Intelligent edge computing device and implementation method of communication connection thereof

Abstract

The invention discloses intelligent edge computing equipment and a communication connection implementation method thereof, and relates to the technical field of equipment communication connection. According to the intelligent edge computing device and the implementation method of the communication connection thereof, the operation data and the to-be-transmitted data of the intelligent edge computing device are obtained through the sensor and the device interface, the calculation task is dynamically allocated by adopting the weighted polling scheduling algorithm, and the to-be-transmitted data is fragmented through the Hash fragmentation algorithm, so that the data transmission efficiency is improved. According to the method and the system, the computing task distribution sequence and the fragmented data set are obtained, the communication protocol of the fragmented data set is matched with the interface type through a dynamic protocol mapping algorithm, and protocol conversion and interface adaptation are carried out, so that the problem of data exchange compatibility among different systems is solved. And monitoring the protocol conversion and interface adaptation process in real time, and adjusting the hardware resource configuration and network setting of the intelligent edge computing equipment by using a fuzzy control algorithm. And through dynamic adjustment, the stability and performance of the system under different load conditions are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of device communication connection, and specifically to an intelligent edge computing device and a method for implementing a communication connection thereof. Background Art

[0002] With the rapid development of the Internet of Things, edge computing technology has gradually occupied an important position in various application scenarios. As an intermediate node connecting the device layer and the cloud layer, the edge computing gateway is responsible for the initial processing and transmission of data close to the data source, which greatly reduces network latency and reduces the pressure on cloud computing and storage. Traditional IoT architecture usually relies on uploading all data to the cloud for processing, but with the rapid increase in the number of IoT devices, this centralized processing method has exposed some technical defects.

[0003] As the number of access devices increases, the amount of data information that the gateway needs to process and transmit also increases. The computing and storage capabilities of existing gateways often cannot meet the needs of high-concurrency data processing, resulting in data delays or loss. There are many types of IoT devices, and the communication protocols and interface types used by various devices are different. Existing gateways often lack sufficient flexibility to support multiple protocol conversions and interface adaptations, leading to compatibility issues in data collection. In actual applications, especially in complex industrial environments, traditional edge computing gateways often have insufficient hardware design and cannot quickly respond to and adapt to changes in equipment and network environments, resulting in insufficient system stability and reliability. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides an intelligent edge computing device and a method for implementing its communication connection, which solves the problems of the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for realizing communication connection of an intelligent edge computing device, comprising: S1. obtaining the operation data of the intelligent edge computing device and the data to be transmitted from the sensor and the device interface, and performing preprocessing; S2. dynamically distributing the load according to the device operation data, allocating the computing tasks through the weighted polling scheduling algorithm, and sharding the data to be transmitted through the hash sharding algorithm to obtain the computing task allocation order and the sharding data set; S3. performing protocol conversion and interface adaptation on the sharding data set through the dynamic protocol mapping algorithm and the computing task allocation order; S4. real-time monitoring of the protocol conversion and interface adaptation process, and adjusting the hardware resource configuration and network settings of the intelligent edge computing device through the fuzzy control algorithm.

[0006] Furthermore, the specific process of obtaining the intelligent edge computing device operation data and the data to be transmitted from the sensor and device interface and performing preprocessing is as follows: the intelligent edge computing device operation data includes device status data and device performance data; the device status data includes the number of online computing units and the device interface status value; the device performance data includes the computing unit CPU load, the computing unit memory usage rate and the computing unit IO load; the device operation data and the data to be transmitted are cleaned and formatted.

[0007] Furthermore, dynamic load distribution is performed according to the equipment operation data, and computing tasks are allocated through a weighted polling scheduling algorithm. The specific process of obtaining the computing task allocation order is as follows: according to the equipment performance data, the load of each computing unit is calculated, and a weight is assigned to each computing unit through the hierarchical analysis method; the number of computing tasks allocated to each computing unit is determined according to the weight value of each computing unit through the weighted polling scheduling algorithm, and all computing units are polled and traversed; according to the output of the weighted polling scheduling algorithm, the computing task allocation order is generated.

[0008] Furthermore, the data to be transmitted is sharded through the hash sharding algorithm, and the specific process of obtaining the computing task allocation order and the sharded data set is as follows: based on the computing task allocation order and the equipment operation data, the sharding rules are set; the data to be transmitted is hashed to generate a sharding identifier; the data to be transmitted is allocated to the corresponding shards, and the sharded data set is generated and organized according to the sharding processing results.

[0009] Furthermore, through the dynamic protocol mapping algorithm and the computing task allocation order, the specific process of protocol conversion and interface adaptation of the sharded data set is as follows: according to the computing task allocation order and the sharded data set, the protocol requirements and interface adaptation requirements of the sharded data set are extracted; according to the protocol requirements and interface adaptation requirements, the dynamic protocol mapping algorithm is used to identify and match the communication protocol and interface type of the sharded data set.

[0010] Furthermore, the specific process of using the dynamic protocol mapping algorithm to identify and match the communication protocol and interface type of the sharded data set is as follows: define the protocol mapping rules; through the dynamic protocol mapping algorithm, according to the defined protocol mapping rules, match the protocol requirements of the data fragments with the target protocol, and identify the required target protocol type and interface type; apply the protocol conversion algorithm to convert the data in the sharded data set from the source protocol to the target protocol, and perform interface adaptation processing on the converted data.

[0011] Furthermore, the specific process of adjusting the hardware resource configuration and network settings of the intelligent edge computing device through the fuzzy control algorithm is as follows: monitor the real-time performance data of protocol conversion and interface adaptation; analyze the load situation and network performance of the intelligent edge computing device, and evaluate the adjustment index of the intelligent edge computing device; take the adjustment index of the intelligent edge computing device as the output variable and divide it into fuzzy sets; formulate fuzzy rules to describe the impact of the adjustment index of the intelligent edge computing device on the hardware resource configuration and network settings; perform fuzzy reasoning according to the fuzzy rules to determine the hardware resource configuration and network settings; the hardware resource configuration includes the performance mode of the CPU frequency and the memory load mode; the network settings include the bandwidth demand mode and the delay optimization mode.

[0012] An intelligent edge computing device comprises the following modules: a data acquisition module, a load distribution module, a protocol adaptation module, and a setting adjustment module; the data acquisition module is used to obtain the intelligent edge computing device operation data and the data to be transmitted from the sensor and the device interface, and perform preprocessing; the load distribution module is used to dynamically distribute the load according to the device operation data, distribute the computing tasks through a weighted polling scheduling algorithm, and perform sharding processing on the data to be transmitted through a hash sharding algorithm to obtain the computing task allocation sequence and the sharding data set; the protocol adaptation module is used to perform protocol conversion and interface adaptation on the sharding data set through a dynamic protocol mapping algorithm and the computing task allocation sequence; the setting adjustment module is used to monitor the protocol conversion and interface adaptation process in real time, and adjust the hardware resource configuration and network settings of the intelligent edge computing device through a fuzzy control algorithm.

[0013] Further,.

[0014] The present invention has the following beneficial effects:

[0015] (1) This method for implementing communication connection of an intelligent edge computing device can ensure that the data has been screened and optimized before transmission by obtaining data from sensors and device interfaces and performing preprocessing, thereby improving the efficiency of data transmission and processing. A weighted round-robin scheduling algorithm is used for dynamic load distribution to ensure balanced distribution of computing tasks among various computing units, thereby improving the utilization of computing resources and the overall performance of the system. A hash sharding algorithm is used to shard the data to be transmitted, thereby effectively managing large-scale data, reducing data transmission delays, and improving the flexibility and reliability of data processing.

[0016] (2) This intelligent edge computing device, by setting up a data acquisition module, a load distribution module, a protocol adaptation module, and a setting adjustment module, uses a dynamic protocol mapping algorithm to perform protocol conversion and interface adaptation on the sharded data set, thereby solving the problem of protocol and interface incompatibility between different devices and systems, and realizing seamless data exchange and communication between systems. By real-time monitoring of the protocol conversion and interface adaptation process, and using a fuzzy control algorithm to dynamically adjust the hardware resource configuration and network settings, the performance of the device can be optimized according to the actual operating conditions, ensuring the stability and efficiency of the system under different load conditions.

[0017] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of a method for implementing a communication connection of an intelligent edge computing device of the present invention.

[0019] Figure 2 This is a flow chart of an intelligent edge computing device of the present invention. DETAILED DESCRIPTION

[0020] The embodiments of the present application solve the problems of insufficient computing power and system compatibility of edge computing gateways when processing high-concurrency data and supporting multiple communication protocols and interface adaptations through an intelligent edge computing device and a method for implementing its communication connection.

[0021] The overall idea of ​​the problem in the embodiment of this application is as follows:

[0022] Obtain intelligent edge computing device operation data and data to be transmitted from sensors and device interfaces, and perform preprocessing.

[0023] Dynamic load distribution is performed according to the equipment operation data, computing tasks are allocated through a weighted polling scheduling algorithm, and the data to be transmitted is sharded through a hash sharding algorithm to obtain the computing task allocation order and sharded data set.

[0024] Through the dynamic protocol mapping algorithm and the order of computing task allocation, the protocol conversion and interface adaptation are performed on the sharded data set.

[0025] Monitor the protocol conversion and interface adaptation process in real time, and adjust the hardware resource configuration and network settings of the intelligent edge computing device through fuzzy control algorithms.

[0026] See also Figure 1, an embodiment of the present invention provides a technical solution: a method for realizing communication connection of an intelligent edge computing device, comprising: S1. obtaining intelligent edge computing device operation data and data to be transmitted from sensors and device interfaces, and performing preprocessing; S2. performing dynamic load distribution according to device operation data, distributing computing tasks through a weighted polling scheduling algorithm, and sharding the data to be transmitted through a hash sharding algorithm to obtain a computing task allocation order and a sharding data set; S3. performing protocol conversion and interface adaptation on the sharding data set through a dynamic protocol mapping algorithm and a computing task allocation order; S4. monitoring the protocol conversion and interface adaptation process in real time, and adjusting the hardware resource configuration and network settings of the intelligent edge computing device through a fuzzy control algorithm.

[0027] In this implementation scheme, the sensor in S1 is a device used to measure physical quantities and can convert these quantities into electrical signals. The device interface includes a transceiver, which is a port or interface for connecting the device to an external system or network, and is used for data input and output. Dynamic load distribution and data sharding processing. In S2, the resource allocation strategy is adjusted according to the current system load to achieve load balancing and ensure that the computing tasks are reasonably allocated to each processing unit; weighted polling scheduling algorithm is a scheduling algorithm that determines the number of processing tasks by assigning weights to each computing unit. The weight is usually set according to the load and performance capabilities of the computing unit. Hash sharding algorithm: an algorithm that divides a data set into multiple small fragments, and each fragment is assigned to a different processing node according to the hash value. In this way, data can be processed in parallel, improving the system processing power and efficiency. The order of computing task allocation in S3: determines the order in which tasks are assigned to each computing unit, which is usually determined based on the load distribution strategy and the actual situation of computing resources. Protocol conversion and interface adaptation in S4: dynamic protocol mapping algorithm: an algorithm used to convert the communication protocol of data from the source protocol to the target protocol to ensure that the data can be correctly transmitted and interpreted between different systems or devices. Protocol conversion: the process of converting data from one communication protocol to another to ensure compatibility between different systems. The protocol usually defines the data format, transmission rules and communication methods; Interface adaptation: adjust the data interface (such as hardware port, software API, etc.) to a form compatible with the target system to ensure that data can be correctly input or output to different systems or devices. Fuzzy control algorithm in S4: a control algorithm used to deal with uncertainty and ambiguity. By defining fuzzy rules and reasoning mechanisms, adjust the configuration of the system to optimize performance; Hardware resource configuration: adjust the allocation of hardware resources in the computer system, such as processor frequency, memory allocation, etc., to improve the performance and efficiency of the system; Network settings: configure network parameters such as bandwidth allocation, connection strategy, network delay, etc. to optimize network performance and data transmission quality.

[0028] Specifically, the specific process of obtaining the intelligent edge computing device operation data and the data to be transmitted from the sensor and device interface and performing preprocessing is as follows: the intelligent edge computing device operation data includes device status data and device performance data; the device status data includes the number of online computing units and the device interface status value; the device performance data includes the computing unit CPU load, the computing unit memory usage rate and the computing unit IO load; the device operation data and the data to be transmitted are cleaned and formatted.

[0029] In this implementation scheme, the number of online computing units: monitor the working status of the computing units and determine which computing units are online. This helps to understand the current working capacity and available resources of the device. Device interface status value: record the working status of each interface to identify whether there is an interface failure or maintenance required. Device performance data: computing unit CPU load: monitor the CPU usage to evaluate its processing capacity and whether there is a performance bottleneck. Computing unit memory usage: track memory usage to determine whether there is a shortage of memory resources or memory leaks. Computing unit IO load: analyze the load of input / output operations to evaluate the data read and write performance and possible bottlenecks of the system. Data cleaning is the process of removing noise, errors or inconsistencies from raw data. Ensure the accuracy and consistency of data and improve the effectiveness of subsequent analysis and processing, including removing duplicate data, filling missing values, correcting erroneous data, etc. Data formatting: convert data into a unified format for subsequent processing and analysis. Make the data conform to the predetermined format standard to improve the compatibility and processability of the data. Including data type conversion (such as converting strings to numerical values), normalization (such as unifying the date format), structuring (such as converting unstructured data into a table form), etc.

[0030] Specifically, dynamic load distribution is performed according to the equipment operation data, and computing tasks are allocated through a weighted round-robin scheduling algorithm. The specific process of obtaining the computing task allocation order is as follows: the load of each computing unit is calculated according to the equipment performance data, and a weight is assigned to each computing unit through the hierarchical analysis method; the number of computing tasks allocated to each computing unit is determined according to the weight value of each computing unit through a weighted round-robin scheduling algorithm, and all computing units are polled and traversed; based on the output of the weighted round-robin scheduling algorithm, the computing task allocation order is generated.

[0031] In this implementation, the load is the workload or resource usage of the computing unit when processing tasks, usually including CPU load, memory usage and IO load. The load L of the computing unit is i It can be calculated by the following formula: i =w CPU CPU i +w Mem Mem i +wIO IO i , where CPU i Indicates the CPU load of computing unit i, Mem i Indicates the memory usage of computing unit i, IO i represents the 10 load of computing unit i, w CPU 、w Mem 、w IO Respectively represent the weight coefficients of CPU, memory and IO loads, reflecting the importance of different load indicators. The analytic hierarchy process is a multi-criteria decision-making method used to assign weights to different options to assist in decision-making. The process of assigning weights to each computing unit using the analytic hierarchy process and constructing a comparison matrix: perform pairwise comparisons on the load of each computing unit, construct a comparison matrix, calculate the eigenvector of the comparison matrix, and obtain the weight of each computing unit. The process of determining the number of computing tasks using the weighted polling scheduling algorithm is as follows. The weighted polling scheduling algorithm determines the number of computing tasks assigned to each computing unit based on the weight value of the computing unit. Determine weights: Based on the weight value obtained by the analytic hierarchy process, set the task ratio of each computing unit. The task allocation formula is as follows: Among them, T i : The number of tasks assigned to computing unit i, W i : Calculate the weight value of unit i, T total : Total number of tasks, ∑ j W j : The sum of the weight values ​​of all computing units. Traverse all computing units, assign tasks according to the weight of each computing unit according to the weighted polling algorithm, and generate a computing task assignment order. The generated computing task assignment order determines the order in which each computing unit processes tasks. According to the output of the weighted polling scheduling algorithm, record the number of tasks assigned to each computing unit. Generate a task assignment order list to specify the order in which each computing unit processes tasks. Assign tasks to each computing unit for processing in the generated order.

[0032] Specifically, the data to be transmitted is sharded through the hash sharding algorithm to obtain the computing task allocation order and the sharded data set. The specific process is as follows: set the sharding rules based on the computing task allocation order and the equipment operation data; perform hash calculation on the data to be transmitted to generate a sharding identifier; allocate the data to be transmitted to the corresponding shards, and generate and organize the sharded data set based on the sharding processing results.

[0033] In this embodiment, the fragmentation rule is a rule for determining how to divide the data to be transmitted into multiple fragments, which is usually based on the characteristics of the data, the design of the hash function and the fragmentation strategy. Determine the hash function used for data fragmentation, such as MD5, SHA-1 or a custom hash function. : Define the fragment size or number according to the data characteristics and system requirements. For example, the maximum size of each fragment can be set to a certain number of bytes, or the number of fragments can be specified. The process of generating a fragmentation identifier by hash calculation is as follows. The hash calculation is to perform hash processing on the data to be transmitted to generate a unique hash value for determining the location of the data fragment. Apply the hash function: take the identifier of the data to be transmitted (such as the index or identifier of the data block) as input and use the preset hash function for calculation. H(d) = HashFunction(d), where H(d) is the hash value of data d, generate a fragmentation identifier: map the hash value to a specific fragment number or position. For example, the fragment to which the data belongs can be determined by performing a modulus operation on the hash value: FragmentID = H(d)modN, where N is the total number of fragments and Fragment ID is the identifier of the data fragment. Data allocation is to allocate the data to be transmitted to the corresponding shards according to the shard identifier. According to the generated shard identifier, the data is stored in the corresponding shard. Each shard may correspond to a storage location or data structure. All generated shards are sorted according to the shard identifier to form a complete shard data set. Each shard contains a part of the data to be transmitted and its metadata.

[0034] Specifically, the specific process of protocol conversion and interface adaptation of sharded data sets through dynamic protocol mapping algorithm and computing task allocation order is as follows: according to the computing task allocation order and the sharded data set, the protocol requirements and interface adaptation requirements of the sharded data set are extracted; according to the protocol requirements and interface adaptation requirements, the dynamic protocol mapping algorithm is used to identify and match the communication protocol and interface type of the sharded data set.

[0035] In this embodiment, the extraction process of protocol requirements and interface adaptation requirements is as follows: Protocol requirements: Extract the protocol requirement information of each data fragment from the fragmented data set. Protocol requirements usually include data transmission protocols (such as MQTT, HTTP, TCP / IP) and data formats (such as JSON, XML, binary). Example: A data fragment may need to be converted from the MQTT protocol to the HTTP protocol. Interface adaptation requirements: Identify the interface type required for each data fragment. For example, a data fragment may need to be transmitted through an RS232 interface, a USB interface, or other specific interfaces. Example: One data fragment needs to be adapted to an RS485 interface, and another needs to be adapted to a USB interface. Use a dynamic protocol mapping algorithm for protocol conversion and interface adaptation. The dynamic protocol mapping algorithm is used to automatically map the source protocol and interface type of the data fragment to the target protocol and interface type according to the protocol requirements and interface adaptation requirements. Create a mapping rule set to define the conversion relationship between different protocols and interfaces. For example, the conversion rule from MQTT to HTTP, and the adaptation rule from RS232 to USB. Matching protocol: Dynamically select the appropriate protocol conversion rule based on the extracted protocol requirements and interface adaptation requirements. For example, converting an MQTT message into an HTTP request. Example: The protocol requirement of the data segment is MQTT, and the target protocol is HTTP. The algorithm converts the data from the MQTT format to the HTTP request format. Perform interface adaptation: According to the interface adaptation requirements, adjust the data transmission method to meet the requirements of the target interface. For example, converting data from RS232 format to USB format. Example: If the data needs to be sent via RS485, but the target device only supports the USB interface, use an adapter to convert the data to USB format.

[0036] Specifically, the specific process of using the dynamic protocol mapping algorithm to identify and match the communication protocol and interface type of the fragmented data set is as follows: define the protocol mapping rules; through the dynamic protocol mapping algorithm, according to the defined protocol mapping rules, match the protocol requirements of the data fragments with the target protocol, and identify the required target protocol type and interface type; apply the protocol conversion algorithm to convert the data in the fragmented data set from the source protocol to the target protocol, and perform interface adaptation processing on the converted data.

[0037] In this embodiment, the process of defining the protocol mapping rules is as follows: The protocol mapping rules are the standards and methods defined for the conversion between different communication protocols and interface types. The rules include: Protocol conversion rules: Describe how to convert data from one communication protocol (such as MQTT) to another (such as HTTP). For example, the data packet format of MQTT is mapped to the HTTP request format. Interface adaptation rules: Define how to adapt data from one interface type (such as RS232) to another (such as USB). For example, convert RS232 signals to USB data transmission standards. Use a dynamic protocol mapping algorithm to match protocols and interfaces, and the matching process is as follows: Extract protocol requirements: Extract the protocol requirements of each data fragment from the fragmented data set. For example, a data fragment may need to be transmitted using the MQTT protocol. Target protocol identification: Use a dynamic protocol mapping algorithm to match the source protocol requirements of the data fragment with the target protocol according to the predefined protocol mapping rules, and identify the applicable target protocol type and interface type. Example: If the data fragment currently uses the MQTT protocol, but the target device requires the HTTP protocol, the dynamic protocol mapping algorithm will select the appropriate conversion rule to convert the data from MQTT to HTTP. Apply the protocol conversion algorithm to perform protocol conversion and interface adaptation. The conversion and adaptation process is as follows: Protocol conversion: Apply the protocol conversion algorithm to convert data from the source protocol to the target protocol. This involves the conversion of data format and protocol syntax. Example: Convert the MQTT message body to the JSON format of the HTTP request. Interface adaptation: After completing the protocol conversion, perform interface adaptation processing according to the target interface type. For example, convert the data format to a format suitable for the target device interface, such as converting serial data to USB data format. Example: Adapt the converted data from the serial interface (RS232) to the USB interface to ensure that the data can be correctly transmitted to the target device.

[0038] Specifically, the specific process of adjusting the hardware resource configuration and network settings of the intelligent edge computing device through the fuzzy control algorithm is as follows: monitor the real-time performance data of protocol conversion and interface adaptation; analyze the load situation and network performance of the intelligent edge computing device, and evaluate the adjustment index of the intelligent edge computing device; take the adjustment index of the intelligent edge computing device as the output variable and divide it into fuzzy sets; formulate fuzzy rules to describe the impact of the adjustment index of the intelligent edge computing device on the hardware resource configuration and network settings; perform fuzzy reasoning according to the fuzzy rules to determine the hardware resource configuration and network settings; the hardware resource configuration includes the performance mode of the CPU frequency and the memory load mode; the network settings include the bandwidth demand mode and the delay optimization mode.

[0039] In this implementation plan, the intelligent edge computing device adjustment index The calculation formula is as follows: Where NI represents the network index, NI = wBW ×BW usage +w Delay ×Net delay , where BW usage :Bandwidth usage, Net delay : Network delay, w BW : Bandwidth usage weight coefficient, w Delay Network delay weight coefficient. The adjustment index is used as the output variable and divided into fuzzy sets. Fuzzy set definition. The output variable adjustment index AI is used as the output variable of the fuzzy controller. Fuzzy set: The adjustment index is divided into multiple fuzzy sets. For example, the performance mode of the CPU frequency. Fuzzy rule example: If the load index is high and the network delay is high, the CPU frequency is high. If the load index is medium and the network delay is medium, the CPU frequency is medium. If the load index is low and the network delay is low, the CPU frequency is low. Definition: High: high frequency (such as 3.0GHz and above) Medium: medium frequency (such as 2.0GHz to 2.9GHz) Low: low frequency (such as 2.0GHz and below), memory load mode fuzzy rule example: If the load index is high and the network bandwidth usage is high, the memory load is high. If the load index is medium and the network bandwidth usage is medium, the memory load is medium. If the load index is low and the network bandwidth usage is low, the memory load is low. Definition: High: high memory load (e.g., greater than 80%), Medium: medium memory load (e.g., 50%-80%), Low: low memory load (e.g., less than 50%). Network settings, Bandwidth demand mode, Fuzzy rule example: If network latency is high and load index is high, then bandwidth demand is high. If network latency is medium and load index is medium, then bandwidth demand is medium. If network latency is low and load index is low, then bandwidth demand is low. Definition: High: high bandwidth demand (e.g., more than 100Mbps), Medium: medium bandwidth demand (e.g., 50-100Mbps), Low: low bandwidth demand (e.g., less than 50Mbps). Latency optimization mode, Fuzzy rule example: If bandwidth demand is high and load index is high, then latency optimization is high. If bandwidth demand is medium and load index is medium, then latency optimization is medium. If bandwidth demand is low and load index is low, then latency optimization is low. Definition: High: high latency optimization (such as reducing latency to less than 10 milliseconds); Medium: medium latency optimization (such as reducing latency to 20-50 milliseconds); Low: low latency optimization (such as latency over 50 milliseconds).

[0040] See also Figure 2, an intelligent edge computing device, comprising the following modules: a data acquisition module, a load distribution module, a protocol adaptation module, and a setting adjustment module; the data acquisition module is used to obtain the intelligent edge computing device operation data and the data to be transmitted from the sensor and the device interface, and perform preprocessing; the load distribution module is used to dynamically distribute the load according to the device operation data, allocate computing tasks through a weighted polling scheduling algorithm, and perform sharding processing on the data to be transmitted through a hash sharding algorithm to obtain the computing task allocation order and the sharding data set; the protocol adaptation module is used to perform protocol conversion and interface adaptation on the sharding data set through a dynamic protocol mapping algorithm and a computing task allocation order; the setting adjustment module is used to monitor the protocol conversion and interface adaptation process in real time, and adjust the hardware resource configuration and network settings of the intelligent edge computing device through a fuzzy control algorithm.

[0041] In this implementation scheme, the data acquisition module obtains the operating data and data to be transmitted of the intelligent edge computing device from the sensor and device interface, and performs preprocessing. This includes the status information of the device (such as the number of online computing units, interface status) and performance data (such as CPU load, memory usage, IO load). Through efficient data acquisition and preprocessing, it is ensured that the subsequent modules can obtain accurate real-time data, providing a reliable data basis for the further operation of the system. The load distribution module performs dynamic load distribution according to the device operation data. The weighted polling scheduling algorithm is used to allocate computing tasks to ensure that each computing unit is load balanced. The data to be transmitted is processed by sharding using the hash sharding algorithm to generate the computing task allocation order and the sharding data set. Through intelligent scheduling and sharding processing, efficient allocation of computing tasks is achieved, system bottlenecks are reduced, device processing capabilities and data transmission efficiency are improved, and overall system performance is optimized. The protocol adaptation module uses a dynamic protocol mapping algorithm and a computing task allocation order to perform protocol conversion and interface adaptation on the sharding data set. This includes identifying and matching the communication protocol and interface type of the sharding data, and performing corresponding protocol conversion and adaptation processing. Enhance the compatibility of devices with various communication protocols and interfaces, achieve seamless docking of multiple protocols, ensure smooth data transmission between different systems and devices, and improve system flexibility and interoperability. Set up adjustment modules to monitor the protocol conversion and interface adaptation process in real time, and dynamically adjust hardware resource configuration (such as CPU frequency, memory load) and network settings (such as bandwidth requirements, latency optimization) through fuzzy control algorithms. Through intelligent fuzzy control algorithms, automatically adjust hardware and network settings according to real-time performance data, optimize device performance, improve system stability and response speed, and ensure optimal performance in a dynamically changing environment.

[0042] In summary, this application has at least the following effects:

[0043] A method for implementing an intelligent edge computing device and its communication connection, which obtains the operating data and data to be transmitted of the intelligent edge computing device from sensors and device interfaces, and dynamically allocates computing tasks using a weighted round-robin scheduling algorithm, slices the data to be transmitted using a hash sharding algorithm, thereby obtaining the computing task allocation order and the sliced ​​data set, and matches the communication protocol of the sliced ​​data set with the interface type through a dynamic protocol mapping algorithm, performs protocol conversion and interface adaptation, and solves the data exchange compatibility problem between different systems. The protocol conversion and interface adaptation process is monitored in real time, and the hardware resource configuration and network settings of the intelligent edge computing device are adjusted using a fuzzy control algorithm. Through dynamic adjustment, the stability and performance of the system under different load conditions are improved.

[0044] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0045] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0046] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0047] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0048] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0049] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for realizing communication connection of an intelligent edge computing device, characterized in that: The following steps are involved: S1. Obtain the intelligent edge computing device operation data and data to be transmitted from the sensor and device interface, and perform preprocessing; S2. Dynamically distribute the load according to the equipment operation data, allocate computing tasks through the weighted polling scheduling algorithm, and slice the data to be transmitted through the hash sharding algorithm to obtain the computing task allocation order and sharding data set; S3. Perform protocol conversion and interface adaptation on the sharded data set through dynamic protocol mapping algorithm and computing task allocation order; S4. Monitor the protocol conversion and interface adaptation process in real time, and adjust the hardware resource configuration and network settings of the intelligent edge computing device through the fuzzy control algorithm.

2. A method for realizing communication connection of an intelligent edge computing device according to claim 1, characterized in that: The specific process of obtaining intelligent edge computing device operation data and data to be transmitted from sensors and device interfaces and preprocessing them is as follows: The intelligent edge computing device operation data includes device status data and device performance data; Equipment status data includes the number of online computing units and equipment interface status values; Device performance data includes computing unit CPU load, computing unit memory usage, and computing unit IO load; Clean and format the equipment operation data and data to be transmitted.

3. A method for realizing communication connection of an intelligent edge computing device according to claim 2, characterized in that: Dynamic load distribution is performed based on the equipment operation data. Computing tasks are assigned through a weighted round-robin scheduling algorithm. The specific process of obtaining the order of computing task assignment is as follows: According to the equipment performance data, the load of each computing unit is calculated, and the weight is assigned to each computing unit through the hierarchical analysis method; The weighted polling scheduling algorithm is used to determine the number of computing tasks assigned to each computing unit according to the weight value of each computing unit, and all computing units are polled in a round-robin manner; Generate the computing task allocation order based on the output of the weighted polling scheduling algorithm.

4. A method for realizing communication connection of an intelligent edge computing device according to claim 3, characterized in that: The specific process of sharding the data to be transmitted by using the hash sharding algorithm and obtaining the computing task allocation order and sharding data set is as follows: Set sharding rules based on the order of computing task allocation and device operation data; Perform hash calculation on the data to be transmitted and generate a shard identifier; Allocate the data to be transmitted to the corresponding shards, and generate and organize the shard data sets based on the shard processing results.

5. A method for realizing communication connection of an intelligent edge computing device according to claim 4, characterized in that: The specific process of protocol conversion and interface adaptation for sharded data sets through dynamic protocol mapping algorithm and computing task allocation order is as follows: Extract the protocol requirements and interface adaptation requirements of the shard data according to the computing task allocation order and shard data set; According to the protocol requirements and interface adaptation requirements, a dynamic protocol mapping algorithm is used to identify and match the communication protocol and interface type of the sharded data set.

6. A method for realizing communication connection of an intelligent edge computing device according to claim 5, characterized in that: The specific process of using the dynamic protocol mapping algorithm to identify and match the communication protocol and interface type of the sharded data set is as follows: Define protocol mapping rules; Through the dynamic protocol mapping algorithm, the protocol requirements of the data fragments are matched with the target protocol according to the defined protocol mapping rules, and the required target protocol type and interface type are identified; Apply the protocol conversion algorithm to convert the data in the sharded data set from the source protocol to the target protocol, and perform interface adaptation processing on the converted data.

7. A method for realizing communication connection of an intelligent edge computing device according to claim 6, characterized in that: The specific process of adjusting the hardware resource configuration and network settings of intelligent edge computing devices through fuzzy control algorithms is as follows: Monitor real-time performance data of protocol conversion and interface adaptation; Analyze the load and network performance of intelligent edge computing devices and evaluate the adjustment index of intelligent edge computing devices; The intelligent edge computing device adjustment index is used as the output variable and divided into fuzzy sets; Formulate fuzzy rules to describe the impact of smart edge computing device adjustment index on hardware resource configuration and network settings; Perform fuzzy reasoning based on fuzzy rules to determine hardware resource configuration and network settings; The hardware resource configuration includes a performance mode of CPU frequency and a memory load mode; The network settings include a bandwidth demand mode and a delay optimization mode.

8. An intelligent edge computing device, applied to a method for realizing communication connection of an intelligent edge computing device according to any one of claims 1 to 7, characterized in that: It includes the following modules: data acquisition module, load distribution module, protocol adaptation module, setting adjustment module; The data acquisition module is used to obtain the intelligent edge computing device operation data and the data to be transmitted from the sensor and the device interface, and perform pre-processing; The load distribution module is used to perform dynamic load distribution according to the equipment operation data, distribute computing tasks through a weighted polling scheduling algorithm, and perform sharding processing on the data to be transmitted through a hash sharding algorithm to obtain the computing task distribution order and the sharding data set; The protocol adaptation module is used to perform protocol conversion and interface adaptation on the sliced ​​data set through a dynamic protocol mapping algorithm and a computing task allocation sequence; The setting adjustment module is used to monitor the protocol conversion and interface adaptation process in real time, and adjust the hardware resource configuration and network settings of the intelligent edge computing device through a fuzzy control algorithm.

Images