TWI889560B - Internet of things information security detection method and internet of things information security detection system - Google Patents

Abstract

An internet of things information security detection method and internet of things information security detection system are provided. The information security detection method includes the following steps: loading standard regulatory information, where the standard regulatory information contains multiple requirement conditions, multiple test items, and multiple compliance data; inputting specification information; using the multiple requirement conditions to compare with the specification information to output the corresponding test items; establishing a testing checklist through the outputted multiple test items; loading self-declaration information containing declaration data; generating test scenarios based on the testing checklist; inputting the declaration data into the test scenarios for issue verification to determine if there is a question-answer correlation; generating corrective feedback if the question-answer correlation does not exist; sending the corrective feedback to the correction end to modify the declaration data; and when the question-answer correlation exists, inputting the declaration data into the test scenarios for compliance verification to determine if the declaration data meets the corresponding compliance data.

Description

Internet of Things Information Security Detection Method and Internet of Things Information Security Detection System

The present invention relates to a detection method, and more particularly to an Internet of Things information security detection method and an Internet of Things information security detection system.

In order to ensure the information security of IoT devices, IoT devices will be inspected by information security regulations (such as EN303645, CNS16190). However, the existing inspection method is to select and interpret IoT devices manually to provide information security reports for IoT devices, but this is time-consuming and labor-intensive.

Therefore, the inventors of the present invention believe that the above defects can be improved, and have conducted intensive research and applied scientific principles to finally propose the present invention which has a reasonable design and effectively improves the above defects.

The technical problem to be solved by the present invention is to provide an Internet of Things information security detection method and an Internet of Things information security detection system in view of the deficiencies of the existing technology.

The present invention discloses an Internet of Things information security detection method, which is applied to a detection system to detect a device to be detected. The information security detection method includes the following steps: loading a standard regulatory information, wherein the standard regulatory information has multiple condition requirements, multiple test items corresponding to the multiple condition requirements, and multiple compliance data matching the multiple test items; inputting a specification information of the device to be detected; using the multiple condition requirements to compare the specification information to output the corresponding test items; establishing a detection list having at least one test item through the output multiple test items; loading a self-declaration information, wherein the self-declaration information has a corresponding test item in the detection list; a statement data of the test item of the test item; using a large language model (LLM) to generate a test scenario according to the detection list; using the large language model to bring the statement data into the test scenario for a question verification to determine whether there is a question-answer relationship between the statement data and the test scenario; generating a correction feedback when the question-answer relationship does not exist; transmitting the correction feedback to a correction end so that the correction end corrects the statement data; and when the question-answer relationship exists, using the large language model to bring the statement data into the test scenario for a compliance verification to determine whether the statement data in the test scenario conforms to the corresponding compliance data.

The embodiment of the present invention also discloses an Internet of Things information security detection system for detecting a device to be detected, the information security detection system comprising: a database, including standard regulatory information, the standard regulatory information having a plurality of condition requirements, a plurality of test items corresponding to the plurality of condition requirements, and a plurality of compliance data matching the plurality of test items; a front-end interface capable of inputting a self-declaration information and a specification information corresponding to the device to be detected; a server connected to the database and the front-end interface, the server using a plurality of the condition requirements to compare the specification information to output the corresponding test items to the front-end interface, so that the front-end interface can be operated to establish a detection list having at least one test item, and the self-declaration The information has a statement data corresponding to the test item in the detection list; and a large language model connected to the server, the large language model can generate a test scenario according to the detection list, and bring the statement data into the test scenario for a question verification to determine whether there is a question-answer relationship between the statement data and the test scenario; when the question-answer relationship does not exist, the large language model generates a correction feedback to be transmitted to a correction end, so that the correction end corrects the statement data; when the question-answer relationship exists, the large language model brings the statement data into the test scenario for a compliance verification to determine whether the statement data in the test scenario conforms to the corresponding compliance data.

In summary, the IoT information security detection method and IoT information security detection system disclosed in the embodiments of the present invention can "use the large language model to bring the statement data into the test scenario to perform a question verification to determine whether there is a question-answer relationship between the statement data and the test scenario" and "use the large language model to bring the statement data into the test scenario to perform a compliance verification to determine whether the statement data in the test scenario complies with the corresponding compliance data". The IoT information security detection method and IoT information security detection system can confirm whether the device to be tested complies with information security regulations without manpower.

To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only used for reference and description and are not used to limit the present invention.

The following is an explanation of the implementation of the "Internet of Things Information Security Detection Method and Internet of Things Information Security Detection System" disclosed in the present invention through specific concrete embodiments. Technical personnel in this field can understand the advantages and effects of the present invention from the content disclosed in this manual. The present invention can be implemented or applied through other different specific embodiments, and the details in this manual can also be modified and changed in various ways based on different viewpoints and applications without deviating from the concept of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted in actual size. Please note in advance. The following implementation will further explain the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

It should be understood that, although the terms "first", "second", "third", etc. may be used in this document to describe various components or signals, these components or signals should not be limited by these terms. These terms are mainly used to distinguish one component from another component, or one signal from another signal. In addition, the term "or" used in this document may include any one or more combinations of the related listed items depending on the actual situation.

In addition, in the following description, if it is indicated to refer to a specific figure or as shown in a specific figure, it is only used to emphasize that most of the related content described in the subsequent description appears in the specific figure, but it does not limit the subsequent description to only refer to the specific figure.

[First embodiment]

Referring to FIG. 1 , this embodiment provides an Internet of Things information security detection method, which is applied to a detection system (for example, the Internet of Things information security detection system of the second embodiment). The Internet of Things information security detection method can be used to detect a device to be detected (not shown), and the device to be detected is detected through a large language model (LLM) during the detection process, so as to confirm whether the device to be detected complies with information security regulations. Among them, the Internet of Things information security detection method includes steps S101 to S119 in this embodiment, but the present invention is not limited to this. In practice, the Internet of Things information security detection method can add or adjust steps as appropriate. Next, the steps of the Internet of Things information security detection method are introduced below.

Step S101: loading a standard regulatory information, wherein the standard regulatory information has a plurality of condition requirements, a plurality of test items corresponding to the plurality of condition requirements, and a plurality of compliance data matching the plurality of test items.

Among them, the standard regulatory information refers to regulatory information about information security, for example, "Cyber Security of Consumer Internet of Things: Benchmark Requirements" (No. CNS16190) published by the National Standards of the Republic of China. In addition, each of the condition requirements refers to a benchmark condition defined in the regulatory information, and each of the benchmark conditions has a corresponding test item (i.e., the test item). Each of the compliance data refers to specific data used to verify whether the test item complies with the regulations. Of course, the standard regulatory information can also be EN303645 in practice, and the present invention does not particularly limit the use of the standard regulatory information.

Step S103: Input specification information of the device to be tested. For example, the specification information may be "having identity verification mechanism".

Step S105: Use the multiple condition requirements to compare the specification information to output the corresponding test items. For example, the specification information records that the device under test has function A, function B, and function C, so the multiple test items outputted correspond to function A, function B, and function C respectively.

Step S107: Create a test list having at least one test item through the outputted multiple test items. It should be noted that the multiple test items actually include necessary test items, recommended test items, and conditional test items, so the test list may include at least one test item.

For example, the multiple test items outputted correspond to necessary function A, recommended function B, and conditional function C respectively. Therefore, there must be one test item in the inspection list that corresponds to function A, the test item that corresponds to recommended function B is selected by the personnel, and the test item that corresponds to conditional function C is selected by the system based on "conditions being met".

Step S109: Load a self-declaration information, wherein the self-declaration information has declaration data corresponding to the test item in the test list.

In practice, the supplier of the equipment to be tested will upload a self-declaration (i.e., the self-declaration information), and the self-declaration will declare each of the test items in the test list.

Step S111: Generate a test scenario according to the test list using a large language model (LLM).

Step S113: Using the large language model to bring the statement data into the test scenario for a question verification to determine whether there is a question-answer relationship between the statement data and the test scenario.

For example, if the test item is: "If a password is used and is in any state other than the factory default setting, all consumer IoT device passwords should be unique to each device or defined by the user", the large language model will be established according to the semantics of the test item as: "When a password is used and is in any state other than the factory default value (such as "admin"), all consumer IoT device passwords should meet one of the following conditions: (a) Each device is unique; (b) Defined by the user, please explain in detail how the password is established." Test scenario.

Assuming that the statement data is: "The user password is set by the user through the wizard and has no correlation with the MAC Address/SN, and cannot be reversed or brute-force calculated", the large language model can import the statement data into the test scenario to perform semantic relationship judgment, thereby determining whether the statement data responds to the question of the test scenario (i.e., the question-answer relevance).

It is worth noting that in order to ensure that the large language model can accurately perform the problem verification, the IoT information security detection method of the present invention can utilize a prompt word related to the critical perspective to require the large language model to perform the problem verification.

For example, using Python programming language for the large language model, it can be implemented in the following code: "merged_data = "Use a critical perspective to judge this test item. The question is: " + str(TestCase.question) + "\n\nThe answer is: " + str(TestCase.answer) + "\n\nReason: " + "\nResult: \nOnly answer pass or fail without other explanations"", where TestCase.question is the test scenario and TestCase.answer is the statement data.

Step S115: generating a correction feedback when the question-answer correlation does not exist.

Step S117: Send the correction feedback to a correction end so that the correction end corrects the declaration data. After the declaration data is corrected, step S113 is executed again.

For example, when the semantics of the statement data are irrelevant to the question of the test scenario (i.e., the statement data does not answer the question in relation to the test item), the correction terminal is issued, allowing the personnel to correct the statement data corresponding to the test item until the semantics of the statement data have the question-answer correlation with the question of the test scenario.

Step S119: When the question-answer correlation exists, the large language model is used to bring the statement data into the test scenario for a compliance verification to determine whether the statement data complies with the corresponding compliance data in the test scenario.

That is, when the semantics of the statement data and the question in the test scenario generate the question-answer correlation, the large language model will again judge whether the semantics of the statement data is correct in the question in the test scenario based on the compliance data.

In practice, when step S119 is completed, the IoT information security detection method can output a detection report of the device to be detected under the standard regulatory information.

[Second embodiment]

Referring to FIG. 2 , this embodiment provides an IoT information security detection system 100, which adopts or implements the IoT information security detection method of the first embodiment. Therefore, the details of each component of the IoT information security detection system 100 related to the IoT information security detection method are not particularly elaborated.

The IoT information security detection system 100 includes a database 1, a front-end interface 2, a server 3 connected to the database 1 and the front-end interface 2, and a large language model 4. Next, the components of the IoT information security detection system 100 and their connection relationships are introduced below.

The database 1 includes standard regulatory information, which has multiple condition requirements, multiple test items corresponding to the multiple condition requirements, and multiple compliance data matching the multiple test items. In practice, the standard regulatory information can be pre-stored or obtained from the cloud in real time, and the standard regulatory information is not limited to a single specific source.

The front-end interface 2 can input a self-declaration information and a specification information corresponding to the device to be tested. The front-end interface 2 can be used through a computer device or other electronic device, but the present invention is not limited thereto.

The server 3 uses a plurality of the condition requirements to match the specification information to output the corresponding test items to the front-end interface, so that the front-end interface can be operated to establish a test list having at least one test item, and the self-declaration information has declaration data corresponding to the test item in the test list.

The large language model 4 can generate a test scenario according to the test list, and bring the statement data into the test scenario for a question verification to determine whether there is a question-answer relationship between the statement data and the test scenario.

When the question-answer correlation does not exist, the large language model 4 generates a correction feedback to be transmitted to a correction end (eg, the front-end interface 2 ) so that the correction end corrects the statement data.

When the question-answer correlation exists, the large language model brings the statement data into the test scenario for compliance verification to determine whether the statement data complies with the corresponding compliance data in the test scenario.

[Technical Effects of the Embodiments of the Invention]

In summary, the IoT information security detection method and IoT information security detection system disclosed in the embodiments of the present invention can "use the large language model to bring the statement data into the test scenario to perform a question verification to determine whether there is a question-answer relationship between the statement data and the test scenario" and "use the large language model to bring the statement data into the test scenario to perform a compliance verification to determine whether the statement data in the test scenario complies with the corresponding compliance data". The IoT information security detection method and IoT information security detection system can confirm whether the device to be tested complies with information security regulations without manpower.

The contents disclosed above are only preferred feasible embodiments of the present invention and are not intended to limit the scope of the patent application of the present invention. Therefore, all equivalent technical changes made using the contents of the specification and drawings of the present invention are included in the scope of the patent application of the present invention.

S101~S119: Steps 100: IoT security detection system 1: Database 2: Front-end interface 3: Server 4: Large language model

FIG. 1 is a schematic diagram of the steps of the IoT information security detection method according to the first embodiment of the present invention.

FIG. 2 is a circuit block diagram of an IoT information security detection system according to a second embodiment of the present invention.

S101~S119: Steps

Claims (4)

A method for detecting information security of the Internet of Things is applied to a detection system to detect a device to be detected, and the method comprises the following steps: Loading a standard regulatory information, the standard regulatory information having a plurality of condition requirements, a plurality of test items corresponding to the plurality of condition requirements, and a plurality of compliance data matching the plurality of test items; Inputting a specification information of the device to be detected; Using the plurality of condition requirements to compare the specification information to output the corresponding test items; Establishing a detection list having at least one test item through the outputted plurality of test items; Loading a self-declaration information, the self-declaration information having a declaration data corresponding to the test item in the detection list; Using a large language model (LLM) to generate a test scenario according to the detection list; Using the large language model to bring the statement data into the test scenario for a question verification to determine whether there is a question-answer relationship between the statement data and the test scenario; generating a correction feedback when the question-answer relationship does not exist; transmitting the correction feedback to a correction end so that the correction end corrects the statement data; and when the question-answer relationship exists, using the large language model to bring the statement data into the test scenario for a compliance verification to determine whether the statement data in the test scenario complies with the corresponding compliance data. The Internet of Things information security detection method as described in claim 1, wherein the problem verification is performed on the large language model using a prompt word related to a critical perspective. An Internet of Things information security detection system is used to detect a device to be detected, and the information security detection system includes: A database, including a standard regulatory information, the standard regulatory information has multiple condition requirements, multiple test items corresponding to the multiple condition requirements, and multiple compliance data matching the multiple test items; A front-end interface, which can input a self-declaration information and a specification information corresponding to the device to be detected; A server, connected to the database and the front-end interface, the server uses the multiple condition requirements to compare the specification information to output the corresponding test items to the front-end interface, so that the front-end interface can be operated to establish a detection list with at least one test item, and the self-declaration information has a declaration data corresponding to the test item in the detection list; and A large language model is connected to the server. The large language model can generate a test scenario according to the test list, and bring the statement data into the test scenario for a question verification to determine whether there is a question-answer relationship between the statement data and the test scenario; When the question-answer relationship does not exist, the large language model generates a correction feedback to be transmitted to a correction end, so that the correction end corrects the statement data; When the question-answer relationship exists, the large language model brings the statement data into the test scenario for a compliance verification to determine whether the statement data in the test scenario conforms to the corresponding compliance data. An Internet of Things information security detection system as described in claim 3, wherein the large language model performs the problem verification based on a prompt word related to a critical perspective.

Images