a test measurement apparatus of Example 10, wherein one or more processors are further configured to execute a program that causes one or more processors to perform the process of transmitting the replication signal to one or more databases.

[0054] Example 12 is a test measurement system according to any of Examples 1 to 11, wherein one or more signal generators include an arbitrary waveform generator.

[0055] Example 13 is a method, The process of receiving a request for a signal with a signal profile from a node in the test measurement system, The process involves sending the above signal profile and the above request to one or more GAN models within a Generative Adversarial Network (GAN) model, The process involves receiving matching signal profiles that match the above signal profile from one or more of the above GAN models, The process involves sending the above matching signal profile to one or more signal generators, and generating and transmitting a matching signal. It is equipped with.

[0056] Example 14 is the method of Example 13, wherein the signal profile includes information about the environment in which the node operates, the metadata of the request, and one or more of the potential targets of the matching signal.

[0057] Example 15 is a method of either Example 13 or 14, further comprising the process of storing the matching signal in a database.

[0058] Example 16 is the method of Example 15, further comprising the process of accessing the database and determining whether the signal profile exists in the database, and the process of sending the signal profile from the database to one or more GAN models.

[0059] Example 17 is the method of Example 16, further comprising the process of operating one or more GAN models to generate the matching signal profile.

[0060] Example 18 is one of the methods from Examples 13 to 17, further comprising the process of distributing the updated GAN model obtained from the above request to other nodes in the system where the updated GAN model resides.

[0061] Example 19 is one of the methods from Examples 13 to 18, wherein the process of receiving the above-mentioned signal request includes the process of receiving an input signal from the device under test (DUT) at the sensor node.

[0062] Example 20 is the method of Example 19, wherein the process of receiving the input signal at the sensor node is as follows: A process for evaluating the characteristics of the above input signal, The process involves using the local GAN ​​model present on the above node to match its characteristics and duplicate the above input signal as a duplicate signal, The process of sending update information for the above local GAN ​​model to the central server. Includes.

[0063] Example 21 is the method of Example 20, further comprising the process of transmitting the above-mentioned replication signal to one or more databases.

[0064] All functions disclosed in the specification, claims, abstract and drawings, and all steps in any method or process disclosed, may be combined in any combination, except where at least some of such functions or steps are mutually exclusive. Each of the functions disclosed in the specification, abstract, claims and drawings may be replaced by an alternative function that serves the same, equivalent or similar purpose, unless otherwise specified.

[0065] 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. For example, if a particular feature is disclosed in relation to a particular aspect or example, that feature may also be available in relation to other aspects and examples, as far as possible.

[0066] 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.

[0067] For the sake of explanation, specific aspects of the disclosed technology have been illustrated and described, but it should be understood that various modifications are possible without deviating from the gist and scope of this disclosure. [Explanation of symbols]

[0068] 10 Test and Measurement Systems 12 nodes 14 Servers 16 processors 18 ports 20 AI Components 22 Databases 24 memory 30 Receiving side 32 Verification Data 34 Training data 36 Contextual Information 38 Parameter Extraction Module 40. Transmitter 42 GAN learning classifiers 44 Waveform Library 46 Selector 47 Comparator 48 GAN generator 49 Filters 50 Device under test (DUT) Channel 52 Channel 53 54. Mixed-Signal Oscilloscope (MSO) 56 filters 57 GAN 58 Arbitrary Waveform Generator (AWG) Channel 59 60 Sensor Nodes 62 Signal Adjustment Section 64 GAN 66 Result Buffer Module 68 Waveform Databases / Libraries 70 Measuring and weighing modules 90 Sensor Nodes 92 Sensor Nodes 94 Sensor Nodes

Claims

1. A test and measurement system, A plurality of nodes, each including one or more nodes having sensors configured to receive signals, One or more Generative Adversarial Network (GAN) models, One or more signal generators configured to generate and transmit signals, One or more processors and Equipped with, The one or more processors The process of receiving a request for a signal with a signal profile from any of the above multiple nodes, The process of sending the above signal profile and the above request to one of the GAN models among the one or more GAN models, The process involves receiving a matching signal profile that matches the above signal profile, The process involves transmitting the above matching signal profile to one of the one or more signal generators, and generating and transmitting a matching signal. A test and measurement system configured to execute a program that causes one or more of the above-mentioned processors to perform the above task.

2. The test measurement system according to claim 1, wherein the signal profile includes the environment in which any of the nodes among the plurality of nodes operates, metadata of the request, and potential targets of the matching signal.

3. One or more of the above processors further, The process of saving the above matching signal to a database, The process involves accessing the above database and determining whether the above signal profile exists in the above database, The process of sending the above signal profile from the above database to one or more above GAN models, The test measurement system according to claim 1, configured to execute a program that causes one or more of the above-mentioned processors to perform the above-mentioned task.

4. The test measurement system according to claim 1, wherein one or more processors are further configured to execute a program that causes one or more processors to perform a process to operate one or more GAN models to generate the matching signal profile.

5. One or more of the above sensors further include a test and measurement device, and the test and measurement device is A port on the channel of the above-mentioned test measurement device receives an input signal from the device under test (DUT), One or more analog-to-digital converters (ADCs) for converting the input signal from the above DUT into a digital representation of the input signal, One or more processors and It has, The one or more processors A process for evaluating the characteristics of the above input signal, Using the local GAN ​​model present in the above test measurement device, the process involves matching the characteristics and duplicating the above input signal as a duplicated signal. The process of sending updates to the above local GAN ​​model to the central server and The test measurement system according to claim 1, configured to execute a program that causes one or more of the above-mentioned processors to perform the above-mentioned task.

6. The process of receiving a request for a signal with a signal profile from a node in the test measurement system, The process involves sending the above signal profile and the above request to one or more Generative Adversarial Network (GAN) models, A process of receiving matching signal profiles that match the above signal profile from one or more of the above GAN models, The process involves sending the above matching signal profile to one or more signal generators, and generating and transmitting a matching signal. A signal generation method comprising the following.

7. The process of saving the above matching signal to a database, The process involves accessing the above database and determining whether the above signal profile exists in the above database, The process of sending the above signal profile from the above database to one or more above GAN models, The signal generation method according to claim 6, further comprising the above.

8. The signal generation method according to claim 6, further comprising the process of distributing the updated GAN model obtained from the above request to other nodes in the system in which the updated GAN model exists.

9. The process that receives the above signal request is, The sensor node processes the input signal from the device under test (DUT), A process for evaluating the characteristics of the above input signal, The process involves using the local GAN ​​model present on the above node to match its characteristics and duplicate the above input signal as a duplicated signal, The process involves sending update information for the above local GAN ​​model to the central server. The signal generation method according to claim 6, having the following characteristics.

10. The signal generation method according to claim 9, further comprising the process of transmitting the above-mentioned duplicated signal to one or more databases.

Images