An edge host system and its interaction method and scene analysis method

Abstract

An edge host system includes an edge host, a protocol device, a server, and smart home devices; the edge host is provided with a protocol processing module, a data management module, a scene linkage module, and a security protection module; the smart home device can feedback its status via a communication link, and the protocol device can convert the data of smart home devices with different communication methods and protocols into a unified format; the smart home device reports data to the protocol device, which is then converted by the protocol device and reported to the edge host, and the edge host reports the status of the smart home device to the server after processing the data; the protocol processing module pre-processes the data reported by each protocol device into a data structure, and the data management module then fuses and processes the data, and the processed data is reported after being screened by the security protection module, and the scene linkage module extracts key data and performs local user behavior analysis to generate corresponding intelligent scenes. The present application provides an edge host system with high reliability, safe use, and easy expansion.

Description

Technical Field

[0001] The present invention relates to smart home technology, and in particular to an edge host system and an interaction method and a scene analysis method thereof. Background Art

[0002] With the rapid development of the Internet of Things, the increasing popularity of wireless networks, and the wave of smart home devices, more and more terminal devices are connected to the Internet. The influx of a large amount of terminal device data into the Internet will inevitably lead to a large amount of computing power consumption in data centers, directly or indirectly affecting the user experience.

[0003] In smart home usage scenarios, it is often necessary to protect user privacy, achieve real-time interactive response between devices and people, analyze user behavior, and realize the intelligence of corresponding terminal device scenarios. Summary of the invention

[0004] In order to solve the defects existing in the above-mentioned prior art, the present application provides an edge host system and its interaction method and scene analysis method, which are highly reliable, safe to use, respond in real time, are easy to expand, can connect to various types of smart home terminal devices, and can adapt to various types of usage scenarios.

[0005] In order to achieve the above technical effects, the specific technical solutions of the present invention are as follows:

[0006] An edge host system includes an edge host, a protocol device, a server, and a smart home device; the edge host is provided with a protocol processing module, a data management module, a scene linkage module, and a security protection module.

[0007] The smart home devices are of various types, and the communication connection method can be wired connection (RS232, RS485, MODBUS, KNX, etc.) or wireless (infrared, WIFI, Bluetooth, LoRa, RF, Zigbee, NB-IOT, etc.). The device is controlled by receiving specific instructions and can feedback the device status via a communication link.

[0008] The protocol device has different communication modes in hardware, and the protocol device matches the basic circuit design of the corresponding smart home device communication; in terms of software protocol, the protocol of each type of smart home device adopts the corresponding protocol parsing method. The protocol device converts the data of smart home devices with different communication modes and protocols into raw data messages with a unified format and easy processing, and reports them to the edge host.

[0009] The server, as the service end for all users, receives data processed by the edge host, provides users with online scene deployment, user data storage, user command distribution, intelligent data analysis, and online equipment upgrade services, and can recommend customized scenes for target users.

[0010] The edge host system, its hardware link layer: smart home devices report data to the protocol device, which is then converted by the protocol device into data that is easy to process and reported to the edge host. The edge host processes the data and then reports the device status to the server; edge signal processing layer: the protocol processing module of the edge host pre-processes the data reported by various protocol devices into a data structure that is easy to manage, and then the data management module integrates and processes the data. The processed data is screened by the security protection module and then reported. During this period, the intelligent scene linkage module extracts key data and performs local user behavior analysis to generate corresponding intelligent scenes.

[0011] The edge host has the following easy-to-expand feature: when an edge host in a home environment cannot meet the user's needs for data storage or smart home device connection, the edge host preset publishing and subscription mechanism interacts with computing power and storage redundancy, thereby achieving load balancing and computing power optimization between edge hosts.

[0012] The security protection module has the following security protection features: the hardware layer is equipped with a media stream data encryption chip to ensure the security of user audio and video monitoring data transmission; the software layer encrypts, locks, stores at the edge and processes at the edge for user privacy and sensitive data to reduce the outflow of user smart home data and enhance the privacy of the system.

[0013] The working characteristics of the data management module are as follows: on the one hand, it integrates the data between smart home terminal devices, extracts the relevant data features and constructs a data model, and realizes the data association between various smart home terminal devices based on the results output by the data model; on the other hand, it compresses and stores the streaming media data generated by users, and counts, classifies and records the user behavior data.

[0014] The working characteristics of the scene linkage module are as follows: based on the user behavior data processed by the data management module, the high-frequency features and low-frequency features are extracted to analyze the user's usage preferences for different smart home terminal devices; combined with the usage associations between smart home devices, the parameter characteristics of the user's usage environment, and mandatory scenes (such as fire scenes, security scenes), customized scenes that conform to the user's usage habits are recommended to the user.

[0015] At the same time, the present invention also provides an edge host system interaction method, and the edge host system based on the above technical solution includes the following steps:

[0016] a. The smart home device status is reported to the protocol device, which converts it into a format that can be parsed by the edge host;

[0017] b. The edge host receives the smart home device data forwarded by the protocol device and determines whether the data is valid. If the data is valid, it proceeds to step c. If the data is invalid, it discards the current data and continues to wait for the next status report;

[0018] c. Execute the security protection module to encrypt and de-privatize the data;

[0019] d. Report the current status to the server, record the data locally on the edge host, manage the local data, and then call the scene linkage module to analyze the locally stored user behavior data, generate corresponding linkage instructions based on the feature status reported by the smart home device, and send them to control the smart home device;

[0020] e. The user command is sent to the edge host. If the user command is valid, the process goes to step f. If the command is invalid, the user is prompted that the current command is invalid and waits for a new command to be issued.

[0021] f. After being converted into a parsable data format by the protocol processing module, the process goes to step c to perform user behavior data reporting and analysis, and step g is performed at the same time;

[0022] g. The protocol device converts the data processed by the edge host into control instruction codes corresponding to controlling the smart home devices, and sends them to control the smart home devices.

[0023] At the same time, the present invention also provides an edge host system scenario analysis method, which can realize appropriate scenario type analysis for different smart home device types, different environmental variables, different smart home device working time periods and network connection conditions.

[0024] The edge host completes each user behavior Q through the following functions n Record the number of actions, each time period of user action is T0, and the number of actions is q:

[0025]

[0026] Count the number of various user control behavior actions {Q1, Q2…Q n} to dataset A:

[0027] A∈{Q1,Q2,...Q n}(n∈[1,+∞])

[0028] Then extract the frequency in set A that is higher than the threshold F h The user behavior instructions are stored in the preprocessing data set B:

[0029]

[0030] Extract the parameters in the B data set with similar operation frequency and analyze the device correlation in the time domain. n The time point of the user behavior is T an , let B n The time point of the device is T bn , operation time correlation parameter set ΔEn:

[0031]

[0032] Add the parameter ΔEn between smart home devices to the data set E, take the median ΔEz of the data in the E set as the expected value, and calculate the root mean square difference RMSD of the E set n :

[0033]

[0034] Finally, the device association θL is generated by the product of the root mean square difference RMSDn and the number of high-frequency actions of the device Bn:

[0035] θL=B n *RMSDn(n∈[1,+∞])

[0036] According to the above method, the coupling degree between the control behaviors of different smart home devices of users can be analyzed through the device association degree. The parameters in the B data set in the above device calculation method can be replaced by climate type, fire protection related parameters, environmental sensor reporting parameters, etc. The system uses the coupling degree as a direct parameter for the linkage of intelligent scene devices and recommends various linked intelligent scenes for users. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present application is further described in detail below through specific implementation methods in conjunction with the accompanying drawings.

[0038] Figure 1 It is a schematic diagram of the edge host system framework in the present invention;

[0039] Figure 2 It is a flow chart of the edge host system interaction method in the present invention;

[0040] Figure 3 This is a flow chart of the edge host system scenario analysis method in the present invention;

[0041] Among them, 1. edge host; 11. protocol processing module; 12. data management module; 13. scene linkage module; 14. security protection module; 2. protocol device; 21. protocol device 1; 22. protocol device 2; 23. protocol device 3; 2n. protocol device n; 3. server; 4. smart home device; 41. light; 42. monitoring; 43. air conditioner; 4n. device n. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of this embodiment more clear, the technical solution in this embodiment will be clearly and completely described below in conjunction with the drawings in this embodiment. Obviously, the described embodiment is a part of the embodiment of this application, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] In the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", "fixation" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0044] Example

[0045] like Figure 1 As shown, an edge host system includes an edge host 1, a protocol device 2, a server 3, and a smart home device 4; the edge host 1 is provided with a protocol processing module 11, a data management module 12, a scene linkage module 13, and a security protection module 14, wherein the protocol device 2 includes a protocol device 1 21, a protocol device 2 22, a protocol device 3 23, and a protocol device n2n, and the smart home device 4 includes a lamp 41, a monitor 42, an air conditioner 43, and a device n4n. Each protocol device (21, 22, 23, 2n) is connected to the same type of smart home device, the protocol device 1 21 corresponds to the lamp 41, the protocol device 2 22 corresponds to the monitor 42, the protocol device 3 23 corresponds to the air conditioner 43, the protocol device n corresponds to the device n, and so on, it can be connected to various types of smart home devices through different protocol devices.

[0046] The edge host system and the smart home device reporting process are as follows: the protocol device (here refers to each protocol device) will record the current working status of the corresponding smart home device in real time. The edge host sends a scan status instruction to the protocol device every time period, and the protocol device reports the working status of the currently connected smart home device.

[0047] In this edge host system, the edge host will analyze and process user behavior, integrate the de-privacy encrypted data into messages and report them to the server. The server will build a user-related database by parsing the messages; it will back up the basic user data on the server to facilitate redeployment after changes in user usage scenarios.

[0048] The edge host system, the scene linkage module is responsible for identifying user behavior characteristics and device linkage characteristics from the data, and building a behavior analysis model based on the correlation of various feature values. Analysis of user habits: For example, users often need to open the curtains during a certain period of time in the morning on weekdays. The edge host recommends such scheduled execution scenes to users based on their habits. Analysis of the correlation between devices: For example, when opening the curtains, it is often necessary to open the corresponding windows. The scene linkage module will recommend users to bind such actions to achieve device linkage.

[0049] like Figure 2 The figure shows a flow chart of the edge host system interaction method, and the interaction process is as follows:

[0050] a. The smart home device status is reported to the protocol device, which converts it into a format that can be parsed by the edge host;

[0051] b. The edge host receives the smart home device data forwarded by the protocol device and determines whether the data is valid. If the data is valid, it proceeds to step c. If the data is invalid, it discards the current data and continues to wait for the next status report;

[0052] c. Execute the security protection module to encrypt and de-privatize the data;

[0053] d. Report the current status to the server, record the data locally on the edge host, manage the local data, and then call the scene linkage module to analyze the locally stored user behavior data, generate corresponding linkage instructions based on the feature status reported by the smart home device, and send them to control the smart home device;

[0054] e. The user instruction is sent to the edge host. If the user instruction is valid, go to step f. If the instruction is invalid, the user is prompted that the current instruction is invalid and waits for a new instruction to be issued.

[0055] f. After being converted into a parsable data format by the protocol processing module, the process goes to step c to perform user behavior data reporting and analysis, and step g is performed at the same time;

[0056] g. The protocol device converts the data processed by the edge host into control instruction codes corresponding to controlling the smart home devices, and sends them to control the smart home devices.

[0057] like Figure 3 As shown, the flow chart of the edge host system scenario analysis method is as follows:

[0058] 1) The edge host completes each user behavior Q through the following functions n Record the number of actions, each time period of user action is T0, and the number of actions is q:

[0059]

[0060] 2) Count the number of each type of user control behavior {Q1, Q2…Qn} to data set A:

[0061] A∈{Q1,Q2,...Q n}(n∈[1,+∞])

[0062] 3) Extract the frequency in set A that is higher than the threshold F h The user behavior instructions are stored in the preprocessing data set B:

[0063]

[0064] 4) Extract the parameters in the B data set with similar operation frequency and analyze the device correlation in the time domain. n The time point of the user behavior is T an , let B n The time point of the device is T bn , operation time correlation parameter set ΔEn:

[0065]

[0066] 5) Add the parameter ΔEn between smart home devices to the data set E, take the median ΔEz of the data in the E set as the expected value and calculate the root mean square difference RMSD of the E set n :

[0067]

[0068] 6) Finally, the root mean square difference RMSD n The number of high-frequency actions of the equipment B n The product of generates device association θ L :

[0069] θL=B n *RMSD n (n∈[1,+∞]).

[0070] According to the above method, the coupling degree between the control behaviors of different smart home devices of users can be analyzed through the device association degree. The parameters in the B data set in the above device calculation method can be replaced by climate type, fire protection related parameters, environmental sensor reporting parameters, etc. The system uses the coupling degree as a direct parameter for the linkage of intelligent scene devices and recommends various linked intelligent scenes for users.

[0071] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art, according to the concept of the present invention, some simple deductions, modifications or substitutions can be made.

Claims

1. An edge host system, It is characterized in that The edge host system includes an edge host, a protocol device, a server, and a smart home device; the edge host is provided with a protocol processing module, a data management module, a scene linkage module, and a security protection module. The smart home device can be controlled by receiving instructions and feedback its status via a communication link. The protocol device can convert the data of smart home devices with different communication modes and protocols into a unified format for easy processing of the original data message to the edge host. The server serves as a service end for all users, which receives the data processed by the edge host and provides users with online scene deployment, user data storage, user instruction distribution, and data analysis. It can recommend customized scenes for target users; the smart home device reports data to the protocol device, which is then converted by the protocol device into data that is easy to process and reported to the edge host. The edge host reports the state of the smart home device to the server after processing the data; the protocol processing module of the edge host pre-processes the data reported by various protocol devices into a data structure that is easy to manage, and then the data management module merges and processes the data. The processed data is screened by the security protection module and reported. During this period, the scene linkage module extracts key data and performs local user behavior analysis to generate corresponding intelligent scenes; The scene linkage module extracts high-frequency and low-frequency features based on the user behavior data processed by the data management module to analyze the user's usage preferences for different smart home devices; combines the usage associations between smart home devices, parameter features of the user's usage environment, and mandatory scenes to recommend customized scenes that meet the user's usage habits; When an edge host in a home environment cannot meet the user's needs for data storage or smart home device connection, the edge host's preset publishing and subscription mechanism interacts with computing power and storage redundancy to achieve load balancing and computing power optimization between edge hosts. The edge host system performs the following scenario analysis method, which includes: 1) The edge host completes the record of the number of user control behavior actions Qn, and each time period for the number of user control behavior actions is T0. 2) Count the number of each type of user control behavior {Q1, Q2…Qn} to data set A: A∈{Q 1 , Q 2 ,...Q n}, n∈[1,+∞] 3) Then extract the user behavior instructions with a frequency higher than the threshold Fh from set A and store them in the preprocessing data set B. If the median value of set A is higher than the threshold Fh, record the valid high-frequency feature data in data set B; 4) Extract the parameters in the B data set with similar operation frequency, analyze the device association in the time domain, set the user behavior time point as Tusri, set the device time point as Tdevi, and calculate the operation time association parameter ΔEn, ； 5) Add the parameter ΔEn of the operation time correlation to the data set E, take the median ΔEz of the data in the E set as the expected value, and calculate the root mean square difference RMSD of the E set: ； 6) Finally, the device association θLn is generated by multiplying the root mean square difference RMSD and the number of high-frequency actions of the device Bn: θLn=Bn*RMSD.

2. The edge host system as claimed in claim 1, It is characterized in that The security protection module is used to ensure the security of user audio and video monitoring data transmission, and encrypts, locks, stores at the edge and processes at the edge for user privacy and sensitive data, thereby reducing the outflow of user smart home data and enhancing the privacy of the system.

3. The edge host system as claimed in claim 1, It is characterized in that On the one hand, the data management module integrates the data between smart home devices, extracts relevant data features and constructs a data model, and realizes the data association between various smart home devices based on the results output by the data model; on the other hand, it compresses and stores the streaming media data generated by users, and counts, classifies and records user behavior data.

4. An edge host system interaction method, based on the edge host system according to any one of claims 1 to 3, It is characterized in that The following steps are involved: a. The smart home device status is reported to the protocol device, which converts it into a format that can be parsed by the edge host; b. The edge host receives the smart home device data forwarded by the protocol device and determines whether the data is valid. If the data is valid, it proceeds to step c. If the data is invalid, it discards the current data and continues to wait for the next status report; c. Execute the security protection module to encrypt and de-privatize the data; d. Report the current status to the server, record the data locally on the edge host, manage the local data, and then call the scene linkage module to analyze the locally stored user behavior data, generate corresponding linkage instructions based on the feature status reported by the smart home device, and send them to control the smart home device; e. The user command is sent to the edge host. If the user command is valid, the process goes to step f. If the command is invalid, the user is prompted that the current command is invalid and waits for a new command to be issued. f. After being converted into a parsable data format by the protocol processing module, the process goes to step c to perform user behavior data reporting and analysis, and step g is performed at the same time; g. The protocol device converts the data processed by the edge host into control instruction codes corresponding to controlling the smart home devices, and sends them to control the smart home devices.

Images