Self-description southbound acquisition method based on global Internet of Things platform

Abstract

The invention discloses a self-description southbound acquisition method based on a global Internet of Things platform, and the method comprises the steps: obtaining original physical feature parameters based on an interface protocol of an edge layer sensor, and constructing a feature vector of the sensor; converting the feature vector into initialization metadata which can be identified by a gateway terminal through a physical feature extraction module; the gateway terminal obtains an object model of the global Internet of Things platform, performs multi-modal binding on the feature vector and the object model, and updates the initialized metadata to obtain standardized metadata; a self-description coding and decoding rule is constructed, and the gateway terminal performs coding and decoding on the original data collected by the sensor based on the coding and decoding rule in combination with the standardized metadata; decoded original data is subjected to structured storage and protocol packaging, and is uploaded to a global Internet of Things platform through a gateway terminal, physical characteristic parameters are converted into metadata which is convenient to identify, and then dynamic adaptation from southbound original data to a global Internet of Things platform model is realized based on a constructed self-description coding and decoding rule.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a self-describing southbound data acquisition method based on a global Internet of Things (IoT) platform. Background Technology

[0002] With the rapid development of IoT technology, the global IoT platform, as the core hub for the access and management of massive heterogeneous devices, places higher demands on the flexibility and adaptability of southbound data acquisition. Gateway terminals, acting as the data hub between the southbound sensor cluster and the northbound global IoT platform, often face high concurrency and multimodal acquisition pressure. In a typical deployment environment, a single gateway needs to simultaneously connect to hundreds of heterogeneous sensor nodes. These nodes not only have different communication protocols, but their data acquisition needs also exhibit significant differentiation: on the one hand, different sensors need to be periodically polled at specific frequencies; on the other hand, a single sensor may contain dozens of data items, requiring differentiated parsing and preprocessing.

[0003] Traditional gateway terminals generally suffer from problems such as complex protocol adaptation, fixed data parsing, and limited acquisition strategies when dealing with multi-protocol and multi-type sensor access scenarios. In particular, for southbound protocols widely used in the Industrial Internet of Things (IIoT) such as Modbus and DLT645-2007, existing solutions use static task scheduling mechanisms and hard-coded methods to implement data acquisition and protocol parsing logic. This results in the need to redevelop drivers, adjust communication parameters, reconstruct uploaded information, and adapt data formats for new sensor devices, leading to prominent problems such as long development cycles, high maintenance costs, and limited system scalability. Summary of the Invention

[0004] In view of this, the present invention proposes a self-describing southbound acquisition method based on a global IoT platform, which can effectively solve the problems of protocol fragmentation and single acquisition strategy in large-scale heterogeneous sensor access scenarios, and significantly improve the device compatibility and deployment efficiency of the IoT edge layer.

[0005] The technical solution of this invention is implemented as follows: A self-describing southbound data acquisition method based on a global IoT platform includes the following steps: Step S1: Based on the interface protocol of the edge layer sensor, obtain the original physical feature parameters and construct the feature vector of the sensor; Step S2: Convert the feature vector into initialization metadata that can be recognized by the gateway terminal through the physical feature extraction module; Step S3: The gateway terminal obtains the object model of the full-domain IoT platform, binds the feature vector to the object model in a multimodal manner, updates the initial metadata, and obtains standardized metadata. Step S4: Construct self-describing encoding and decoding rules. The gateway terminal encodes and decodes the raw data collected by the sensor based on the encoding and decoding rules and standardized metadata. Step S5: The decoded raw data is stored in a structured manner and encapsulated in a protocol, and then sent to the global IoT platform through the gateway terminal.

[0006] Preferably, the feature vector in step S1 The expression is: ; in Let i be the feature vector of the i-th sensor. A unique identifier for the sensor. For physical port mapping, For specification type, For communication address, For communication baud rate parameters, For sensor object model, The sensor's self-describing information set, i.e., the expression is: ,in For data identification, For data accuracy, For data length, For the sampling frequency, This is a set of platform attributes.

[0007] Preferably, the specific steps of step S2 are as follows: Perform validity checks on the feature vectors; Will Combined into a communication configuration object; according to Find the predefined object model configurations on the global IoT platform; Will The sub-attributes in the data are transformed into structured data; Generate initialization metadata that can be recognized by the gateway terminal. .

[0008] Preferably, the steps for validating the feature vector are as follows: check Is it an integer and within the range {1-n}? check Is it one of RS485-1 to RS485-4? check Is it Modbus or DLT645-2007? check Is it a 12-digit integer? check Is it 2400 or 9600? check Does it belong to the set of object models defined by the global IoT platform? check Check whether the sub-attributes are complete and conform to the format.

[0009] Preferably, the initialization metadata The expression is: ; in For physical feature extraction module, Let be the feature vector of the nth sensor.

[0010] Preferably, the specific steps of step S3 are as follows: The gateway terminal obtains the object model of the global IoT platform through MQTT topic subscription; Parse the JSON structure of the object model in the global IoT platform to extract the attribute list, data type, and constraints; Data identification Multimodal binding is performed with the object model attributes of the global IoT platform, and the data identification binding information is updated in the initial metadata to obtain standardized metadata.

[0011] The preferred multimodal binding relationship is: ; in Bind information to the j-th data identifier of the i-th sensor. and These are different binding modes, for Mapping functions in the pattern for Mapping function in pattern, pattern Map a single data identifier to a single attribute, schema Aggregate multiple data items into a composite attribute.

[0012] Preferably, the specific steps in step S4 where the gateway terminal encodes the raw data collected by the sensor according to the encoding rules are as follows: Gateway terminal according to encoding rules Generate self-describing protocol frames and identify the data Fill into self-describing protocol frames The position of D in the middle, where H is the frame header, A is the device address, and is the communication address of the sensor. FC stands for function code, D for data field, CS for checksum, which is the sum of all data from H in the frame header to CS, and T for frame trailer. Self-describing protocol frames The data is transmitted to the sensor, which initiates a data acquisition request. The sensor collects the actual data, encapsulates the collected raw data into a response frame, and sends it back to the gateway terminal.

[0013] Preferably, the gateway terminal follows the decoding rules. The length of the data in the function code FC and the self-describing information set self is determined by the function code FC. The data field D of the response frame is parsed to obtain identifiable raw data.

[0014] Preferably, the data sent to the global IoT platform is: ; in Let j be the collected value of the j-th attribute of the i-th sensor. for The corresponding specific data.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention discloses a self-describing southbound data acquisition method based on a global IoT platform. After acquiring the original physical characteristic parameters of the sensor, a feature vector is constructed. The feature vector includes port mapping, protocol type, communication address, etc. Through the constructed physical feature extraction module, the feature vector can be converted into identifiable initialization metadata. The gateway terminal can bind the feature vector to the object model based on the object model of the global IoT platform in multiple modes. At the same time, the binding result will update the initialization metadata, thereby obtaining standardized metadata that is easy to process and realizing data unification. Then, self-describing encoding and decoding rules are constructed to encode standardized metadata into self-describing protocol frames. Acquisition commands can then be sent to sensors. After the sensors acquire the raw data, they process it to obtain response frames. The gateway terminal can decode the response frames to obtain identifiable raw data. Finally, the raw data can be cleaned, stored, and encapsulated before being sent to the global IoT platform. Based on the innovative self-describing encoding and decoding rules, dynamic adaptation of southbound sensor data to northbound global IoT platform object models can be achieved. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only preferred embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a self-describing southward data acquisition method based on a global Internet of Things platform according to the present invention. Detailed Implementation

[0018] To better understand the technical content of this invention, a specific embodiment is provided below, and the invention will be further described in conjunction with the accompanying drawings.

[0019] See Figure 1 The present invention provides a self-describing southbound data acquisition method based on a global IoT platform, comprising the following steps: Step S1: Based on the interface protocol of the edge layer sensor, obtain the original physical feature parameters and construct the feature vector of the sensor; Step S2: Convert the feature vector into initialization metadata that can be recognized by the gateway terminal through the physical feature extraction module; Step S3: The gateway terminal obtains the object model of the full-domain IoT platform, binds the feature vector to the object model in a multimodal manner, updates the initial metadata, and obtains standardized metadata. Step S4: Construct self-describing encoding and decoding rules. The gateway terminal encodes and decodes the raw data collected by the sensor based on the encoding and decoding rules and standardized metadata. Step S5: The decoded raw data is stored in a structured manner and encapsulated in a protocol, and then sent to the global IoT platform through the gateway terminal.

[0020] The southbound sensor cluster and the northbound global IoT platform are connected via a gateway terminal to transmit data. However, due to the significant differences in the number and types of southbound sensors, not only are the communication protocols different, but the data acquisition also exhibits significant differentiation. Therefore, it is necessary to address the issues of protocol fragmentation and a single acquisition strategy under large-scale abnormal sensor access. In the self-describing southbound acquisition method based on the global IoT platform of this invention, the original physical feature parameters, including sensor unique identifiers, physical port mappings, and protocol types, are first obtained according to the interface protocol of the southbound edge layer sensors. The obtained parameters are directly constructed into the sensor feature vector. Then, the constructed physical feature extraction module can convert the feature vector into initialization metadata that can be recognized by the gateway terminal. The gateway terminal then maps the object model and feature vector of the global IoT platform through multimodal binding to update the initialization metadata. Finally, standardized metadata can be obtained, which facilitates subsequent data processing.

[0021] After obtaining standardized metadata, based on the constructed self-describing encoding and decoding rules, the gateway terminal first encodes the data according to the encoding rules and generates acquisition instructions, which are then sent to the sensors. The sensors can collect raw data, process it, and send it back to the gateway terminal. The gateway terminal can then decode the data sent back by the sensors based on the decoding rules to obtain the raw data from the sensors. Finally, after cleaning, storing, and encapsulating the raw data, it can be uploaded to the global IoT platform, realizing dynamic adaptation of the raw data from southbound sensors to the object model of the northbound global IoT platform. Through the physical feature extraction module, the physical feature parameters of the sensors are converted into metadata that the gateway terminal can recognize, realizing multimodal data fusion of "device-protocol-data-service", improving data interoperability and business efficiency. The proposed self-describing encoding and decoding rules can support rapid adaptation of different versions or variants of the same protocol family without modifying the core code by modifying variable data and implementing a decoupled design of the protocol. When new devices are connected, the configuration development time for new device access is significantly shortened compared to the traditional hard-coding mode. Operation and maintenance personnel can complete operations such as device replacement and acquisition strategy adjustment by maintaining the feature vectors of the sensors, significantly reducing the dependence on professional protocol developers.

[0022] Preferably, the feature vector in step S1 The expression is: ; in Let i be the feature vector of the i-th sensor. A unique identifier for the sensor. For physical port mapping, For specification type, For communication address, For communication baud rate parameters, For sensor object model, The sensor's self-describing information set, i.e., the expression is: ,in For data identification, For data accuracy, For data length, For the sampling frequency, This is a set of platform attributes.

[0023] The original physical characteristic parameters include the sensor's unique identifier, physical port mapping, protocol type, communication address, communication baud rate parameter, sensor object model (adapting to the sensor object model name defined by the global IoT platform), and sensor self-description information set. Then, a feature vector can be constructed based on these original physical characteristic parameters. .

[0024] Preferably, the specific steps of step S2 are as follows: Perform validity checks on the feature vectors; Will Combined into a communication configuration object; according to Find the predefined object model configurations on the global IoT platform; Will The sub-attributes in the data are transformed into structured data; Generate initialization metadata that can be recognized by the gateway terminal. .

[0025] The purpose of validating the feature vector is to ensure that it conforms to predefined rules. After the validity check is passed, the valid parameters need to be converted into a standardized format that the gateway terminal can recognize. This mainly includes three steps: The first step is to map communication parameters: The four parameters are combined to form a communication configuration object; The second step is object model mapping: based on Search for predefined object model configurations in the global IoT voucher; The third step is self-describing information processing: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] The sub-attributes in the data are transformed into structured data; Finally, the above transformation results can be combined to form recognizable initialization metadata.

[0026] Preferably, the steps for validating the feature vector are as follows: check Is it an integer and within the range {1-n}? check Is it one of RS485-1 to RS485-4? check Is it Modbus or DLT645-2007? check Is it a 12-digit integer (padded with zeros if necessary, e.g., 000000000001)? check Is it 2400 or 9600? check Does it belong to the set of object models defined by the global IoT platform? check Check whether the sub-attributes are complete and conform to the format.

[0027] Preferably, the initialization metadata The expression is: ; in For physical feature extraction module, Let be the feature vector of the nth sensor.

[0028] The physical feature extraction module converts the sensor's feature vectors into initial metadata, establishes digital identity information for each connected sensor, and realizes the self-description of the sensor's physical features.

[0029] Preferably, the specific steps of step S3 are as follows: The gateway terminal obtains the object model of the global IoT platform through MQTT topic subscription and receives updates to the object model definition; Parse the JSON structure of the object model in the global IoT platform to extract the attribute list, data type, and constraints; Data identification Multimodal binding is performed with the object model attributes of the global IoT platform, and the data identification binding information is updated in the initial metadata to obtain standardized metadata.

[0030] The attribute list, for example, tgUa (phase A voltage), and data type, for example, float, are used. The sensor data identifier format is standardized, for example, hexadecimal code 0x0201FF00, clearly defining the physical quantity or data type it represents for easier subsequent mapping. Based on the above-mentioned construct model mapping mechanism, sensor data identifiers and object model attributes are multimodally bound. The multimodal binding relationship is as follows: ; in Bind information to the j-th data identifier of the i-th sensor. and These are different binding modes, for Mapping functions in the pattern for Mapping function in pattern, pattern To establish a single data identifier mapping to a single attribute (e.g., 0x02010100 → tgUa), direct association is required without calculation, and the pattern is as follows. To aggregate multiple data items into a composite attribute (e.g., 0x0201FF00 → [tgUa, tgUb, tgUc]), the aggregation method is explicitly set to split data block mode, with its contents representing the three-phase voltages A, B, and C in sequence. Each data element is sequentially represented as an attribute name bound to a data block.

[0031] Preferably, the specific steps in step S4 where the gateway terminal encodes the raw data collected by the sensor according to the encoding rules are as follows: Gateway terminal according to encoding rules Generate self-describing protocol frames and identify the data Fill into self-describing protocol frames The position of D in the middle, where H is the frame header, A is the device address, and is the communication address of the sensor. FC stands for function code, D for data field, CS for checksum, which is the sum of all data from H in the frame header to CS, and T for frame trailer. Self-describing protocol frames The data is transmitted to the sensor, which initiates a data acquisition request. The sensor collects the actual data, encapsulates the collected raw data into a response frame, and sends it back to the gateway terminal.

[0032] This invention proposes a self-describing encoding and decoding rule based on Modbus and DLT645-2007 protocol data units, wherein the sensor's self-describing information set... Data identifiers are defined in the document. (e.g., 0x0201FF00 corresponds to the three-phase voltages tgUa, tgUb, and tgUc of ABC), the gateway terminal generates a self-describing protocol frame according to the encoding rules. Then, based on the data identifier The predefined location mapping relationship with data field D will identify the data. Fill into self-describing protocol frames The corresponding position in data field D is used, and other sub-segments undergo table conversion processing. For example, the frame header (H = 0x68) and frame trailer (T = 0x16) are fixed values, and the device address A is set as the sensor's communication address. The function code FC indicates the operation type (e.g., FC=0x11 indicates reading data).

[0033] Based on data identification After modifying data field D, the data acquisition and encoding are completed, and the self-describing protocol frame at this point... It can transmit data to the sensor, which can receive the data acquisition command, collect the actual data, and encapsulate the raw data into a response frame. This response frame contains a self-describing protocol frame. The structures are basically the same, the difference lies in the content of the data field D. The data field D of the response frame contains the raw data of the sensor, which is decoded by the gateway terminal into object model attribute values.

[0034] Preferably, the gateway terminal follows the decoding rules. The length of the data in the function code FC and the self-describing information set self is determined by the function code FC. The data field D of the response frame is parsed to obtain identifiable raw data.

[0035] During decoding, the data results Function codes (FC) and self-describing information sets can be used. Data length defined in Map the data field D to the data result (e.g., FC = 0x11, which means reading data and extracting the value at the corresponding position from D); The mode directly reads the data value of D (tgUa = D[0,2]) and converts it into a physical quantity; Pattern based on data length Parse the composite data values ​​of the data block type (tgUa = D[0,2], tgUb = D[2,4], tgUc = D[4,6]) to complete the data decoding.

[0036] Preferably, the data sent to the global IoT platform is: ; in Let j be the collected value of the j-th attribute of the i-th sensor. for The corresponding specific data.

[0037] Based on the attribute information bound to the object model, the standardized and valid data is subjected to structured storage and protocol encapsulation, and finally the data is reliably transmitted through the gateway terminal.

[0038] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A self-describing southbound data acquisition method based on a global IoT platform, characterized in that, Includes the following steps: Step S1: Based on the interface protocol of the edge layer sensor, obtain the original physical feature parameters and construct the feature vector of the sensor; Step S2: Convert the feature vector into initialization metadata that can be recognized by the gateway terminal through the physical feature extraction module; Step S3: The gateway terminal obtains the object model of the full-domain IoT platform, binds the feature vector to the object model in a multimodal manner, updates the initial metadata, and obtains standardized metadata. Step S4: Construct self-describing encoding and decoding rules. The gateway terminal encodes and decodes the raw data collected by the sensor based on the encoding and decoding rules and standardized metadata. Step S5: The decoded raw data is stored in a structured manner and encapsulated in a protocol, and then sent to the global IoT platform through the gateway terminal.

2. The self-describing southbound data acquisition method based on a global IoT platform according to claim 1, characterized in that, The feature vector in step S1 The expression is: ； in Let i be the feature vector of the i-th sensor. A unique identifier for the sensor. For physical port mapping, For specification type, For communication address, For communication baud rate parameters, For sensor object model, The sensor's self-describing information set, i.e., the expression is: ,in For data identification, For data accuracy, For data length, For the sampling frequency, This is a set of platform attributes.

3. The self-describing southbound data acquisition method based on a global IoT platform according to claim 2, characterized in that, The specific steps of step S2 are as follows: Perform validity checks on the feature vectors; Will Combined into a communication configuration object; according to Find the predefined object model configurations on the global IoT platform; Will The sub-attributes in the data are transformed into structured data; Generate initialization metadata that can be recognized by the gateway terminal. .

4. The self-describing southbound data acquisition method based on a global IoT platform according to claim 4, characterized in that, The specific steps for validating the feature vector are as follows: check Is it an integer and within the range {1-n}? check Is it one of RS485-1 to RS485-4? check Is it Modbus or DLT645-2007? check Is it a 12-digit integer? check Is it 2400 or 9600? check Does it belong to the set of object models defined by the global IoT platform? check Check whether the sub-attributes are complete and conform to the format.

5. The self-describing southbound data acquisition method based on a global IoT platform according to claim 4, characterized in that, The initialization metadata The expression is: ； in For physical feature extraction module, Let be the feature vector of the nth sensor.

6. The self-describing southbound data acquisition method based on a global IoT platform according to claim 2, characterized in that, The specific steps of step S3 are as follows: The gateway terminal obtains the object model of the global IoT platform through MQTT topic subscription; Parse the JSON structure of the object model in the global IoT platform to extract the attribute list, data type, and constraints; Data identification Multimodal binding is performed with the object model attributes of the global IoT platform, and the data identification binding information is updated in the initial metadata to obtain standardized metadata.

7. The self-describing southbound data acquisition method based on a global IoT platform according to claim 6, characterized in that, The relationship for multimodal binding is: ； in Bind information to the j-th data identifier of the i-th sensor. and These are different binding modes, for Mapping functions in the pattern for Mapping function in pattern, pattern Map a single data identifier to a single attribute, schema Aggregate multiple data items into a composite attribute.

8. The self-describing southbound data acquisition method based on a global IoT platform according to claim 2, characterized in that, The specific steps in step S4 where the gateway terminal encodes the raw data collected by the sensor according to the encoding rules are as follows: Gateway terminal according to encoding rules Generate self-describing protocol frames and identify the data Fill into self-describing protocol frames The position of D in the middle, where H is the frame header, A is the device address, and is the communication address of the sensor. FC stands for function code, D for data field, CS for checksum, which is the sum of all data from H in the frame header to CS, and T for frame trailer. Self-describing protocol frames The data is transmitted to the sensor, which initiates a data acquisition request. The sensor collects the actual data, encapsulates the collected raw data into a response frame, and sends it back to the gateway terminal.

9. A self-describing southbound data acquisition method based on a global IoT platform according to claim 8, characterized in that, The gateway terminal follows the decoding rules. The length of the data in the function code FC and the self-describing information set self is determined by the function code FC. The data field D of the response frame is parsed to obtain identifiable raw data.

10. A self-describing southbound data acquisition method based on a global IoT platform according to claim 2, characterized in that, The data sent to the global IoT platform is: ； in Let j be the collected value of the j-th attribute of the i-th sensor. for The corresponding specific data.

Images