CN117834454A

CN117834454A - Internet of things equipment protocol adaptive quick access method

Info

Publication number: CN117834454A
Application number: CN202410239832.2A
Authority: CN
Inventors: 司志兵; 梁超; 刘勃; 赵静文; 赵邦国; 朱宏博; 游�明; 陈文尹; 程维国; 刘道学; 耿天宝; 甄黎明; 贺强; 田炳坤
Original assignee: Anhui Shuzhi Construction Research Institute Co ltd; China Tiesiju Civil Engineering Group Co Ltd CTCE Group
Current assignee: Anhui Shuzhi Construction Research Institute Co ltd; China Tiesiju Civil Engineering Group Co Ltd CTCE Group
Priority date: 2024-03-04
Filing date: 2024-03-04
Publication date: 2024-04-05
Anticipated expiration: 2044-03-04

Abstract

The invention relates to the field of Internet of things, in particular to a rapid access method for protocol adaptation of Internet of things equipment, which comprises the steps of identifying high-priority Internet of things equipment through weight scanning, obtaining the type and the position of the high-priority Internet of things equipment, distributing weight values, and constructing an adaptive network, wherein the adaptive network comprises local data processing and cooperatively calculated edge nodes; the adapter converts the data of different protocols into a unified format and collects real-time and historical data; evaluating a data transmission mechanism, and optimizing transmission parameters and network bandwidth; after data cleaning, aggregation and conversion, extracting a feature training model; and the visual interface displays the model output and the data analysis result. The invention effectively solves the problems of long equipment access time and low data processing efficiency through rapid equipment identification, self-adaptive network construction, protocol conversion, data acquisition, transmission mechanism optimization and feature training.

Description

Internet of things equipment protocol adaptive quick access method

Technical Field

The invention relates to the technical field of the Internet of things, in particular to a rapid access method for protocol adaptation of equipment of the Internet of things.

Background

With the progress of technology, the number of devices and applications of the internet of things has increased dramatically, which requires efficient and reliable communication between the devices. However, due to historical reasons and requirements of different application scenarios, communication protocols supported by the internet of things device are various, such as MQTT, coAP, HTTP, AMQP, and incompatibility among the protocols brings certain challenges to deployment and application of the internet of things.

The patent document with the publication number of CN110113359A discloses an Internet of things platform protocol adaptation method, which comprises the steps that an Internet of things platform supports equipment to access by using a CoAP protocol or an MQTT protocol, the Internet of things platform issues digital certificates applicable to the CoAP protocol and the MQTT protocol for the equipment, the digital certificates are associated with the Topic use authorities of the Internet of things platform, the equipment accesses the Internet of things platform through the digital certificates to authenticate, the Internet of things platform judges whether the equipment obtains the Topic use authorities corresponding to the digital certificates or not, the CoAP protocol and the MQTT protocol are adapted based on the Topic use authorities, and the equipment utilizes the Topic use authorities to carry out Topic message communication to realize communication.

It follows that the following problems are present: the internet equipment in the prior art has long access time and low data processing efficiency.

Disclosure of Invention

Therefore, the invention provides a rapid access method for protocol adaptation of Internet of things equipment, which is used for solving the problems of long access time and low data processing efficiency in the prior art.

In order to achieve the above object, the present invention provides a method for fast accessing protocol adaptation of an internet of things device, the method comprising:

the method comprises the steps of scanning internet of things equipment with high priority through weights, obtaining the type and the position of any internet of things equipment, and distributing a weight value for any internet of things equipment according to the type and the function of the internet of things equipment;

According to the weight and the position information of any one of the Internet of things equipment, constructing an adaptive network comprising a first edge node and a second edge node of the Internet of things equipment, and distributing different edge nodes for any one of the Internet of things equipment; the first edge node is used for carrying out local data processing, and the second edge node is used for carrying out cooperative calculation and data aggregation;

according to the weight and the protocol type of the Internet of things equipment, selecting an adapter and loading the protocol analysis module, and adopting the self-adaptive adapter to perform protocol conversion to convert data of different protocols into a uniform format;

collecting data of the Internet of things equipment through an adapter, wherein the data comprises real-time data and historical data;

evaluating a current data transmission mechanism according to the collected real-time data and historical data and the weight value of the Internet of things equipment, and optimizing data transmission parameters, network bandwidth and a buffer zone of the data transmission mechanism according to an evaluation result;

carrying out data cleaning, data aggregation and data conversion on the acquired data according to the weight of the data, and extracting the characteristics of the data according to the weight of the data to obtain a data set;

Training the model by training the data set, and adjusting parameters of the model and a training process according to the weight of the data;

and displaying the output and data analysis results of the model through a visual interface.

Further, scanning the internet of things equipment with high priority through the weight, obtaining the type and the position of any internet of things equipment, and distributing a weight value to any internet of things equipment according to the type and the function of the internet of things equipment comprises:

identifying and acquiring a list of all connected internet of things devices through network scanning;

according to the type and the function of the Internet of things equipment, a weight value is distributed to any Internet of things equipment;

and according to the weight value of the equipment of the Internet of things, preferentially distributing resources for the equipment with high weight.

Further, constructing an adaptive network including a first edge node and a second edge node of the internet of things device according to the weight and the position information of any one of the internet of things devices includes:

collecting information of the Internet of things equipment;

according to the weight and the position information of the equipment, selecting the equipment with high stability as a first edge node, and selecting the excellent geographic position as a second edge node;

According to the performance, the position and the stability of the Internet of things equipment, weight is distributed to any edge node;

the network device is configured for the first edge node and the second edge node.

Further, the first edge node performing local data processing includes:

collecting data from the internet of things device in real time through the self-adaptive adapter;

performing preliminary processing on the acquired data at the first edge node;

compressing the data after preliminary processing;

storing the processed and compressed data in a local memory of the first edge node;

periodically synchronizing locally stored data to the second edge node;

performing real-time analysis on the first edge node;

making a local decision on the first edge node according to the real-time analysis and preliminary processing results;

monitoring the processing speed and the storage service condition of local data processing, and optimizing the processing strategy and the resource allocation according to the monitored data;

encryption and access control are implemented during the processing of the local data.

Further, the second edge node is configured to perform collaborative computing and data aggregation including:

receiving primarily processed and compressed data from an edge node of an internet of things device in the building field;

Decompressing the transmitted data at the second edge node;

aggregating data from edge nodes of different building domain internet of things devices;

performing a collaborative computing task on the second edge node;

storing the result of the collaborative calculation and the data aggregation on the second edge node;

and analyzing and mining the data after the data aggregation and the collaborative calculation.

Further, performing protocol conversion using the adaptive adapter includes:

according to the equipment type and the protocol, the self-adaptive adapter loads a corresponding protocol analysis module;

the protocol analysis module is in charge of receiving data sent by the equipment and analyzing the data according to protocol specifications to obtain analysis results;

extracting key information according to the protocol analysis result, and converting the extracted key information into a predefined unified data format according to the format;

encapsulating the converted data into data packets with uniform format, and transmitting the encapsulated data packets to a data processing platform;

the self-adaptive adapter dynamically adjusts the protocol analysis module and the data conversion logic according to the feedback of the building field Internet of things equipment;

Verifying whether the converted data meets the compatibility requirement of the target system;

and monitoring the performance of the self-adaptive adapter in the data conversion process, and optimizing the self-adaptive adapter according to the monitoring result.

Further, according to the device type and the protocol, the loading of the corresponding protocol parsing module by the adaptive adapter includes:

metadata defining a device type and a communication protocol;

creating an interface of the protocol analysis module to realize a factory mode of the protocol analysis module;

designing a loading mechanism of the self-adaptive adapter framework to realize dynamic updating of the protocol analysis module;

the exception processing of the protocol analysis module is realized to obtain a processing result;

testing and verifying the protocol analysis module to obtain a test result;

and integrating the protocol analysis module into the adaptive adapter framework according to the processed result and the tested result.

Further, evaluating the current data transmission mechanism according to the collected real-time data and historical data and the weight value of the internet of things device, and optimizing the data transmission parameters, the network bandwidth and the buffer area of the data transmission mechanism according to the evaluation result comprises the following steps:

Collecting the equipment state, performance index and network flow of the Internet of things in real time;

analyzing the data transmission rate, delay and packet loss rate of the current data transmission mechanism to obtain an analysis result;

adjusting the transmission rate according to the analysis result, and optimizing the data coding mode;

according to the weight value and the data transmission requirement of the Internet of things equipment, adjusting the network bandwidth;

according to the historical data flow mode and the real-time network load, the size of the buffer area is adjusted;

testing the adjusted data transmission parameters, network bandwidth and buffer area;

and verifying whether the optimized data transmission is improved according to the test result.

Further, extracting features of the data according to the weights of the data to obtain a data set includes:

calculating a weight value for each feature in the dataset to obtain a feature weight;

selecting the characteristic with high weight according to the characteristic weight;

converting the non-numerical features into a mode of processing by a machine learning algorithm;

using the extracted features, the dataset is constructed.

Further, by training the data set training model, adjusting parameters of the model and training process according to the weight of the data includes:

selecting a model and training the model through the data set;

Evaluating the weight of each feature according to the performance of the model;

adjusting parameters of the model according to the characteristic weights;

evaluating the performance of the model using the validation set;

and deploying the trained model to actual Internet of things equipment.

Compared with the prior art, the method has the beneficial effects that the internet of things equipment with high access priority can be rapidly identified and accessed through the weight scanning mechanism, and the priority access and processing of key equipment are ensured. According to the type and position information of the Internet of things equipment, a self-adaptive network is constructed, the first edge node and the second edge node are reasonably distributed, and effective utilization of resources is achieved. And the self-adaptive adapter is adopted to carry out protocol conversion, so that data of different protocols are converted into a unified format, the data processing flow is simplified, and the complexity of protocol adaptation is reduced. The data of the Internet of things equipment, including real-time data and historical data, are collected through the adapter, and the integrity and accuracy of data collection are improved. And evaluating the current data transmission mechanism according to the collected real-time data and historical data and the weight value of the Internet of things equipment, optimizing the current data transmission mechanism, and improving the efficiency and stability of data transmission. And performing data cleaning, data aggregation and data conversion on the acquired data, extracting the characteristics of the data, obtaining a high-quality data set, and providing a reliable data basis for model training. Through training the data set training model, parameters and training processes of the model are adjusted according to the weight of the data, the performance of the model is optimized, and the processing capacity of the model on the data of the Internet of things is improved. And the output and data analysis results of the model are displayed through the visual interface, so that a user can intuitively know the running condition and the data analysis results of the model, and the satisfaction degree of the user on the Internet of things system is improved.

In particular, by assigning different weight values to different types of internet of things devices, it can be ensured that resources are preferentially assigned to devices critical to the operation of the intelligent system. High weight devices typically represent the most important part of the system, and preferentially allocating resources to them ensures that critical functions of the system are optimized and supported, thereby improving overall efficiency.

In particular, by deploying edge nodes at locations near the data source, data transmission delays can be reduced and data processing efficiency can be improved. Using geographically superior devices as the second edge node may provide faster data access speeds. The equipment with high stability is selected as the first edge node, so that the reliability of the network can be ensured, and faults and downtime are reduced. The weight is distributed according to the performance, the position and the stability of the equipment, so that the reasonable distribution of the resources can be ensured, and the utilization efficiency of the network resources is improved.

In particular, real-time data acquisition ensures timeliness of the data, enabling the system to quickly respond to changes in device state. The data compression reduces the storage and transmission requirements, saves resources, and accelerates the data processing speed. The local storage ensures that the data can be accessed at any time without losing, and is convenient for quick retrieval and analysis. The data synchronization ensures the backup and centralized management of the data and improves the safety and reliability of the data. Real-time analysis enables the system to discover problems and react in real time, and intelligence and adaptability of the system are enhanced. The local decision reduces delay, improves response speed, and is beneficial to timely adjusting the running states of equipment and a system. The continuous monitoring and optimization ensures the high-efficiency operation of the system, avoids the waste of resources and prolongs the service life of equipment. The data encryption and access control protect the security and privacy of the data and prevent unauthorized access and data disclosure.

In particular, by performing data decompression and aggregation on the second edge node, the transmission amount of data in the network can be reduced, the bandwidth usage is reduced, and the overall data processing efficiency is improved. The aggregation of real-time data and the storage of calculation results enable building managers to make more accurate decisions based on the latest data. The second edge node can utilize the computing resource to process data, reduce the dependence on the central server and optimize the resource utilization. And the data processing is performed on the edge node, so that the transmission distance of the data in the network can be reduced, and the risk of data leakage is reduced.

In particular, by means of an adaptive adapter, the system is able to accommodate a plurality of different protocol devices without having to write specific interface programs for each device. The adapter can dynamically load the corresponding analysis module according to the equipment type and the protocol, so that the flexibility and the expandability of the system are improved. The protocol analysis module ensures that the data is accurately analyzed according to the protocol specification, extracts key information and improves the accuracy of data processing. The unified interface reduces development costs and maintenance effort to write separate interfaces for each device. And the data conversion and packaging process is automatic, so that the data circulation speed is increased, and the manual intervention is reduced. And dynamically adjusting a protocol analysis module and a data conversion logic according to the equipment feedback, and optimizing a data processing flow.

In particular, by defining metadata for device types and communication protocols, the system is able to easily support new devices and protocols. Dynamic updating and optimizing of the protocol analysis module ensures that the data processing always meets the latest requirements and standards. The exception handling mechanism ensures that the system does not crash when the data resolution fails, but rather is able to gracefully handle errors. By realizing the testing and verification framework, the testing process of the new protocol analysis module can be simplified, and the quality of the module is ensured. The dynamic update mechanism supports continuous integration and continuous deployment, making integration of new modules and updating of old modules easier.

In particular, by adjusting the transmission rate and optimizing the data encoding, the efficiency of data transmission can be improved, and the transmission time and resource consumption can be reduced. The network bandwidth and buffer management are optimized, delay and packet loss in the data transmission process can be reduced, and the user experience is improved. According to the actual demand, the network bandwidth is dynamically adjusted, the network resources can be more effectively utilized, and the overall performance of the network is improved. By optimizing the data transmission mechanism, unnecessary waste of network resources and equipment can be reduced, thereby reducing the operation cost.

In particular, by selecting important features, we can build a more efficient model, reduce the complexity of the model, and improve the accuracy of the predictions. Reducing the number of features can also reduce the computational complexity of the model, thereby improving the computational efficiency of the model.

In particular, by selecting a model and adjusting model parameters according to data characteristics, the accuracy of the model in predicting unknown data can be improved. When the trained model is deployed on the Internet of things equipment, the data can be processed more intelligently, and the risk of data leakage is reduced. Accurate predictions can help the internet of things devices provide more intelligent services. The model can quickly respond to data changes and support real-time decision making.

Drawings

Fig. 1 is a flow chart of a fast access method for protocol adaptation of an internet of things device according to an embodiment of the present invention;

fig. 2 is a schematic flow chart of a method for scanning internet of things equipment with high priority through weights in the fast access method for internet of things equipment protocol adaptation, which is provided by the embodiment of the invention, obtaining the type and the position of any one of the internet of things equipment, and distributing a weight value for any one of the internet of things equipment according to the type and the function of the internet of things equipment;

fig. 3 is a schematic flow chart of collaborative computing and data aggregation performed by the second edge node in the fast access method for internet of things device protocol adaptation according to the embodiment of the present invention;

fig. 4 is a schematic flow chart of protocol conversion by adopting the adaptive adapter in the fast access method for protocol adaptation of the internet of things device provided by the embodiment of the invention.

Detailed Description

In order that the objects and advantages of the invention will become more apparent, the invention will be further described with reference to the following examples; it should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.

It should be noted that, in the description of the present invention, terms such as "upper," "lower," "left," "right," "inner," "outer," and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the apparatus or elements must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.

Furthermore, it should be noted that, in the description of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those skilled in the art according to the specific circumstances.

Referring to fig. 1, the method for fast accessing protocol adaptation of internet of things equipment provided by the invention includes:

s100, scanning the internet of things equipment with high priority through the weight, acquiring the type and the position of any one of the internet of things equipment, and distributing a weight value for any one of the internet of things equipment according to the type and the function of the internet of things equipment;

s200, constructing an adaptive network comprising a first edge node and a second edge node of the Internet of things equipment according to the weight and the position information of any one of the Internet of things equipment, and distributing different edge nodes for any one of the Internet of things equipment; the first edge node is used for carrying out local data processing, and the second edge node is used for carrying out cooperative calculation and data aggregation;

s300, according to the weight and the protocol type of the Internet of things equipment, selecting an adapter and a loading protocol analysis module, and adopting a self-adaptive adapter to perform protocol conversion to convert data of different protocols into a uniform format;

s400, collecting data of the Internet of things equipment through an adapter, wherein the data comprises real-time data and historical data;

s500, evaluating a current data transmission mechanism according to the collected real-time data and historical data and the weight value of the Internet of things equipment, and optimizing data transmission parameters, network bandwidth and a buffer zone of the data transmission mechanism according to an evaluation result;

S600, carrying out data cleaning, data aggregation and data conversion on the acquired data according to the weight of the data, and extracting the characteristics of the data according to the weight of the data to obtain a data set;

s700, training the data set training model, and adjusting parameters of the model and a training process according to the weight of the data;

s800, displaying the output and data analysis result of the model through a visual interface.

In particular, all internet of things devices connected to a network are identified using a network scanning tool or a device discovery protocol (such as UPnP). The type and location information of the device is obtained by means of the MAC address, IP address, UDN or other identifier of the device. Devices are classified according to their type (e.g., temperature sensor, camera, bulb, etc.) and function. A weighting algorithm is designed to assign a weight value to the device based on its classification, importance, and anticipated data processing requirements. For example, a higher weight may be assigned to critical devices such as security cameras. According to the weight and location information of the device, the edge node that is closest or has the highest processing power is selected. Each device is assigned a first edge node (for local data processing) and a second edge node (for collaborative computing and data aggregation) to construct an adaptive network comprising at least one first edge node for local data processing and one second edge node for collaborative computing and data aggregation. And selecting a proper adapter and a protocol analysis module matched with the selected adapter according to the weight and the protocol type of the equipment. For example, for a Zigbee enabled device, a Zigbee adapter is selected. And selecting a protocol analysis module matched with the adapter. And converting the data of different protocols into a uniform format by adopting an adaptive adapter. Real-time data and historical data are collected from the internet-enabled device through the adapter. The converted data is transmitted to the corresponding edge node through the adapter. Real-time data and historical data are stored on the edge nodes for local processing and subsequent analysis. The efficiency, delay and reliability of data transmission are analyzed, taking into account network congestion and bandwidth limitations. And adjusting data transmission parameters such as transmission rate, retransmission mechanism and network buffer size according to the evaluation result. And cleaning the acquired data according to the weight of the data, such as removing noise and abnormal values. Data aggregation and data conversion processes, such as data compression and data normalization, are performed. And extracting the characteristics of the data according to the weight of the data to obtain a data set suitable for model training. A model, such as a machine learning model or a deep learning model, is trained using the training dataset. Parameters of the model, such as learning rate, iteration number, etc., are adjusted according to the weights of the data. The model training process is monitored, and the model can be ensured to accurately and efficiently process data.

Specifically, through a weight scanning mechanism, the internet of things equipment with high access priority can be rapidly identified, and the priority access and processing of key equipment are ensured. According to the type and position information of the Internet of things equipment, a self-adaptive network is constructed, the first edge node and the second edge node are reasonably distributed, and effective utilization of resources is achieved. And the self-adaptive adapter is adopted to carry out protocol conversion, so that data of different protocols are converted into a unified format, the data processing flow is simplified, and the complexity of protocol adaptation is reduced. The data of the Internet of things equipment, including real-time data and historical data, are collected through the adapter, and the integrity and accuracy of data collection are improved. And evaluating the current data transmission mechanism according to the collected real-time data and historical data and the weight value of the Internet of things equipment, optimizing the current data transmission mechanism, and improving the efficiency and stability of data transmission. And performing data cleaning, data aggregation and data conversion on the acquired data, extracting the characteristics of the data, obtaining a high-quality data set, and providing a reliable data basis for model training. Through training the data set training model, parameters and training processes of the model are adjusted according to the weight of the data, the performance of the model is optimized, and the processing capacity of the model on the data of the Internet of things is improved. And the output and data analysis results of the model are displayed through the visual interface, so that a user can intuitively know the running condition and the data analysis results of the model, and the satisfaction degree of the user on the Internet of things system is improved.

Specifically, as shown in fig. 2, scanning the internet of things device with high priority through the weight, obtaining the type and the position of any one of the internet of things device, and distributing a weight value to any one of the internet of things device according to the type and the function of the internet of things device includes:

s101, identifying and acquiring a list of all connected Internet of things equipment through network scanning;

s102, distributing a weight value to any Internet of things equipment according to the type and the function of the Internet of things equipment;

s103, according to the weight value of the Internet of things equipment, resources are preferentially allocated to the equipment with high weight.

Specifically, a network scanner is started to identify and obtain a list of all internet of things devices currently connected. The information including, but not limited to, IP address, MAC address, device identifier, etc. is scanned to build a complete list of devices. Weight values are assigned according to the type (such as a temperature sensor, a humidity sensor, a camera and the like) and the function (such as key equipment, auxiliary equipment and the like) of the equipment of the internet of things. For example, a critical device may be assigned a higher weight value to ensure its priority and resource allocation. And sorting according to the weight values of the devices of the Internet of things, and giving priority to the devices with higher weight values. High weight devices are allocated resources such as computing power, storage space, network bandwidth, etc. The resource allocation can be dynamically adjusted according to the real-time requirements of the device and the system resource status. The type and location information of the device is acquired before the resources are allocated. Such information may help determine an optimal resource allocation strategy, for example, to place certain devices on edge nodes near the data processing center. The state of the Internet of things equipment, such as an online state, a workload, a fault condition and the like, is monitored in real time. The weight value is dynamically adjusted according to the state of the device to reflect the actual priority of the device. And collecting feedback information of resource allocation, such as equipment performance, data transmission efficiency and the like. And optimizing and adjusting the weight distribution strategy and the resource distribution mechanism according to the feedback information.

In particular embodiments, a series of internet of things devices, such as traffic lights, environmental monitoring sensors, smart meters, etc., are maintained. The internet of things equipment data comprises the following indexes: the operating state of the equipment (online/offline), the failure rate of the equipment (failure times/1000 hours), the energy consumption of the equipment (unit: kilowatt-hours), the age of the equipment (unit: month), the location of the equipment (e.g. urban center, suburban area).

First, the data needs to be purged to remove invalid or incomplete records. The data is converted to a unified format, for example, date and time is converted to YYYY-MM-DD HH: MM: SS format.

Setting the running state of equipment to distribute 40% weight; setting equipment failure rate, equipment energy consumption and equipment age, and respectively distributing weights of 20%; the location of the device is set to assign a weight of 10%.

And calculating the comprehensive score of each Internet of things device by using a weighted average method. Composite score = (device operating status x 40%) + (device failure rate x 20%) + (device energy consumption x 20%) + (device age x 20%) + (device location x 10%).

Ranking the devices according to the composite score may be the most stable and efficient device, while the lowest scoring device may require further analysis or maintenance.

For the equipment with highest score, the company can continue to popularize and apply the equipment to improve the operation efficiency of the intelligent city.

For the lowest scoring device, the company may perform maintenance or replacement to improve the stability and performance of the device.

For example, the data for device a and device B are as follows:

the running state of the equipment A is on-line, the failure rate is 5 times/1000 hours, the energy consumption is 100 kilowatt-hours, the age is 12 months, and the equipment A is positioned in the city center.

The operating state of the equipment B is on-line, the failure rate is 10 times/1000 hours, the energy consumption is 120 kilowatt-hours, the age is 6 months, and the equipment B is positioned in suburbs.

From the above weight assignments we can calculate the composite score for device a and device B:

the composite score for device a = (online x 40%) + (5 times/1000 hours x 20%) + (100 kw hours x 20%) + (12 months x 20%) + (center of market x 10%) =0.4+0.1+2+0.24+0.1=0.84.

The composite score for device B = (online x 40%) + (10 times/1000 hours x 20%) + (120 kw hours x 20%) + (6 months x 20%) + (suburb x 10%) =0.4+0.2+2.4+0.12+0.06=0.78).

According to the comprehensive score, the score of the equipment A is higher than that of the equipment B, and the maintenance and popularization of the equipment A are prioritized.

Specifically, by assigning different weight values to different types of internet of things devices, it can be ensured that resources are preferentially assigned to devices that are critical to the operation of the intelligent system. High weight devices typically represent the most important part of the system, and preferentially allocating resources to them ensures that critical functions of the system are optimized and supported, thereby improving overall efficiency.

Specifically, constructing an adaptive network including a first edge node and a second edge node of the internet of things device according to the weight and the position information of any one of the internet of things devices includes:

collecting information of the Internet of things equipment;

Specifically, information of all the devices of the internet of things is collected through the network scanning and device management API, including device types, functions, performance indexes, positions, stability data and the like. According to the weight value and stability data of the equipment of the Internet of things, selecting the equipment with high weight and good stability as the first edge node. The weight value may reflect the importance, performance, and current state of the device. In selecting the second edge node, a geographically advantageous device is prioritized, e.g. a device located at the edge of the network coverage or a device with efficient data transfer capabilities. The first edge node and the second edge node are assigned weight values in combination with their performance, location and stability. This ensures that network resources are allocated reasonably and that network performance is optimized. Network devices, including IP addresses, subnet masks, gateways, DNS servers, etc., are configured for the first and second edge nodes to ensure that they can effectively communicate and process internet of things data.

In particular, by disposing the edge node at a position close to the data source, the data transmission delay can be reduced, and the data processing efficiency can be improved. Using geographically superior devices as the second edge node may provide faster data access speeds. The equipment with high stability is selected as the first edge node, so that the reliability of the network can be ensured, and faults and downtime are reduced. The weight is distributed according to the performance, the position and the stability of the equipment, so that the reasonable distribution of the resources can be ensured, and the utilization efficiency of the network resources is improved.

Specifically, the local data processing by the first edge node includes:

performing preliminary processing on the acquired data at the first edge node;

compressing the data after preliminary processing;

periodically synchronizing locally stored data to the second edge node;

performing real-time analysis on the first edge node;

In particular implementations, in a large commercial building, there is a first edge node that is responsible for collecting and managing data from various building domain internet of things devices. These devices include temperature sensors, humidity sensors, lighting controllers, door access systems, and the like. The first edge node collects temperature data from the temperature sensor in real time through the adaptive adapter, collects humidity data from the humidity sensor, collects illumination state data from the illumination controller, and collects access control state data from the access control system. And at the first edge node, performing preliminary processing on the acquired data. For example, the temperature data is denoised, the illumination state data is formatted, and the access state data is parsed. And compressing the data after preliminary processing to reduce the data transmission quantity and the storage space requirement. For example, the temperature data is compressed using a snap compression algorithm. The processed and compressed data is stored in a local memory of the first edge node. For example, temperature data is stored on a Solid State Disk (SSD). The locally stored data is periodically synchronized to a second edge node (e.g., cloud server). For example, the temperature data is synchronized to the cloud server every minute. Real-time analysis is performed on the first edge node, temperature data is analyzed in real time by using a data stream processing tool, and abnormal temperature fluctuation is detected. If abnormal temperature fluctuations are detected, the first edge node may trigger the opening of the air conditioning system. The data processing algorithm is adjusted or the storage capacity is increased according to the processing speed and the storage usage. The temperature data is encrypted using the AES encryption algorithm, and access control is implemented by setting user rights and roles.

Specifically, as shown in fig. 3, the second edge node is configured to perform collaborative computing and data aggregation, including:

s201, receiving the primarily processed and compressed data from an edge node of the Internet of things equipment in the building field;

S202, decompressing the transmitted data on the second edge node;

s203, aggregating data from the first edge nodes of the different building field Internet of things devices;

s204, executing a cooperative computing task on the second edge node;

s205, storing the result of the collaborative calculation and the data aggregation on the second edge node;

s206, analyzing and mining the data after the data aggregation and the collaborative calculation.

In particular, the second edge node receives the preliminary processed and compressed data sent from the first edge node via the network interface. And ensuring the integrity and the security of the data, and carrying out data verification and encryption. And deploying a decompression algorithm on the second edge node, decompressing the received compressed data, and recovering the original data. And selecting a proper decompression algorithm to ensure the efficiency and accuracy of data decompression. Key information in the decompressed data, such as time stamps, sensor readings, etc., is extracted. Classifying and summarizing the data according to the need to form a data aggregation result. It is desirable to use database or data warehouse technology to store and manage aggregated data. And executing cooperative computing tasks, such as data filtering, pattern recognition and the like, on the aggregated data by using the computing resources of the second edge node. Collaborative computing may involve joint processing of multiple data streams or data sets, requiring the use of a distributed computing framework. The result of the co-computation and data aggregation is stored in a local store or memory of the second edge node. Ensuring the security and reliability of data, the need for data encryption and backup requires the realization of data consistency and persistent storage mechanisms. The data stored on the second edge node is further analyzed using data analysis tools and algorithms, such as machine learning, deep learning, etc. And mining modes, trends and correlations in the data, and providing basis for decision support and system optimization. The data analysis flow and algorithm need to be customized to the analysis requirements.

In a specific embodiment, the second edge node receives 1000 data packets per second from different building domain internet of things devices. The average size of each data packet is 100 bytes, and the decompressed data size is 200 bytes. The amount of decompressed data per second reaches 200,000 bytes. Each data packet contains 3 key information fields, time stamp, temperature reading and humidity reading, respectively. 1000 packets per second are aggregated to generate 3000 key information records. For example, an aggregate chart containing all temperature readings, and an aggregate chart containing all humidity readings may be generated. And executing a data filtering task, processing 3000 key information records per second, and screening out abnormal data. For example, if the temperature reading exceeds 30 ℃, then the flag is abnormal. A pattern recognition task was performed that recognizes 50 associated patterns of temperature and humidity per second. For example, if a combination of temperature and humidity occurs frequently within a particular time period, it may be indicative of a particular environmental condition. 3000 key information records per second and 50 pattern recognition results are stored in the memory of the second edge node. The capacity of the memory is 1GB, and 1 hour of data can be stored. The data is stored in a memory in an efficient data structure, such as a hash table or a time-series database. The stored data is further analyzed using a machine learning algorithm, processing 100 records per second to identify energy consumption patterns within the building. For example, by analyzing data of temperature and humidity, energy consumption trend in a building can be predicted. An analysis report was generated every 10 minutes, containing energy consumption trends and optimization advice.

Specifically, by performing data decompression and aggregation on the second edge node, the transmission amount of data in the network can be reduced, the bandwidth usage is reduced, and the overall data processing efficiency is improved. The aggregation of real-time data and the storage of calculation results enable building managers to make more accurate decisions based on the latest data. The second edge node can utilize the computing resource to process data, reduce the dependence on the central server and optimize the resource utilization. And the data processing is performed on the edge node, so that the transmission distance of the data in the network can be reduced, and the risk of data leakage is reduced.

Specifically, as shown in fig. 4, the protocol conversion using the adaptive adapter includes:

s301, loading a corresponding protocol analysis module by the self-adaptive adapter according to the equipment type and the protocol;

s302, the protocol analysis module is responsible for receiving data sent by the equipment and analyzing the data according to protocol specifications to obtain analysis results;

s303, extracting key information according to the protocol analysis result, and converting the extracted key information into a predefined unified data format according to the format;

s304, packaging the converted data into data packets with a uniform format, and transmitting the packaged data packets to a data processing platform;

S305, the self-adaptive adapter dynamically adjusts the protocol analysis module and the data conversion logic according to the feedback of the building field Internet of things equipment;

s306, verifying whether the converted data meets the compatibility requirement of the target system;

s307, monitoring performance of the self-adaptive adapter in the data conversion process, and optimizing the self-adaptive adapter according to the monitoring result.

Specifically, an extensible adaptive adapter framework is designed, and a corresponding protocol analysis module can be dynamically loaded according to the equipment type and the communication protocol. Each protocol parsing module is responsible for parsing data of a specific protocol and extracting useful information. And realizing a protocol analysis module so that the protocol analysis module can receive the data stream sent by the equipment. And analyzing the data according to the protocol specification, and extracting the equipment information and the data content. Key information such as device type, data value, time stamp, etc. is extracted from the parsing result. The extracted information is converted according to a predefined unified data format. The data packaging module is responsible for packaging the converted data into a data packet with a uniform format. The data packets are transmitted over a network to a data processing platform, such as a cloud platform or a data warehouse. A device feedback mechanism is implemented that allows a device to provide feedback to the adaptive adapter regarding data formats and protocols. And dynamically adjusting the protocol analysis module and the data conversion logic according to the feedback so as to optimize the data conversion process. Verifying whether the converted data meets the compatibility requirement of the target system or not: and verifying the received data packet in the data processing platform to ensure that the data packet meets the compatibility requirement of the target system. And if the data packet does not meet the requirements, triggering the self-adaptive adapter to carry out corresponding adjustment. And deploying a monitoring system, and collecting performance data of the self-adaptive adapter in the data conversion process in real time. And optimizing the self-adaptive adapter according to monitoring results such as conversion speed, error rate, resource consumption and the like.

In a specific embodiment, there is a temperature sensor that transmits data using the Modbus protocol. The adapter framework will first identify the type of device and the protocol it uses. The framework will then load a protocol parsing module designed specifically for the Modbus protocol. Device type: a temperature sensor; protocol: modbus TCP; data format: a 16-bit register, small-end endian; protocol parsing module interface: receiveData (stream): receiving a data stream; paramata (): analyzing data according to Modbus specifications; extractDeviceInfo (): extracting device information and data content; convertData (format): converting the data into a unified data format; when the temperature sensor is connected, the adapter framework detects Modbus TCP protocol and loads a corresponding protocol analysis module. The analysis module extracts a temperature value of 25 ℃ and a timestamp of 2023-04-01T12:00:00Z.

The data packaging module packages the data into a JSON format as follows:

{"deviceType":"temperatureSensor","value":25,"timestamp":"2023-04-01T12:00:00Z"}。

the data packet is sent to the cloud platform through the network. The device sends feedback indicating that the data format needs to be changed, e.g., using big-end endian. The adapter adjusts the Modbus protocol parsing module to use the correct byte order. And after the cloud platform receives the data packet, verifying whether the JSON format and the fields accord with expectations. If not, the cloud platform informs the adapter to adjust. The monitoring system collects performance data during data conversion, for example, conversion speed is 1000 data packets/second, and error rate is 0.1%. And optimizing the adapter according to the monitoring result to improve the conversion speed and reduce the error rate.

In particular, by means of an adaptive adapter, the system is able to accommodate a plurality of different protocol devices without having to write a specific interface program for each device. The adapter can dynamically load the corresponding analysis module according to the equipment type and the protocol, so that the flexibility and the expandability of the system are improved. The protocol analysis module ensures that the data is accurately analyzed according to the protocol specification, extracts key information and improves the accuracy of data processing. The unified interface reduces development costs and maintenance effort to write separate interfaces for each device. And the data conversion and packaging process is automatic, so that the data circulation speed is increased, and the manual intervention is reduced. And dynamically adjusting a protocol analysis module and a data conversion logic according to the equipment feedback, and optimizing a data processing flow.

Specifically, according to the device type and the protocol, the adaptive adapter loads a corresponding protocol parsing module, which includes:

metadata defining a device type and a communication protocol;

designing a loading mechanism of an adapter framework to realize dynamic updating of the protocol analysis module;

the exception handling of the protocol analysis module is realized;

Testing and verifying the protocol analysis module to obtain a test result;

and integrating the protocol analysis module into the adapter framework according to the processed result and the tested result.

Specifically, a class or structure is created to store information about the device type and communication protocol, such as protocol name, data format, supported commands, etc. An interface is defined listing the methods that all protocol parsing modules must implement, such as parseData, extractKeyInfo, convertToFormat, etc. A factory class is created and is responsible for creating and returning corresponding protocol parsing module instances according to the device type and the communication protocol. A loading mechanism is implemented in the adapter framework that dynamically loads the corresponding protocol parsing module according to the device type and communication protocol. A mechanism is designed that allows the protocol resolution module to update its resolution logic at run-time to accommodate protocol changes or new device types. Each protocol analysis module can be ensured to process possible abnormal conditions in the analysis process, and an appropriate error processing mechanism can be provided. A set of test framework is developed for verifying the correctness and performance of the protocol parsing module. And adjusting and optimizing the protocol analysis module according to the test result. The protocol analysis module after test verification is integrated into the adapter framework, so that the protocol analysis module can cooperate and a unified data conversion interface is provided.

Specifically, evaluating a current data transmission mechanism according to the collected real-time data and historical data and the weight value of the internet of things device, and optimizing data transmission parameters, network bandwidth and a buffer zone of the data transmission mechanism according to an evaluation result comprises:

and verifying whether the efficiency of the optimized data transmission is improved or not according to the test result.

Specifically, a monitoring tool is deployed to collect the states of the internet of things equipment in real time, including performance indexes such as CPU utilization rate, memory occupation, sensor reading and the like. Network traffic analysis tools are used to collect network traffic data such as number of data packets, packet size, data transmission rate, etc. The collected data is analyzed to determine the performance of the data transmission mechanism, including data transmission rate, delay, and packet loss rate. Bottlenecks and problem areas in the network, such as high-delay transmission paths or frequently lost network segments, are identified. And adjusting the data transmission rate according to the analysis result to ensure that the network resources are reasonably utilized. More efficient data encoding schemes, such as employing more advanced encoding algorithms to increase the efficiency and reliability of data transmission, are explored and implemented. And reallocating the network bandwidth according to the weight value and the data transmission requirement of the Internet of things equipment. Ensuring that high weight devices get adequate bandwidth support while also taking into account the basic bandwidth requirements of low weight devices. Buffer sizes are adjusted according to historical data traffic patterns and real-time network loads to reduce data queuing and delay. During peak flow, the buffer size is increased to absorb additional water level fluctuations. And testing the adjusted data transmission parameters, network bandwidth and buffer zone settings in an actual network environment. And evaluating the test results, including the change of key indexes such as transmission rate, delay, packet loss rate and the like, so as to verify the optimization effect. And comparing the performance data before and after the optimization, and evaluating whether a data transmission mechanism is improved. User feedback and system stability data are collected to ensure that optimization does not negatively impact other network services. Network performance is continuously monitored, and the persistence of the optimization effect is ensured. Parameters are periodically re-evaluated and adjusted to accommodate changes in network conditions and traffic demands.

Specifically, by adjusting the transmission rate and optimizing the data encoding, the efficiency of data transmission can be improved, and the transmission time and resource consumption can be reduced. The network bandwidth and buffer management are optimized, delay and packet loss in the data transmission process can be reduced, and the user experience is improved. According to the actual demand, the network bandwidth is dynamically adjusted, the network resources can be more effectively utilized, and the overall performance of the network is improved. By optimizing the data transmission mechanism, unnecessary waste of network resources and equipment can be reduced, thereby reducing the operation cost.

Specifically, extracting features of the data according to the weights of the data to obtain a dataset includes:

using the extracted features, the dataset is constructed.

Specifically, using existing machine learning models (e.g., random forests, gradient hoists, etc.), the contribution of each feature to the model's predictions is evaluated, and the feature weights are set to the feature importance scores given by the model. And selecting the feature with higher weight according to the calculated feature weight. A threshold may be set that only retains features whose weights exceed the threshold. The selected features are ranked, with top ranked features being prioritized. The classification features are converted into numerical features, with each class corresponding to a unique number. A binary vector is created for each class, with only one bit being 1 and the remaining bits being 0. The categories are converted to binary representations, each category having a unique binary code. The selected features are combined into a matrix, each row representing a sample and each column representing a feature. The target variables (labels) of the samples are combined into a vector corresponding to the feature matrix. The feature matrix and the target vector are partitioned into a training set, a validation set and a test set, and the partitioning is performed using a ratio of 70-15-15.

Specifically, by training the data set training model, the parameters of the model and the training process according to the weight of the data are adjusted, including:

selecting a model and training the model through the data set;

adjusting parameters of the model according to the characteristic weights;

evaluating the performance of the model using the validation set;

and deploying the trained model to actual Internet of things equipment.

In an implementation, we have chosen a classification model based on decision trees. First, we train the model using the full dataset. For example, we have 1000 data samples, each sample having 5 features. After model training is completed, we evaluate the contribution of each feature to model performance. In decision trees, the "importance" of a feature is a common indicator. Assuming our model shows that home appliance power consumption is the most important feature to predict whether a user is at home. Based on the importance of the features, we can adjust the parameters of the model, such as the maximum depth of the decision tree, to better utilize the most important features. If the power consumption is high, we may set a deeper tree to take full advantage of this information. After model tuning, we use a separate validation set to evaluate the performance of the model. Assuming that the accuracy of the model after the parameters are adjusted on the verification set is improved by 5%, once the model performs well on the verification set and passes various tests, the model can be deployed on actual intelligent household equipment. The device will collect data in real time and use the model to predict whether the user is at home.

Specifically, a prediction model for intelligent home internet of things equipment is used for predicting whether a user is at home. The dataset included the following features: power consumption (W) of home appliances, door and window sensor status (on/off), wi-Fi connected status (connected/disconnected) of mobile device, ambient light intensity (LUX), indoor temperature (°c). Characteristic value: the household electrical appliance power consumption is 500W, the door and window sensor state is 1 (open), the Wi-Fi connection state of the mobile device is 1 (connected), the ambient light intensity is 200 LUX, and the indoor temperature is 22 ℃.

Through model training and evaluation we get the weights for each feature. For example, the weight of power consumption is 0.7, the door and window sensor is 0.1, the wi-Fi connection is 0.1, the light intensity is 0.05, and the temperature is 0.05.

Based on the feature weights, we adjust the parameters of the model, such as the maximum depth of the decision tree, making the model more dependent on the power consumption feature. On the adjusted model, we tested with an independent validation set and found an improvement in accuracy from 80% to 85%.

Specifically, by selecting a model and adjusting model parameters according to data characteristics, the accuracy of the model in predicting unknown data can be improved. When the trained model is deployed on the Internet of things equipment, the data can be processed more intelligently, and the risk of data leakage is reduced. Accurate predictions can help the internet of things devices provide more intelligent services. The model can quickly respond to data changes and support real-time decision making.

Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.

The foregoing description is only of the preferred embodiments of the invention and is not intended to limit the invention; various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. The fast access method for the protocol adaptation of the Internet of things equipment is characterized by comprising the following steps of:

According to the weight and the protocol type of the Internet of things equipment, an adapter and a loading protocol analysis module are selected, protocol conversion is carried out by adopting a self-adaptive adapter, and data of different protocols are converted into a uniform format;

2. The method for quickly accessing protocol adaptation of internet of things equipment according to claim 1, wherein the step of scanning the internet of things equipment with high priority through weights, the step of obtaining the type and the position of any one of the internet of things equipment, and the step of assigning a weight value to any one of the internet of things equipment according to the type and the function of the internet of things equipment comprises the steps of:

3. The method for quickly accessing protocol adaptation of an internet of things device according to claim 2, wherein constructing an adaptive network including a first edge node and a second edge node of the internet of things device according to weight and location information of any one of the internet of things devices comprises:

collecting information of the Internet of things equipment;

according to the weight and the position information of the Internet of things equipment, selecting the Internet of things equipment with high stability as a first edge node, and selecting a superior geographic position as a second edge node;

and configuring network equipment for the first edge node and the second edge node.

4. The method for fast accessing protocol adaptation of internet of things equipment according to claim 3, wherein the local data processing by the first edge node comprises:

Collecting data from the internet of things equipment in real time through an adapter;

performing preliminary processing on the acquired data on the first edge node to obtain a second processing result;

compressing the data after preliminary processing;

periodically synchronizing locally stored data to the second edge node;

performing real-time analysis on the first edge node;

making a local decision on the first edge node according to the real-time analysis and the second processing result;

5. The method for fast access for internet of things device protocol adaptation according to claim 4, wherein the second edge node performing collaborative computing and data aggregation comprises:

decompressing the transmitted data at the second edge node;

Performing a collaborative computing task on the second edge node;

6. The method for fast accessing protocol adaptation of an internet of things device according to claim 5, wherein the protocol conversion using the adaptive adapter comprises:

according to the equipment type and the protocol, the adapter loads a corresponding protocol analysis module;

verifying whether the converted data meets the compatibility requirement of a target system;

7. The method for fast accessing protocol adaptation of internet of things equipment according to claim 6, wherein the loading of the corresponding protocol parsing module by the adaptive adapter according to the equipment type and the protocol comprises:

metadata defining a device type and a communication protocol;

testing and verifying the protocol analysis module to obtain a test result;

8. The method for quickly accessing protocol adaptation of internet of things equipment according to claim 7, wherein evaluating a current data transmission mechanism according to the collected real-time data and historical data and the weight value of the internet of things equipment, and optimizing data transmission parameters, network bandwidth and buffer area according to the evaluation result comprises:

9. The method for fast access for internet of things device protocol adaptation according to claim 8, wherein extracting features of data according to weights of the data to obtain a data set comprises:

using the extracted features, the dataset is constructed.

10. The method for quickly accessing protocol adaptation of internet of things equipment according to claim 9, wherein by training the data set training model, adjusting parameters of the model and training process according to weight of data comprises:

Selecting a model and training the model through the data set;

adjusting parameters of the model according to the characteristic weights;

evaluating the performance of the model using the validation set;

and deploying the trained model to actual Internet of things equipment.