Waveform generation system and method

Abstract

To artificially generate waveform data for machine learning.SOLUTION: An AI assistant 10 receives, from a user 20, input such as patterns and definitions relating to waveforms to be generated, and accesses a memory 14 that stores knowledge of test measurement, waveform fault information 16, descriptions 18 of test measurement devices, and various protocol standards 26, and provides knowledge required for waveform generation to a generative AI model 12. Further, via an assistant API, a vector database and retrieval-augmented generation (RAG) may be connected to a system so as to provide the generative AI model 12 with knowledge / information of specific domains relating to waveforms for test measurement. Using these, the generative AI model 12 is capable of generating a large amount of waveform data for machine learning, with labels assigned to respective waveform data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a waveform generation system and method for use in test measurements, and more particularly to a waveform generation system and method for creating waveform data for machine learning in order to utilize machine learning in test measurements. [Background technology]

[0002] Using machine learning to test and measure the performance of electronic devices will greatly speed up and improve design testing and validation and manufacturing line production. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] US Patent Application Publication No. 2022 / 0373598 [Patent Document 2] US Patent Application Publication No. 2024 / 0169210 [Patent Document 3] US Patent Application Publication No. 2024 / 0393918 [Non-patent literature]

[0004] [Non-Patent Document 1] "What is RAG (Retrieval-Augmented Generation)?", ITmedia, Inc., [online], [Retrieved May 16, 2025], Internet<https: / / atmarkit.itmedia.co.jp / ait / articles / 2403 / 13 / news035.html> [Non-patent document 2] "What is a Vector Database / Vector Store?", ITmedia, Inc., [online], [Retrieved May 16, 2025], Internet<https: / / atmarkit.itmedia.co.jp / ait / articles / 2402 / 29 / news042.html> [Non-patent document 3] "Mixed Signal Oscilloscope" introduction site, Tektronix, [online], [Retrieved May 16, 2025], Internet<https: / / www.tek.com / ja / oscilloscope-mixed-signal-oscilloscope> Summary of the Invention [Problem to be solved by the invention]

[0005] In some machine learning environments, problems can arise due to the lack of large training waveform data sets associated with the standard, the various signal impairments used in the test, and the various equipment. Especially for emerging standards, the lack of training waveform data sets can be severe due to the lack or unavailability of hardware. [Means for solving the problem]

[0006] A waveform generation system according to embodiments of the present disclosure may use a generative AI model, such as a large language model (LLM), in conjunction with knowledge and capabilities related to standards, impairments, and test and measurement equipment to create a waveform file. In some examples, the present disclosure may use the generative AI model disclosed herein to create a tensor.

[0007] Some machine learning systems and processes use tensors as inputs to the machine learning system. Tensors consist of images of waveforms generated by a device under test (DUT). Tensors may also include bar graphs and other data. Examples of tensors are disclosed in, for example, U.S. Patent Application Publication No. 2022 / 0373598 (Patent Document 1), U.S. Patent Application Publication No. 2024 / 0169210 (Patent Document 2), and U.S. Patent Application Publication No. 2024 / 0393918 (Patent Document 3).

[0008] Some generative AI model providers offer large-scale language models (LLMs) for users, and some providers have begun offering AI assistant extensions. AI assistant extensions include several application programming interfaces (APIs) that allow users to access additional functionality. These extensions include prompts and files for adding additional information / knowledge to the AI ​​assistant's knowledge, as well as functions that allow the AI ​​assistant to interact with local or web resources. The ability to interact with resources is expected to form bidirectional relationships, not only requesting actions but also responding to them, a relatively new feature. Users can add files that can extend the AI ​​assistant's knowledge. Additionally, systems can connect to Retrieval Augmentation Generation (RAG) via APIs, which can extend the capabilities of generative AI models with knowledge / information about specific domains, enabling more accurate waveform generation based on that knowledge / information.

[0009] A generative AI model undergoes a form of training, where the generative AI model learns basic patterns in training data and uses this learning to generate new data, including waveform data, based on input, which in this embodiment is input from a user using an AI assistant. Search Augmentation Generation (RAG) differs from training in that the information for the RAG may originate from the generative AI model itself, which is seeking information. RAG makes domain-specific data and information (including updates) available to the generative AI model. The generative AI model can use RAG to search for information before providing a response. A system may have resources behind a firewall or otherwise controlled resources that the generative AI model can access.

[0010] Additionally, the system may connect via an API to and utilize vector databases that store information using embeddings (also called numerical vector representations). Vector databases are a key component of the RAG system, but they are separate technologies. Vector databases allow data such as images, graphics, text, and numbers to be represented as vectors that can be compared for similarity, making it possible to search for semantically related data (see Non-Patent Document 2). In RAG, embeddings are retrieved from the vector database and used to augment language models with domain-specific knowledge.

[0011] The API function allows users to access sufficient information / knowledge to describe the characteristics of a large number of waveforms. The assistant API and generative AI model then input parameters into the standard workflow for creating waveforms. In addition, the knowledge / information about waveforms entered by the user and the knowledge / information collected by the system about waveforms can be used as labels (information that represents the correct answer for waveform data). This allows labeled waveform data to be generated. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 shows an example of a system for generating waveforms for a test system. [Figure 2] FIG. 2 shows a flowchart of one embodiment of a method for generating waveforms in a test system. [Figure 3] FIG. 3 shows a flow chart of an alternative embodiment of a method for generating waveforms in a test system. DETAILED DESCRIPTION OF THE INVENTION

[0013] FIG. 1 illustrates an embodiment of an AI assistant system configuration for a testing system. A user 20 interacts with a generative model 12 through an AI assistant 10. The AI ​​assistant 10 may be implemented, for example, as a chat AI application. The system's functionality can be extended through an application programming interface (API) for the AI ​​assistant, as described above. Therefore, interaction with the AI ​​assistant 10 can be performed directly, for example, by the user 20 providing a script that invokes the AI ​​assistant 10, or through a user interface, such as a chatbot interface or other user input window that accepts prompt input. Furthermore, as described above, a RAG or vector database can be incorporated into the system via the assistant API, thereby extending the AI ​​assistant 10 to utilize domain-specific knowledge / information, thereby enabling it to generate waveforms (waveform data, or simply "waveforms") with greater accuracy.

[0014] The AI ​​Assistant 10 may have access to general test and measurement (T&M) knowledge 14. Based on user configuration, the AI ​​Assistant 10 may also have access to information 16 about waveform impairments such as jitter, noise, and the like, as well as information describing test and measurement equipment (specification descriptions) 18, as described in more detail below. Furthermore, the AI ​​Assistant 10 may have access to various protocol standards 26, such as PCI Gen X, USB Gen X, and LAN (Local Area Network). Note that "Gen X" refers to any generation. The system may also include one or more test and measurement instruments, such as a test and measurement instrument 28. Examples of test and measurement instruments include, but are not limited to, oscilloscopes, such as mixed-signal oscilloscopes (see Non-Patent Document 3), digital multimeters, and spectrum analyzers. Such test and measurement instruments may also have waveform generation capabilities, such as an arbitrary waveform generator. A test and measurement instrument may comprise any device capable of receiving signals from a device under test (DUT) and generating waveforms.

[0015] The generative AI model 12 may comprise any LLM and may be comprised of a computer program running on one or more processors, such as processor 24. Processor 24 may reside on one or more computing devices, including test and measurement equipment 28, a server operating remotely from a user, or a local computing device of user 20. The system may include any combination of the above and other computing devices. The resulting waveforms (waveform data) may be stored in store 22, which may be used to maintain a vector database.

[0016] Using embodiments of the system disclosed herein, a user 20 can request a set of waveforms in a variety of ways. Examples include, but are not limited to, the following: One example might be, "Create 10,000 PCIe Gen7 waveforms using a short stress pattern random quaternary (SSPRQ) pattern for all legal symbol rates and amplitudes. Randomly vary the time interval error (TIE) of each waveform using a normal distribution with a standard deviation (stddev) of 1 / 4 unit interval (UI). Randomly vary noise between 1% and 10%. Add the parameters used to define the waveforms to the waveform metadata." Another example might be, "Create 35,000 waveforms based on the symbols, data rates, and amplitudes derived from the waveform file 'pcie_gen7_03_30.wfm'. Randomly vary the TIE of each waveform using a normal distribution with a standard deviation of 0.1 UI (stddev=0.1UI)." "Randomly select an s4p file from the folder "cable_models" and use it as a waveform model to vary the impairments. Add the parameters used to define the waveform to the waveform's metadata." Another example might be "Randomly select an s4p file from the folder "PCIe7_Examples" and create 60,000 waveforms using their data rates, bit sequences, and amplitudes. Then vary the TIE of the waveforms so that the transition times for patterns 001 and 110 are [...] and the transition times for patterns 101 and 010 are [...]..."

[0017] Outputs of embodiments according to the present disclosure include analog time-domain waveforms, fast frames of analog data (where "fast frames" refers to a segmented memory acquisition mode designed to conserve memory when acquiring a large number of low duty cycle events), a waveform database obtained by overlaying created data sets, digital time-domain waveforms, and radio frequency (RF) IQ (in-phase and quadrature) waveforms.

[0018] User 20 may use a wide variety of parameters to define a waveform. The following discussion provides some non-limiting examples. User 20 may define a test pattern, such as a pseudorandom binary sequence (PRBS7, 9, 15, 23, etc.), a set of patterns or symbols that represent data in a pattern format, such as SSPRQ, with one or more repetitions of the pattern or symbol. User 20 may also define modulation techniques, such as quaternary pulse amplitude modulation (PAM4), non-return-to-zero (NRZ), or quadrature amplitude modulation (QAM4, QAM16). Some test patterns may indicate corresponding modulations, such as SSPRQ indicating PAM4 modulation.

[0019] Some parameters may include the bit rate or symbol rate of the pattern, the frequency (cycles per second) of the carrier of the RF signal, the signal amplitude, the signal offset, and whether the signal is differential or single-ended. For differential signals, users may also include the difference in timing between the two sides of the differential signal (skew). Parameters may also include timing error between lanes in multi-channel systems (inter-channel skew), noise, jitter such as random jitter and deterministic jitter, inter-symbol interference (ISI), and signal impairments such as bit error rate and symbol error rate.

[0020] Other parameters may include whether to use spread spectrum clocking (SSC), whether to read a file containing S-parameters, waveforms, or other data, or whether to write a file specifying waveforms or other files to be written. Parameters may include the number of waveforms to generate and statistical specifications for the occurrence rate of randomly specified events. Additionally, user 20 may specify a test and measurement instrument, such as an oscilloscope, or an arbitrary waveform generator (AWG), depending on the desired waveform. User 20 may also provide the signal power (e.g., in dB), the slope of the signal power over the duration of the RF signal, and the RF pulse repetition interval (PRI). As shown in Figure 1, the system may take into account standards such as Peripheral Component Interface (PCI) or PCI Express (PCIe), or Universal Serial Bus (USB), including the packets and fields associated with the protocols.

[0021] The user 20 can provide any of the above-described inputs to the AI ​​assistant 10. In some embodiments, the user inputs may be: Pattern definition: This may list the signal definitions such as PRBS7, PRBS10, etc. The actual value and, optionally, its units: for example, a data rate such as 2.7 GHz. In this case, the value you provide is 2.7e9, along with the unit Hertz. The count value (such as the number of times a loop operation) that an item is requested. Random selection: such as randomly selecting an option from a set of multiple objects, such as files in a folder Dependency relationships: Define the relationship between parameters (e.g., the correlation between TIE and rising edge is -0.6) Statistical description: For example, the mean value is 0.1 UI (unit interval) and the standard deviation is 0.05 UI. Statistical description involves randomly picking values ​​based on statistical specifications. If a large enough population is measured from the definition, the AI ​​assistant 10 and AI model can approximate and find values ​​based on the specific statistical definition. For example, if you request 100 waveforms with a mean data rate of 2.7 GHz and a standard deviation of 160 MHz, the AI ​​assistant 10 will generate 100 waveforms, each with a different data rate, but with a data rate that is normally distributed and centered around 2.7 GHz.

[0022] In some embodiments, templates and creation rules can be used to define more complex behavior. For example, the example prompt "100 I2C Packets for 3 10-bit devices where 1 device is not responding" might use a template for 10-bit data packets and a creation rule that specifies how to convert the data packets into "clock" and "data" analog signals. Templates and creation rules allow complex behavior to be added without requiring a lot of code. Template definitions in XML, JSON, YAML, or other description language formats allow data templates and creation rules to be defined in a generic way.

[0023] FIG. 2 shows an example workflow flowchart for generating a waveform using a system such as that shown in FIG. 1. In the embodiment of FIG. 2, the method begins with defining a pattern in step 30. The user can define the pattern according to any of the methods described above. The definition may include, for example, the data rate, bit error, and other parameters described above. The assistant API and generative AI model 12 may call functions that provide parameters related to the output waveform. If the user does not specify any specific items, the system may provide reasonable default settings. The system provides a timing definition in step 32, and then the AI ​​assistant 10 sends the pattern and definition based on the user input to the generative AI model 12 in step 34. The generative AI model 12 returns the generated waveform to the AI ​​assistant 10 in step 36.

[0024] In step 38, the user may validate the set of waveforms to ensure that they meet the specifications specified in the input to the process. For example, if the user's input to the process sets a TIE with a particular standard deviation, the entire waveform set may be evaluated to determine if the waveforms comply with that specification. After validation, the system may store the waveforms for later use in step 39 or utilize the waveforms in a test and measurement device.

[0025] After a user requests a set of waveforms, the assistant API and generative AI model 12 collects information from the user's input and generates the waveforms using the workflow of FIG. 2. In some examples, the assistant API and generative AI model 12 uses the workflow of FIG. 2 for each requested waveform. For example, if a user requests 100 waveforms, the assistant API and AI model 12 generates 100 waveforms using the workflow of FIG. 2 100 times. In the example workflow shown in FIG. 2, functions invoked by the system provide parameters related to the output waveforms. As described above, if the user does not specify any parameters, reasonable default settings are used. Furthermore, the generative AI model 12 can also adjust the generated waveforms to take into account specific characteristics, capabilities, and limitations, such as sample rate, memory depth, and cycle-related performance.

[0026] If the system uses a waveform as input, the workflow is modified as shown in Figure 3. In Figure 3, the input waveform is run through a waveform analyzer at step 40. The process then continues as described in Figure 2, defining the pattern at step 42, determining timing definitions at step 44, sending information (pattern and definition) to the generative AI model 12 at step 46, and receiving the waveform through the AI ​​assistant 10 at step 48. At step 50, the waveform is verified by the user to ensure it meets the desired specifications, and the waveform is saved or used at step 52.

[0027] The waveform analyzer uses analysis functions to determine waveform elements such as pattern, data rate, amplitude, etc. These elements are passed to AI Assistant 10 if the prompt mentions connecting to an AI Assistant. For example, a user may provide measured characteristics of an existing waveform as part of the input, and the generative AI model will provide as output a set of waveforms with these measured characteristics, with missing parameters filled in by AI Assistant 10.

[0028] The waveforms generated by a generative AI model may not simply create new waveform sets, but may also augment existing waveform sets. For example, if the current TIE for a particular set or repository of waveforms is 5 ps peak-to-peak, and a user desires a TIE of 5.6 ps peak-to-peak with a standard deviation of 3.2 ps for the entire waveform set, the generative AI model may take the current TIE measurement for the waveform set and the desired final TIE measurement and generate additional waveforms to shift the TIE measurement for the waveform set so that the final TIE measurement meets the desired value.

[0029] While waveform generation is a general task, combining AI models with domain-specific knowledge allows for the creation of large numbers of distinct, labeled waveforms for machine learning. These waveforms can be generated artificially without using a real DUT, but they may be indistinguishable from waveforms captured using an oscilloscope and a device under test (DUT). These waveforms can be particularly useful for machine learning or in situations where waveforms based on actual manufactured DUTs are not readily available, such as in the early stages of a new bus standard.

[0030] The present disclosure includes loading protocol standard knowledge and test and measurement equipment knowledge into an AI assistant, allowing users to reference key characteristics of protocol standards and signals without having to specifically program this knowledge as part of a waveform creation workflow. Furthermore, the present disclosure provides a standard workflow for generating output data, where a generative AI model can incorporate specific values ​​for one or more parameters controllable by the workflow into the output data. The present disclosure also includes a workflow that defines parameters as statistical definitions rather than specific values, such that each time the workflow is executed, a new, unique waveform is created based on the provided statistical definitions. The present disclosure also includes a process for creating, labeling, and storing multiple waveforms based on parameters defined in prompts. Knowledge / information about the waveforms entered by the user and knowledge / information collected by the system about the waveforms can be used as labels (information representing the correct waveform data). The system can then generate labels corresponding to each generated waveform data.

[0031] Aspects of the disclosed technology may operate on specially created hardware, firmware, digital signal processors, or specially programmed general-purpose computers, including processors that operate according to programmed instructions. The terms "controller" or "processor" herein contemplate microprocessors, microcomputers, ASICs, and dedicated hardware controllers, among others. Aspects of the disclosed technology may be implemented with computer-usable data and computer-executable instructions, such as one or more program modules, executed by one or more computers (including a monitoring module) or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc., which, when executed by a processor in a computer or other device, perform particular tasks or implement particular abstract data types. Computer-executable instructions may be stored in computer-readable storage media, such as hard disks, optical disks, removable storage media, solid-state memory, RAM, etc. Those skilled in the art will appreciate that the functionality of the program modules may be combined or distributed as desired in various embodiments. Furthermore, such functionality may be embodied in whole or in part in firmware or hardware equivalents, such as integrated circuits, field programmable gate arrays (FPGAs), etc. Certain data structures may be used to more effectively implement one or more aspects of the disclosed technology, and such data structures are considered within the scope of the computer-executable instructions and computer-usable data described herein.

[0032] The disclosed aspects may, in some cases, be implemented in hardware, firmware, software, or any combination thereof. The disclosed aspects may also be implemented as instructions carried by or stored on one or more computer-readable media, which may be read and executed by one or more processors. Such instructions may be referred to as a computer program product. As used herein, computer-readable media refers to any medium that can be accessed by a computing device. By way of example, and not limitation, computer-readable media may include computer storage media and communication media.

[0033] "Computer storage media" means any medium that can be used to store computer-readable information. By way of example and not limitation, computer storage media may include 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), digital video disc (DVD) and other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage and other magnetic storage devices, and any other volatile or nonvolatile, removable or non-removable medium implemented in any technology. "Computer storage media" excludes signals themselves and transitory forms of signal transmission.

[0034] A communication medium means any medium usable for communicating computer-readable information. By way of example, and not limitation, communication media may include coaxial cable, fiber optic cable, air, or any other medium suitable for communicating electrical, optical, radio frequency (RF), infrared, acoustic, or other types of signals. Example

[0035] The following examples are provided to aid in understanding the technology disclosed in this application. Embodiments of the technology may include one or more of the examples described below, and any combination thereof.

[0036] Example 1 is a waveform generation system comprising one or more memories having test and measurement knowledge, a generative artificial intelligence (AI) model that can access the one or more memories, a display, a user interface that allows a user to provide input, and one or more processors, the one or more processors being configured to execute a program that causes the one or more processors to perform the following operations: access an application programming interface (API) of an AI assistant for the generative AI model to allow the user to interact with the AI ​​assistant; receive one or more user inputs via one or more of the APIs or user interfaces that provide descriptions of one or more waveforms to be generated; use the AI ​​assistant to create waveform definitions from the descriptions and access the generative AI model; receive one or more waveforms from the AI ​​assistant; and store the one or more waveforms.

[0037] Example 2 is the waveform generation system of Example 1, wherein the one or more processors are further configured to execute a program that causes the one or more processors to perform a process of connecting via the API to a search extension generation (RAG) for accessing information outside the one or more memories.

[0038] A third embodiment is the waveform generation system of any one of the first and second embodiments, wherein the one or more memories store a corpus of test and measurement knowledge (a large database of specialized knowledge).

[0039] Example 4 is the waveform generation system of any of Examples 1 to 3, wherein the user input includes one or more of a list of signal definitions, real values, statistical descriptions, number of iterations, and option selections.

[0040] Example 5 is a waveform generation system of any of Examples 1 to 4, wherein the program that causes the one or more processors to process receiving the one or more user inputs includes a program that causes the one or more processors to process presenting a user interface having a template with associated creation rules.

[0041] Example 6 is a waveform generation system according to any one of Examples 1 to 5, wherein the program causing the one or more processors to perform a process of receiving the one or more user inputs includes a program causing the one or more processors to perform a process of providing a template and a definition of the input.

[0042] Example 7 is a waveform generation system according to any one of Examples 1 to 6, wherein the program that causes the one or more processors to perform the process of creating the waveform definition includes a program that causes the one or more processors to perform the process of using initial settings when the user input is not received.

[0043] Example 8 is a waveform generation system according to any one of Examples 1 to 7, wherein the program causing the one or more processors to perform the process of creating the waveform definition includes a program that receives measured characteristics of an existing waveform as input and causes the one or more processors to perform the process of generating a new waveform having specified characteristics.

[0044] Example 9 is a waveform generation system of Examples 1 to 8, wherein the one or more processors are further configured to execute a program that causes the one or more processors to perform a process of verifying the one or more waveforms received from the AI ​​assistant against the waveform definition.

[0045] Example 10 is a waveform generation method for generating one or more waveforms, comprising: accessing an AI assistant's application programming interface (API) for the generative artificial intelligence (AI) model to enable a user to interact with the AI ​​assistant; receiving, via one or more of said APIs or user interfaces, one or more user inputs providing a description of one or more waveforms to be generated; creating a waveform definition from the description using the AI ​​assistant and accessing the generative AI model; receiving one or more waveforms from the AI ​​assistant; a process of utilizing or storing said one or more waveforms in one or more memories; It is equipped with.

[0046] Example 11 is the waveform generation method of Example 10, further comprising utilizing Retrieval Augmentation Generation (RAG) via the API to access information external to the system in which the generative AI model resides.

[0047] Example 12 is the waveform generation method of either example 10 or 11, wherein the one or more memories store test and measurement knowledge accessible by the generative AI model.

[0048] Example 13 is the waveform generation method of any of Examples 10 to 12, wherein the user input includes one or more of a waveform definition enumeration, a real value, a statistical description, a number of iterations, and an optional selection.

[0049] Example 14 is a waveform generation method according to Examples 10 to 13, wherein the process of receiving one or more user inputs includes a process of presenting a user interface having a template with associated creation rules.

[0050] Example 15 is a waveform generation method of Examples 10 to 14, wherein the process of receiving the one or more user inputs via the user interface includes a process of providing templates and definitions for the one or more user inputs.

[0051] A sixteenth embodiment is the waveform generating method according to the tenth to fifteenth embodiments, wherein the process of creating the waveform definition includes a process of using an initial setting when the user input is not received.

[0052] Example 17 is the waveform generation method of Examples 10 to 16, wherein creating the waveform definition includes receiving measured characteristics of an existing waveform as input and generating a new waveform having the specified characteristics.

[0053] Example 18 is the waveform generation method of Examples 10 to 17, further comprising: validating the one or more waveforms received from the AI ​​assistant against the waveform definition.

[0054] Although the above-described versions of the presently disclosed subject matter have many advantages that have been described or that will be apparent to those skilled in the art, not all of these advantages or features are required in every version of the disclosed devices, systems, or methods.

[0055] All features disclosed in the specification, claims, abstract and drawings, and all steps in any disclosed method or process, may be combined in any combination, except where at least some of such features or steps are mutually exclusive combinations. Each feature disclosed in the specification, abstract, claims and drawings may be replaced by an alternative feature serving the same, equivalent or similar purpose, unless expressly stated otherwise.

[0056] Additionally, the description in this application refers to specific features. All features disclosed in this specification, including the claims, abstract, and drawings, and all steps in all disclosed methods or processes, may be combined in any combination, unless they are at least partially mutually exclusive. Each feature disclosed in this specification, including the claims, abstract, and drawings, may be replaced with an alternative feature serving the same, equivalent, or similar purpose, unless otherwise specified.

[0057] 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, unless the circumstances do not preclude this possibility.

[0058] Although specific embodiments of the invention have been illustrated and described for purposes of illustration, it will be appreciated that various modifications can be made therein without departing from the spirit and scope of the invention. Accordingly, the invention should not be limited except as by the appended claims. [Explanation of symbols]

[0059] 10 AI Assistant APIs 12 Generative AI Models 14 General Testing and Measurement Knowledge 16 Waveform Disturbance Information 18 Information describing test and measurement equipment 20 users 22 stores 24 processors 26 Various protocol standards 28 Test and measurement equipment

Claims

1. 1. A waveform generation system, comprising: one or more memories containing test and measurement knowledge; a generative artificial intelligence (AI) model having access to the one or more memories; The display and a user control that allows a user to provide input; one or more processors Equipped with The one or more processors accessing an AI assistant application programming interface (API) for the generative AI model to enable the user to interact with the AI ​​assistant; receiving one or more user inputs via one or more of said APIs or user interfaces providing a description of one or more waveforms to be generated; using the AI ​​assistant to create each waveform definition from the description and access a generative AI model; receiving one or more waveforms from the AI ​​assistant; storing said one or more waveforms; a waveform generation system configured to execute a program that causes the one or more processors to perform the above steps.

2. 2. The waveform generation system of claim 1, wherein the one or more processors are further configured to execute a program that causes the one or more processors to interface via the API with a search extension generation (RAG) to access information outside the one or more memories.

3. 3. The waveform generation system of claim 1, wherein the one or more processors are further configured to execute a program that causes the one or more processors to generate labels corresponding to each of the one or more waveforms using information collected via one or more of the API or the user interface.

4. 2. The waveform generation system of claim 1, wherein the program that causes the one or more processors to process receiving the one or more user inputs includes a program that causes the one or more processors to process presenting a user interface having templates with associated creation rules.

5. 2. The waveform generation system of claim 1, wherein the program causing the one or more processors to process the one or more user inputs includes a program causing the one or more processors to process the one or more user inputs by providing templates and definitions of the inputs.

6. 2. The waveform generation system of claim 1, wherein the program that causes the one or more processors to perform the process of creating the waveform definition includes a program that receives measured characteristics of an existing waveform as input and causes the one or more processors to perform the process of generating a new waveform having specified characteristics.

7. 2. The waveform generation system of claim 1, wherein the one or more processors are further configured to execute a program that causes the one or more processors to perform a process of verifying the one or more waveforms received from the AI ​​assistant against the waveform definition.

8. 1. A waveform generation method for generating one or more waveforms, comprising: accessing an AI assistant's application programming interface (API) for a generative artificial intelligence (AI) model to enable a user to interact with the AI ​​assistant; receiving one or more user inputs via one or more of said APIs or user interfaces providing a description of one or more waveforms to be generated; creating a waveform definition from the description using the AI ​​assistant and accessing the generative AI model; receiving one or more waveforms from the AI ​​assistant; utilizing or storing said one or more waveforms in one or more memories; A waveform generation method comprising:

9. 9. The method of claim 8, wherein receiving one or more user inputs comprises presenting a user interface having templates with associated creation rules.

10. 9. The method of claim 8, wherein the step of creating the waveform definition includes the step of utilizing a default setting if the user input is not received.

11. 9. The method of claim 8, wherein creating the waveform definition comprises receiving measured characteristics of an existing waveform as input and generating a new waveform having the specified characteristics.

Images