Hardware circuit function identification method and device

Abstract

The invention discloses a hardware circuit function identification method and device, and the method comprises the steps: segmenting a to-be-identified circuit, and obtaining to-be-identified local circuits; for each to-be-identified local circuit, analyzing the to-be-identified local circuit to obtain text information of the to-be-identified local circuit; screening out candidate circuits and circuit parameter information from a vector database, wherein the similarity between the candidate circuits and the text information of the to-be-identified local circuit is greater than a similarity threshold; determining function information of the to-be-identified local circuit based on the text information of the to-be-identified local circuit, the candidate circuit and the circuit parameter information; and analyzing the to-be-identified circuit according to the function information of all the to-be-identified local circuits to obtain the circuit function of the to-be-identified circuit. The candidate circuit information with high similarity with the to-be-identified local circuit is screened out for function identification, so that the function information of the to-be-identified local circuit can be accurately extracted. And according to the function information, the function of the whole circuit is further inferred, so that accurate identification of the function of the circuit to be identified is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electronic design automation, and particularly relates to a hardware circuit function identification method and device. BACKGROUND

[0002] With the increasing complexity of electronic devices, the scale and diversity of circuit design are also rapidly growing. Traditional electronic design automation tools mainly focus on the physical implementation and simulation verification of circuits, while the ability in circuit function identification is relatively limited.

[0003] At present, the identification of circuit function still mainly depends on the experience and professional knowledge of designers. Designers usually need to manually analyze the schematic diagram and netlist file, understand the topological structure and device connection relationship of the circuit, so as to infer the function of the circuit. This process is not only inefficient, but also easily affected by human factors, leading to incorrect identification or omission of the circuit function.

[0004] Therefore, how to accurately identify the circuit function has become a problem to be solved in the field. SUMMARY

[0005] The present application provides a hardware circuit function identification method and device, aiming to accurately identify the circuit function.

[0006] In order to achieve the above purpose, the present application provides the following technical solutions:

[0007] A hardware circuit function identification method, comprising:

[0008] segmenting a to-be-identified circuit to obtain one or more to-be-identified local circuits;

[0009] For each to-be-identified local circuit, analyzing the to-be-identified local circuit to obtain text information of the to-be-identified local circuit;

[0010] screening a candidate circuit and circuit parameter information with a similarity greater than a similarity threshold from a vector database according to the text information of the to-be-identified local circuit;

[0011] determining function information of the to-be-identified local circuit based on the text information of the to-be-identified local circuit, the candidate circuit and the circuit parameter information;

[0012] analyzing the to-be-identified circuit according to the function information of all to-be-identified local circuits to obtain the circuit function of the to-be-identified circuit.

[0013] Optionally, the analyzing the to-be-identified local circuit to obtain the text information of the to-be-identified local circuit comprises:

[0014] extracting an identifier, a topology, a main chip pin connection relationship and a device parameter of the local circuit to be identified;

[0015] converting the identifier, the topology, the main chip pin connection relationship and the device parameter into initial text information;

[0016] inputting a chip document corresponding to the identifier into a multi-modal large model to obtain a pin function table, electrical characteristic parameters and function description text;

[0017] converting a chart in the chip document to obtain structured parameters;

[0018] determining the structured parameters, the pin function table, the electrical characteristic parameters and the function description text as peripheral chip analysis parameters;

[0019] converting the initial text information and the peripheral chip analysis parameters to obtain text information of the local circuit to be identified.

[0020] Optionally, before the candidate circuits and the candidate circuit parameter information with a similarity greater than a similarity threshold are filtered out from the vector database, the method further includes:

[0021] obtaining netlist information and function information of all candidate circuits; the netlist information at least includes a topology, a main chip pin connection relationship and device parameters;

[0022] inputting a complex component into a large model to obtain component text information;

[0023] determining the component text information, the netlist information and the function information as circuit parameter information;

[0024] converting the circuit parameter information into a semantic vector, and storing the all candidate circuits and the semantic vectors corresponding to the candidate circuits into a vector database.

[0025] Optionally, the candidate circuits and the circuit parameter information with a similarity greater than a similarity threshold are filtered out from the vector database, including:

[0026] converting the text information of the local circuit to be identified into a vector;

[0027] calculating a similarity between the vector and each candidate circuit;

[0028] filtering out a similarity greater than a similarity threshold from all similarities, and identifying the similarity as a target similarity;

[0029] Obtain the candidate circuit and circuit parameter information corresponding to the target similarity from the vector database.

[0030] Optionally, determining the functional information of the local circuit to be identified based on the text information of the local circuit to be identified, the candidate circuit, and the circuit parameter information includes:

[0031] Get preset prompt words;

[0032] The preset prompt words, the text information of the local circuit to be identified, the candidate circuit, and the circuit parameter information are input into the large model to obtain the functional information of the local circuit to be identified; the functional information includes the circuit function and key parameters of the local circuit to be identified.

[0033] Optional, also includes:

[0034] Based on the functional information of the local circuit to be identified, a design document is generated;

[0035] or,

[0036] An optimization scheme is generated based on the functional information of the local circuit to be identified.

[0037] A hardware circuit function identification device, comprising:

[0038] A segmentation unit is used to segment the circuit to be identified to obtain one or more local circuits to be identified.

[0039] The analysis unit is used to analyze each of the local circuits to be identified to obtain text information of the local circuits to be identified.

[0040] The filtering unit is used to filter out candidate circuits and circuit parameter information from the vector database that have a textual similarity greater than a similarity threshold with the local circuit to be identified.

[0041] The determining unit is used to determine the functional information of the local circuit to be identified based on the text information of the local circuit to be identified, the candidate circuit, and the circuit parameter information;

[0042] The functional analysis unit is used to analyze the circuit to be identified based on the functional information of all the local circuits to be identified, and to obtain the circuit function of the circuit to be identified.

[0043] Optionally, the analysis unit is specifically used for:

[0044] Extract the identifier, topology, main chip pin connection relationship and device parameters of the local circuit to be identified;

[0045] The identifier, the topology, the main chip pin connection relationship, and the device parameters are converted into initial text information.

[0046] Input the chip document corresponding to the identifier into the multimodal large model to obtain the pin function table, electrical characteristic parameters and functional description text;

[0047] The diagrams in the chip documentation are converted to obtain structured parameters;

[0048] The structured parameters, the pin function table, the electrical characteristic parameters, and the function description text are determined as peripheral chip parsing parameters;

[0049] The initial text information and the peripheral chip parsing parameters are converted to obtain the text information of the local circuit to be identified.

[0050] Optional, also includes:

[0051] The acquisition unit is used to acquire netlist information and functional information of all candidate circuits; the netlist information includes at least the topology, main chip pin connection relationship and device parameters;

[0052] The input unit is used to input complex components into the large model to obtain component text information;

[0053] An information determination unit is used to determine the component text information, the netlist information, and the functional information as circuit parameter information;

[0054] The storage unit is used to convert the circuit parameter information into semantic vectors and store all candidate circuits and their corresponding semantic vectors in the vector database.

[0055] Optionally, the filtering unit is specifically used for:

[0056] The text information of the local circuit to be identified is converted into a vector;

[0057] Calculate the similarity between the vector and each candidate circuit;

[0058] Filter out similarities greater than the similarity threshold from all similarities and mark them as target similarities;

[0059] The technical solution provided in this application obtains candidate circuits and circuit parameter information corresponding to the target similarity from a vector database. The circuit to be identified is segmented to obtain local circuits to be identified; for each local circuit, the text information of the local circuit to be identified is obtained through analysis; candidate circuits and circuit parameter information with a similarity greater than a similarity threshold to the text information of the local circuit to be identified are selected from the vector database; the functional information of the local circuit to be identified is determined based on the text information, candidate circuits, and circuit parameter information; and the circuit to be identified is analyzed based on the functional information of all local circuits to be identified to obtain the circuit function. By selecting candidate circuit information with high similarity to the local circuit to be identified and performing functional identification, the functional information of the local circuit to be identified can be accurately extracted. Based on this functional information, the function of the entire circuit can be further inferred, thereby achieving accurate identification of the function of the circuit to be identified. Attached Figure Description

[0060] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0061] Figure 1 A flowchart illustrating a hardware circuit function identification method provided in this application embodiment;

[0062] Figure 2 A schematic diagram illustrating the storage of local circuit information into a vector database, as provided in an embodiment of this application;

[0063] Figure 3 A schematic diagram of a local circuit function identification method provided in an embodiment of this application;

[0064] Figure 4 This is a schematic diagram of the architecture of a hardware circuit function identification device provided in an embodiment of this application. Detailed Implementation

[0065] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0066] In this application, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0067] like Figure 1 The flowchart shown is a hardware circuit function identification method provided in an embodiment of this application, including the following steps:

[0068] S101: Divide the circuit to be identified into one or more local circuits to be identified.

[0069] Optionally, the circuit to be identified can be segmented into one or more local circuits according to functional modules (such as power management, communication interface, etc.) using a netlist parsing algorithm or a visual segmentation model.

[0070] S102: For each local circuit to be identified, analyze the local circuit to be identified to obtain the text information of the local circuit to be identified.

[0071] The text information of the local circuit to be identified includes at least the topology text information, the main chip pin connection relationship, device parameters, and peripheral chip parsing parameters.

[0072] Optionally, in another embodiment of this application, the specific implementation of step S102 includes processes A1 to A6.

[0073] A1: Extract the identifier, topology, main chip pin connection relationship and device parameters of the local circuit to be identified.

[0074] The identifiers include at least the chip model, the topology includes at least the device connection relationships and node voltage / current constraints, and the device parameters include at least the resistance value and capacitance value.

[0075] Specifically, the pin connection relationship of the main chip refers to the connection method of each pin of the main chip (such as a microprocessor, integrated circuit, etc.) with other circuit parts in the circuit.

[0076] A2: Convert identifiers, topology, main chip pin connections, and device parameters into initial text information.

[0077] Understandably, the purpose of converting text information is to clearly display the various components, connections, functions, and control logic of the circuit, so as to facilitate understanding of the circuit's working principle, quickly identify the key functional modules of the circuit, and provide accurate information for subsequent debugging, maintenance, and replacement.

[0078] A3: Input the chip document corresponding to the identifier into the multimodal large model to obtain the pin function table, electrical characteristic parameters and function description text.

[0079] The chip documentation can be in formats including, but not limited to, PDF and image formats.

[0080] Optionally, the multimodal large model can be GPT-4V.

[0081] Specifically, the pin function table should include at least the Serial Peripheral Interface (SPI) interface definition, and the electrical characteristic parameters should include at least the SPI data length, the Most Significant Bit (MSB) configuration, and the Least Significant Bit (LSB) configuration.

[0082] Specifically, the functional description text refers to the detailed explanation and usage description of each pin or functional module in the chip documentation.

[0083] A4: Convert the diagrams in the chip documentation to obtain structured parameters.

[0084] The diagrams include, but are not limited to, pin layout diagrams and timing waveform diagrams.

[0085] Optionally, the diagrams in the chip documentation can be converted using a visual large model (e.g., OCR) and a diagram understanding module to obtain structured parameters (e.g., idle level: high level).

[0086] A5: The structured parameters, pin function table, electrical characteristic parameters, and function description text are identified as peripheral chip parsing parameters.

[0087] A6: Convert the initial text information and peripheral chip parsing parameters to obtain the text information of the local circuit to be identified.

[0088] Specifically, the initial text information and peripheral chip parsing parameters are converted into a standardized text sequence (such as JSON format or natural language description) to obtain the text information of the local circuit to be identified.

[0089] S103: Select candidate circuits and circuit parameter information from the vector database that have a similarity greater than the similarity threshold with the text information of the local circuit to be identified.

[0090] Among them, the candidate circuit is the circuit that is most similar to the local circuit to be identified.

[0091] Optionally, before step S103, candidate circuits and their circuit parameter information are stored in a vector database. The vector database can quickly find the candidate circuit most similar to the local circuit to be identified by calculating similarity. This method helps reduce human error and omissions, thereby improving the reliability of circuit identification. Therefore, another embodiment of this application provides a circuit information storage method, including processes B1 to B4.

[0092] B1: Obtain netlist and functional information for all candidate circuits.

[0093] The netlist information includes at least the topology, main chip pin connections, and device parameters.

[0094] B2: Input complex components into the large model to obtain component text information.

[0095] Among them, component text information refers to the detailed description of complex components.

[0096] B3: Determine the component text information, netlist information, and functional information as circuit parameter information.

[0097] B4: Convert circuit parameter information into semantic vectors and store all candidate circuits and their corresponding semantic vectors in the vector database.

[0098] It is understandable that encoders convert text information from various sources or in different forms into a format that computers can understand, especially by using text embedding models (such as BERT-base) to convert text into digital representations of semantic vectors, and storing these semantic vectors in a vector database for easy subsequent querying and analysis.

[0099] For a better explanation of the contents shown in processes B1 through B4, please refer to... Figure 2 The diagram illustrates a method for storing local circuit information in a vector database. First, candidate circuits (i.e., local circuits in the diagram) are analyzed to obtain their topology, main chip pin connections, device parameters, and corresponding functional information (i.e., local circuit descriptions in the diagram). Complex components are then input into a large model to obtain component text information (i.e., complex component descriptions in the diagram). The component text information, netlist information, and functional information are then used to determine circuit parameter information. This circuit parameter information is input into a text embedding model to obtain semantic vectors. Finally, all candidate circuits and their corresponding semantic vectors are stored in the vector database.

[0100] Optionally, the vector database can be Milvus.

[0101] Optionally, in another embodiment of this application, the specific implementation of step S104 includes processes C1 to C4.

[0102] C1: Convert the text information of the local circuit to be identified into a vector.

[0103] Specifically, the text information of the local circuit to be identified is input into the text embedding model to obtain the vector output by the text embedding model.

[0104] C2: Calculate the similarity between the vector and each candidate circuit.

[0105] Optionally, the cosine similarity between the vector and each candidate circuit can be calculated, or the Euclidean distance between the vector and each candidate circuit can be calculated, to obtain the similarity between the vector and each candidate circuit.

[0106] C3: Filter out similarities greater than the similarity threshold from all similarities and mark them as target similarities.

[0107] The similarity threshold can be set according to the actual situation, and no specific restrictions are imposed here.

[0108] For example, there are three similarities: first similarity, second similarity, and third similarity. The first similarity is 0.5, the second similarity is 0.7, and the third similarity is 0.9. The similarity threshold is 0.6. From all similarities, the similarities greater than the similarity threshold are selected as the second and third similarities, and the second and third similarities are identified as the target similarities.

[0109] C4: Obtain candidate circuits and circuit parameter information corresponding to the target similarity from the vector database.

[0110] It is understandable that processes C1 to C4 involve selecting the K most similar candidate circuits and their circuit parameter information from the vector database, which provides a basis for determining the circuit function of the local circuit to be identified based on these candidate circuits and their circuit parameters.

[0111] S104: Based on the text information of the local circuit to be identified, the candidate circuit and the circuit parameter information, determine the functional information of the local circuit to be identified.

[0112] It is understandable that the functional information of the local circuit to be identified is obtained by reasoning about the circuit function of the local circuit to be identified based on the text information of the local circuit to be identified, the candidate circuit, and the circuit parameter information.

[0113] Optionally, in another embodiment of this application, the specific implementation of step S104 includes processes D1 to D2.

[0114] D1: Get preset prompt words.

[0115] The preset prompts include, but are not limited to, inferring the main function of a circuit based on the following circuit parameters, identifying the type of circuit, determining its function based on the circuit connection method and parameters, and analyzing the circuit's input and output characteristics to infer its application scenarios.

[0116] D2: Input the preset prompts, the text information of the local circuit to be identified, the candidate circuit and the circuit parameter information into the large model to obtain the functional information of the local circuit to be identified.

[0117] The functional information includes the circuit functions of the local circuit to be identified (e.g., amplification, filtering, and oscillation) and key parameters (e.g., operating voltage, current, input and output signal characteristics, and gain).

[0118] Optionally, after step S104, the method further includes:

[0119] Based on the functional information of the local circuit to be identified, a design document is generated.

[0120] For example, the identified motor control circuit can automatically generate control algorithm documentation.

[0121] Alternatively, an optimization scheme can be generated based on the functional information of the local circuit to be identified.

[0122] For example, the identified low-pass filter circuit may suggest using lower-cost capacitors and inductors.

[0123] Furthermore, in the early stages of circuit design, functional identification is used to review the schematic diagram, thereby identifying potential design problems in advance. For example, the identified motor control circuit can automatically check the integrity of the bridge drive structure and the rationality of the component parameters.

[0124] Meanwhile, when programming the chip, software engineers need to have a detailed understanding of the functions and parameters of each local circuit module in the hardware schematic. This process can be greatly simplified by automatically identifying circuit functions, thus automating part or all of the software programming. For example, the SPI control logic of the main chip can be automatically generated based on the pin configuration of the driver chip's Serial Peripheral Interface (SPI).

[0125] S105: Analyze the circuit to be identified based on the functional information of all the local circuits to be identified, and obtain the circuit function of the circuit to be identified.

[0126] Understandably, by analyzing the functional information of each local circuit module of the circuit to be identified, the overall function of the entire circuit can be inferred. Each local circuit (such as an amplifier, filter, power supply, etc.) has its own unique function, and by combining this functional information, the overall function of the entire circuit can be determined.

[0127] In order to better understand Figure 1 The content shown is explained below. See also: Figure 3 The diagram illustrates a method for identifying the function of a local circuit. First, the circuit to be identified is split into local circuits to be identified. Then, the circuit to be identified is analyzed to obtain text information of the local circuits to be identified. Candidate circuits and circuit parameter information corresponding to the text information of the local circuits to be identified are selected from a vector database. The text information, candidate circuits, and circuit parameter information of the local circuits to be identified are input into a large model to obtain the functional information of the local circuits to be identified. Based on the functional information of the local circuits to be identified, automatic circuit review, automatic code writing, automatic document writing, and optimization suggestion generation can be performed.

[0128] In summary, by selecting candidate circuits with high similarity to the local circuit to be identified and performing functional identification, the functional information of the local circuit to be identified can be accurately extracted. Based on this functional information, the function of the entire circuit can be further inferred, thereby achieving accurate identification of the function of the circuit to be identified.

[0129] like Figure 4 The diagram shown is an architectural schematic of a hardware circuit function identification device provided in an embodiment of this application. The function identification device includes: a segmentation unit 100, an analysis unit 200, a filtering unit 300, a determination unit 400, and a function analysis unit 500.

[0130] The segmentation unit 100 is used to segment the circuit to be identified to obtain one or more local circuits to be identified.

[0131] The analysis unit 200 is used to analyze each local circuit to be identified and obtain the text information of the local circuit to be identified.

[0132] The analysis unit 200 is specifically used for: extracting the identifier, topology, main chip pin connection relationship, and device parameters of the local circuit to be identified; converting the identifier, topology, main chip pin connection relationship, and device parameters into initial text information; inputting the chip document corresponding to the identifier into the multimodal large model to obtain the pin function table, electrical characteristic parameters, and functional description text; converting the diagrams in the chip document to obtain structured parameters; determining the structured parameters, pin function table, electrical characteristic parameters, and functional description text as peripheral chip parsing parameters; and converting the initial text information and peripheral chip parsing parameters to obtain the text information of the local circuit to be identified.

[0133] The filtering unit 300 is used to filter out candidate circuits and circuit parameter information from the vector database that have a similarity greater than a similarity threshold with the text information of the local circuit to be identified.

[0134] The filtering unit 300 is specifically used for: converting the text information of the local circuit to be identified into a vector; calculating the similarity between the vector and each candidate circuit; filtering out similarities greater than the similarity threshold from all similarities and marking them as target similarities; and obtaining the candidate circuit and circuit parameter information corresponding to the target similarity from the vector database.

[0135] The determining unit 400 is used to determine the functional information of the local circuit to be identified based on the text information of the local circuit to be identified, the candidate circuit and the circuit parameter information.

[0136] The determining unit 400 is specifically used for: obtaining preset prompt words; inputting the preset prompt words, text information of the local circuit to be identified, candidate circuits and circuit parameter information into the large model to obtain the functional information of the local circuit to be identified; the functional information includes the circuit function and key parameters of the local circuit to be identified.

[0137] The functional analysis unit 500 is used to analyze the circuit to be identified based on the functional information of all the local circuits to be identified, and to obtain the circuit function of the circuit to be identified.

[0138] In summary, by converting local circuits and chip manuals into structured text information and using text embedding models for similarity matching in a vector database, accurate identification of circuit functions can be achieved. This method not only improves the efficiency of circuit analysis but also provides strong support for subsequent processes such as review and documentation.

[0139] Combination Figure 4 The device, as shown, also includes:

[0140] The acquisition unit is used to acquire netlist information and functional information of all candidate circuits; the netlist information includes at least the topology, main chip pin connection relationship and device parameters.

[0141] The input unit is used to input complex components into the large model to obtain component text information.

[0142] The information determination unit is used to determine the component text information, netlist information and functional information as circuit parameter information.

[0143] The storage unit is used to convert circuit parameter information into semantic vectors and store all candidate circuits and their corresponding semantic vectors in the vector database.

[0144] CombinationFigure 4 The device, as shown, also includes:

[0145] The first generation unit is used to generate a design document based on the functional information of the local circuit to be identified.

[0146] Alternatively, the second generation unit is used to generate an optimized scheme based on the functional information of the local circuit to be identified.

[0147] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. In particular, for system or system embodiments, since they are fundamentally similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. Components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0148] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0149] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying the function of a hardware circuit, characterized in that, include: The circuit to be identified is divided into one or more local circuits to be identified. For each of the local circuits to be identified, the local circuits to be identified are analyzed to obtain the text information of the local circuits to be identified; Candidate circuits and circuit parameter information with a textual similarity greater than a similarity threshold to the local circuit to be identified are selected from the vector database. Based on the text information of the local circuit to be identified, the candidate circuit, and the circuit parameter information, the functional information of the local circuit to be identified is determined; The circuit to be identified is analyzed based on the functional information of all the local circuits to be identified, and the circuit function of the circuit to be identified is obtained.

2. The method according to claim 1, characterized in that, The analysis of the local circuit to be identified to obtain its text information includes: Extract the identifier, topology, main chip pin connection relationship and device parameters of the local circuit to be identified; The identifier, the topology, the main chip pin connection relationship, and the device parameters are converted into initial text information. Input the chip document corresponding to the identifier into the multimodal large model to obtain the pin function table, electrical characteristic parameters and functional description text; The diagrams in the chip documentation are converted to obtain structured parameters; The structured parameters, the pin function table, the electrical characteristic parameters, and the function description text are determined as peripheral chip parsing parameters; The initial text information and the peripheral chip parsing parameters are converted to obtain the text information of the local circuit to be identified.

3. The method according to claim 1, characterized in that, Before selecting candidate circuits and candidate circuit parameter information from the vector database that have a textual similarity greater than a similarity threshold with the local circuit to be identified, the process further includes: Obtain netlist and functional information for all candidate circuits; the netlist information includes at least the topology, main chip pin connections, and device parameters. Complex components are input into a large model to obtain component text information; The component text information, the netlist information, and the functional information are determined as circuit parameter information; The circuit parameter information is converted into semantic vectors, and all candidate circuits and their corresponding semantic vectors are stored in a vector database.

4. The method according to claim 1, characterized in that, The step of filtering candidate circuits and circuit parameter information from the vector database that have a textual similarity greater than a similarity threshold with the local circuit to be identified includes: The text information of the local circuit to be identified is converted into a vector; Calculate the similarity between the vector and each candidate circuit; Filter out similarities greater than the similarity threshold from all similarities and mark them as target similarities; Obtain the candidate circuit and circuit parameter information corresponding to the target similarity from the vector database.

5. The method according to claim 1, characterized in that, The step of determining the functional information of the local circuit to be identified based on the text information of the local circuit to be identified, the candidate circuit, and the circuit parameter information includes: Get preset prompt words; The preset prompt words, the text information of the local circuit to be identified, the candidate circuit, and the circuit parameter information are input into the large model to obtain the functional information of the local circuit to be identified; the functional information includes the circuit function and key parameters of the local circuit to be identified.

6. The method according to claim 1, characterized in that, Also includes: Based on the functional information of the local circuit to be identified, a design document is generated; or, An optimization scheme is generated based on the functional information of the local circuit to be identified.

7. A hardware circuit function identification device, characterized in that, include: A segmentation unit is used to segment the circuit to be identified to obtain one or more local circuits to be identified. The analysis unit is used to analyze each of the local circuits to be identified to obtain text information of the local circuits to be identified. The filtering unit is used to filter out candidate circuits and circuit parameter information from the vector database that have a textual similarity greater than a similarity threshold with the local circuit to be identified. The determining unit is used to determine the functional information of the local circuit to be identified based on the text information of the local circuit to be identified, the candidate circuit, and the circuit parameter information; The functional analysis unit is used to analyze the circuit to be identified based on the functional information of all the local circuits to be identified, and to obtain the circuit function of the circuit to be identified.

8. The apparatus according to claim 7, characterized in that, The analysis unit is specifically used for: Extract the identifier, topology, main chip pin connection relationship and device parameters of the local circuit to be identified; The identifier, the topology, the main chip pin connection relationship, and the device parameters are converted into initial text information. Input the chip document corresponding to the identifier into the multimodal large model to obtain the pin function table, electrical characteristic parameters and functional description text; The diagrams in the chip documentation are converted to obtain structured parameters; The structured parameters, the pin function table, the electrical characteristic parameters, and the function description text are determined as peripheral chip parsing parameters; The initial text information and the peripheral chip parsing parameters are converted to obtain the text information of the local circuit to be identified.

9. The apparatus according to claim 7, characterized in that, Also includes: The acquisition unit is used to acquire netlist information and functional information of all candidate circuits; the netlist information includes at least the topology, main chip pin connection relationship and device parameters; The input unit is used to input complex components into the large model to obtain component text information; An information determination unit is used to determine the component text information, the netlist information, and the functional information as circuit parameter information; The storage unit is used to convert the circuit parameter information into semantic vectors and store all candidate circuits and their corresponding semantic vectors in the vector database.

10. The apparatus according to claim 7, characterized in that, The filtering unit is specifically used for: The text information of the local circuit to be identified is converted into a vector; Calculate the similarity between the vector and each candidate circuit; Filter out similarities greater than the similarity threshold from all similarities and mark them as target similarities; Obtain the candidate circuit and circuit parameter information corresponding to the target similarity from the vector database.

Images