Edge-end-oriented low-delay data pushing system and method thereof
By configuring and matching messages on the edge, the accuracy and delay problems of data push systems in the existing technology are solved, faster and more accurate message push is achieved, reducing the burden on the central server and improving the user experience.
Patent Information
- Application Number
- CN202411088064.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-08-09
AI Technical Summary
The existing data push system has problems such as inaccurate push objects, pushing by cloud central servers, small throughput of message push, and small support groups, resulting in extensive information push, affecting the experience and effectiveness of message push, and having relatively high delays. high.
Design a low-latency data push system for the edge end. By configuring and matching messages on the edge end, it reduces data transmission delay, improves response speed, and increases the matching degree of customer message push through the edge end matching algorithm to achieve the reach The object is more accurate and fast.
It realizes that message information is closer to the data source, reduces data transmission delay, improves response speed, reduces the burden on the central server, enhances the matching degree and low latency of customer message push, and improves user experience.
Smart Images

Figure CN119094603A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a low-latency data push system and method for an edge terminal. Background Art
[0002] Edge computing is a distributed computing architecture whose core idea is to transfer data processing and analysis from centralized data centers to the edge of the network; APP data push is a very important means of reaching users, and it plays an important role in increasing product activity, improving functional usage experience, improving user stickiness, and improving user retention rate.
[0003] At this stage, data push is particularly prominent when operations are concentrated on activities. The amount of data pushed is often in the tens of millions or even hundreds of millions. The push tasks must be received by users as quickly as possible. However, the traditional message push has problems such as inaccurate push targets, push by cloud center servers, low message push throughput, and a small number of people supported by message push. In particular, the matching degree of push condition information for different objects is not up to standard, resulting in a large amount of information being pushed in a flood-like manner, which not only brings unnecessary harassment to customers, but also affects the experience and efficiency of message push, making data push high-latency. For this reason, we propose a low-latency data push system and method for the edge. Summary of the invention
[0004] In view of the above problems existing in the existing data processing, the present invention is proposed.
[0005] Therefore, one of the objects of the present invention is to provide a low-latency data push system and method for the edge. The present invention pushes messages to the edge for analysis and then matches the corresponding edge and node, and then pushes the messages to each node one by one, thereby reducing the delay in data transmission, improving the response speed, and reducing the burden on the central server. By configuring the matching algorithm at the edge, the matching degree of customer message push can be increased and the latency can be reduced, making the reaching of the object more accurate and faster.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0007] On the one hand, the present invention provides a low-latency data push system for edge terminals, including a cloud processing center and an edge computing processing push system, wherein the edge computing processing push system includes:
[0008] The edge data collection module is used to collect, push requests, aggregate, and transmit node message data to the cloud processing center through scheduled task scheduling at the distributed edge;
[0009] An edge matching push calculation module is used to receive message data of the node of the edge data acquisition module, complete the matching calculation, placeholder replacement, paging sending and statistical information recording of the message data, and the matching push calculation module includes an edge matching analysis unit for calculating the comprehensive matching score of the network topology structure of different nodes of the current edge computing end, an edge node analysis unit for calculating the weight correction coefficient of different nodes of the current edge computing end, and an edge node membership analysis unit for analyzing the membership of different nodes of the edge end; wherein the edge end forms a push message information analysis result according to the comprehensive matching score of the network topology structure of different nodes;
[0010] The edge configuration module is used to respond to the push message information analysis results of the edge matching push calculation module, and push message information in batch or single form after matching the corresponding nodes according to the push message information analysis results. The edge configuration module includes a batch event processing unit and a single event processing unit; that is, the edge configuration module is used to receive message trigger events and corresponding push message information analysis results, and calculate whether the edge node batch or single message push trigger requirements are met according to the conditions of the configured trigger event to push the message.
[0011] As a preferred solution of the present invention, wherein: the push request of the edge data acquisition module includes user identification, business identification, push scenario and message information, the edge matching push calculation module configures the corresponding edge push strategy according to different push scenarios, and the push strategy configures the edge push channel, the number of edge notification devices of the push channel and the range to execute push according to the push scenario, that is, according to the protocols of different push edge channels, the relevant parameters required for the edge push of each channel are assembled, and the push edge channel services of each push are called to execute push.
[0012] As a preferred solution of the present invention, wherein: the message push conditions of the batch event processing unit are based on batch event attributes, support complex Boolean expressions of event attributes, the batch push conditions are freely configured, and the freely configured conditions are composed of multiple sub-items; the message push conditions of the single event processing unit are based on single event attributes, support complex Boolean expressions of event attributes, the single push conditions are freely configured, and the freely configured conditions are composed of multiple sub-items.
[0013] As a preferred solution of the present invention, wherein: the edge configuration module also includes an edge message queue matching unit, the edge message queue matching unit divides the topic of the message queue into two dimensions, the two dimensions are the identification with the topic and the priority with the topic, the priority of the topic is preset by default as high priority message, medium priority message and low priority message, the edge message queue matching unit is composed of identification information and priority, for messages that cannot match the identification information, a low priority message queue is used, and for messages that cannot match the identification information priority, a high priority message queue is used.
[0014] As a preferred solution of the present invention, the cloud processing center uses a distributed SQL database to record the triggering status of the message configuration of each node on the edge and perform data partition caching.
[0015] As a preferred solution of the present invention, the edge computing processing push system also includes a message push record query module and a log management module, which are connected to the edge matching push computing module and are used to query the message push event processing status, message push status, message push content and time of the edge.
[0016] As a preferred solution of the present invention, the edge end matching analysis unit is used to determine the network topology structure of different nodes of the current edge computing end, and calculate the comprehensive matching score of the network topology structure of different nodes of the current edge computing end according to the predicted scores of different nodes of the current edge computing end and the matched attribute value information, as shown in the following formula:
[0017]
[0018] in, m Determine the number of k-th nodes in the domain for the edge computing end, score(Q) k To determine the comprehensive score of nodes in the domain, α k is the weight correction coefficient of the kth node, β k is the attribute value of the kth node, w k is the weight coefficient of the kth node, score(q) k Score the prediction of the kth node;
[0019] The node analysis unit is used to calculate the weight correction coefficients of different nodes at the current edge computing end, as shown in the following formula:
[0020]
[0021] Among them, A and B represent the identification data subsets in the edge topology chain, A i ∩B iRepresents the i-th subset A of the identified data in the edge topology chain i and the i-th subset B of the identification data i The intersection of k i ), p(B k i ) represent the matching probability corresponding to the k-th node identification data subset of the edge topology chain;
[0022] P(A) represents the corrected probability of identifying a data subset by the edge topology chain, The empty set symbol.
[0023] As a preferred solution of the present invention, the edge node membership analysis unit is used to analyze the membership of different edge nodes to reflect the attribute values of different nodes, and calculate the proportion p of different edge nodes according to different preset indicators through the formula k,ij , as follows:
[0024]
[0025] Where n is the number of the ith subset of the identified data in the edge topology chain, y k,ij Identify the value of the data in the i-th subset under the j-th index for the k-th node in the edge topology chain, p k,ij The weight of the i-th subset of the data identified by the k-th node in the edge topology chain under the j-th indicator;
[0026] The entropy value E of the jth indicator is calculated by j , as follows:
[0027]
[0028] Among them, ln(p ij ) is p ij The natural logarithm of ln(n) is the natural logarithm of n. lnp ij =0; the weight coefficient w of the kth node at the edge under the jth index is calculated by the formula k , and the membership R of the edge corresponding to the kth node under the jth index is calculated by ij , as follows:
[0029]
[0030] Among them, E k,j is the entropy value of the jth indicator of the kth node at the edge.
[0031] As a preferred solution of the present invention, after the events triggered by the preset messages are formed into expressions, the prediction scores of different nodes are analyzed according to the number of triggers, the number of messages and the trigger time interval, as shown in the following formula:
[0032]
[0033] Among them, X represents the message data with the most occurrences in the kth node, K represents the change value of the message data collected by the current node as the trigger time interval changes, ω i represents the weight value of the message data collected by the current node for the i-th time, and N represents the mixed weight, which represents the weight of the message data with the largest number of occurrences of the event triggered by the message at the current node.
[0034] On the one hand, the present invention provides a method for a low-latency data push system for an edge terminal, comprising the following steps:
[0035] Step 1: The distributed edge end collects, pushes requests, aggregates, and transmits message data from nodes to the cloud processing center through scheduled task scheduling;
[0036] Step 2: Receive the message data of the node, complete the matching calculation, placeholder replacement, paging sending and statistical information recording of the message data, calculate the comprehensive matching score of the network topology structure of different nodes of the current edge computing end, calculate the weight correction coefficient of different nodes of the current edge computing end and analyze the membership of different nodes of the edge end. The edge end forms the push message information analysis result according to the comprehensive matching score of the network topology structure of different nodes;
[0037] Step three, respond to the analysis results of the push message information, and push the message information in batches or in a single form after matching the corresponding nodes according to the analysis results of the push message information, that is, calculate whether the conditions of the configured trigger event meet the edge node batch or single message push trigger requirements for push work.
[0038] Beneficial effects of the present invention: The present invention configures message push to the edge end for analysis and then matches the corresponding edge end and node, and then pushes node messages one by one, so that the message information is closer to the data source, thereby reducing the delay in data transmission, improving the response speed, and reducing the burden on the central server. In addition, by configuring the matching algorithm at the edge end, the matching degree of customer message push can be increased and the delay can be reduced, making the target reach more accurate and faster. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:
[0040] Figure 1 It is a schematic diagram of the modular structure of the present invention;
[0041] Figure 2 This is an application scenario diagram of the present invention;
[0042] Figure 3 is a flow chart of the method of the present invention;
[0043] Numbers in the figure: 10, edge data acquisition module; 20, edge matching push calculation module; 30, edge configuration module; 21, edge matching analysis unit; 22, edge node analysis unit; 23, edge node membership analysis unit; 31, batch event processing unit; 32, single event processing unit; 33, edge message queue matching unit. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.
[0045] Reference Figure 1-Figure 3 , is an embodiment of the present invention, which provides a low-latency data push system and method for edge terminals. It includes a cloud processing center and an edge computing processing push system, and the edge computing processing push system includes:
[0046] The edge data collection module 10 is used for the distributed edge end to collect, push requests, aggregate, and transmit message data of nodes to the cloud processing center through scheduled task scheduling;
[0047] The edge matching push calculation module 20 is used to receive the message data of the node of the edge data acquisition module 10, complete the matching calculation, placeholder replacement, paging sending and statistical information recording of the message data, and the matching push calculation module includes an edge matching analysis unit 21 for calculating the comprehensive matching score of the network topology structure of different nodes of the current edge computing end, an edge node analysis unit 22 for calculating the weight correction coefficient of different nodes of the current edge computing end, and an edge node membership analysis unit 23 for analyzing the membership of different nodes of the edge end; wherein the edge end forms a push message information analysis result according to the comprehensive matching score of the network topology structure of different nodes calculated;
[0048] The edge configuration module 30 is used to respond to the push message information analysis results of the edge matching push calculation module 20, and push message information in batch or single form after matching the corresponding nodes according to the push message information analysis results. The edge configuration module 30 includes a batch event processing unit 31 and a single event processing unit 32; that is, the edge configuration module 30 is used to receive message trigger events and corresponding push message information analysis results, and calculate whether the edge node batch or single message push trigger requirements are met according to the conditions of the configured trigger event to perform push work.
[0049] The push request of the edge data collection module 10 includes user identification, business identification, push scenario and message information. The edge matching push calculation module 20 configures the corresponding edge push strategy according to different push scenarios. The push strategy configures the edge push channel, the number of edge notification devices of the push channel and the range of execution of push according to the push scenario. That is, according to the protocols of different push edge channels, the relevant parameters required for the edge push of each channel are assembled, and the push edge channel services of each push are called to execute push.
[0050] This embodiment needs to be further explained that the message push conditions of the batch event processing unit 31 are based on batch event attributes, support complex Boolean expressions of event attributes, and the batch push conditions are freely configured. The freely configured conditions are composed of multiple sub-items; the message push conditions of the single event processing unit 32 are based on single event attributes, support complex Boolean expressions of event attributes, and the single push conditions are freely configured. The freely configured conditions are composed of multiple sub-items.
[0051] This embodiment requires that the edge configuration module 30 also includes an edge message queue matching unit 33. The edge message queue matching unit 33 divides the topic of the message queue into two dimensions. The two dimensions are the identification with the topic and the priority with the topic. The priority of the topic is preset as high priority message, medium priority message and low priority message by default. The edge message queue matching unit 33 consists of identification information and priority. For messages that cannot match the identification information, a low priority message queue is used. For messages that cannot match the identification information priority, a high priority message queue is used.
[0052] In this embodiment, the cloud processing center uses a distributed SQL database to record the triggering status of each node message configuration at the edge and perform data partition caching.
[0053] In addition, the edge computing processing push system also includes a message push record query module 40 and a log management module 50, which are connected to the edge matching push computing module 20 and are used to query the message push event processing status, message push status, message push content and time of the edge.
[0054] It should be emphasized in this embodiment that, due to the problems of inaccurate push targets, push by cloud center server, and low message push throughput in the current message push, the edge end used in this embodiment performs push service after node information matching according to the matching degree. The edge end matching analysis unit 21 is used to determine the network topology structure of different nodes of the current edge computing end, and calculates the comprehensive matching score of the network topology structure of different nodes of the current edge computing end according to the predicted scores of different nodes of the current edge computing end and the matched attribute value information, as shown in the following formula:
[0055]
[0056] Where m is the number of nodes in the k-th domain determined by the edge computing end, score(Q) k To determine the comprehensive score of nodes in the domain, α k is the weight correction coefficient of the kth node, β k is the attribute value of the kth node, w k is the weight coefficient of the kth node, score(q) k Score the prediction of the kth node;
[0057] The node analysis unit 22 is used to calculate the weight correction coefficients of different nodes at the current edge computing end, as shown in the following formula:
[0058]
[0059]
[0060] Among them, A and B represent the identification data subsets in the edge topology chain, A i ∩B i Represents the i-th subset A of the identified data in the edge topology chain i and the i-th subset B of the identification data i The intersection of k i ), p(B k i ) represent the matching probability corresponding to the k-th node identification data subset of the edge topology chain;
[0061] P(A) represents the corrected probability of identifying a data subset by the edge topology chain, The empty set symbol.
[0062] The edge node membership analysis unit 23 is used to analyze the membership of different edge nodes to reflect the attribute values of different nodes. The proportion p of different edge nodes is calculated according to different preset indicators through formula 5. k,ij ,as follows:
[0063]
[0064] Where n is the number of the ith subset of the identified data in the edge topology chain, y k,ij Identify the value of the data in the i-th subset under the j-th index for the k-th node in the edge topology chain, p k,ij The weight of the i-th subset of the data identified by the k-th node in the edge topology chain under the j-th indicator;
[0065] At the same time, the entropy value E of the jth indicator is calculated by the following formula: j ,as follows:
[0066]
[0067] Among them, ln(p ij ) is p ij The natural logarithm of ln(n) is the natural logarithm of n. lnp ij =0; The weight coefficient w of the kth node at the edge under the jth index is calculated by formula 7 k , and the membership R of the edge corresponding to the kth node under the jth index is calculated by formula 8 ij , as follows:
[0068]
[0069] Among them, E k,jis the entropy value of the jth indicator of the kth node at the edge.
[0070] Further based on the above description, after the events triggered by the preset messages are formed into expressions, the prediction scores of different nodes are analyzed according to the number of triggers, the number of messages and the trigger time interval, as shown in the following formula:
[0071]
[0072] Among them, X represents the message data with the most occurrences in the kth node, K represents the change value of the message data collected by the current node as the trigger time interval changes, ω i represents the weight value of the message data collected by the current node for the i-th time, and N represents the mixed weight, which represents the weight of the message data with the largest number of occurrences of the event triggered by the message at the current node.
[0073] Based on the above, this embodiment is summarized as follows: Message push configuration to the edge: The logic and algorithm of message push are deployed on the edge device instead of the traditional central server. Analysis and matching of edge and node: After the edge device receives the message, it analyzes and determines the edge and node that are most suitable for receiving these messages according to the preset rules or algorithms. Node message push: After determining the target node, the message is pushed directly on the edge, reducing the transmission distance of data in the network.
[0074] Furthermore, compared with the traditional message push method, this embodiment can achieve the following effects:
[0075] 1. Reduce latency: Messages are processed and pushed at the edge, which reduces the time for data transmission and thus reduces response time.
[0076] 2. Reduce the burden on the central server: The central server does not need to handle all message push tasks, which reduces its computing and storage pressure.
[0077] 3. Improve matching: By configuring more sophisticated matching algorithms on the edge, messages can be pushed to target users more accurately.
[0078] 4. Improve user experience: Fast and accurate message push can improve user satisfaction and interactivity.
[0079] like Figure 3 As shown, based on the above system structure, this embodiment also provides a method for a low-latency data push system for an edge terminal, including the following steps:
[0080] Step 1: The distributed edge end collects, pushes requests, aggregates, and transmits message data from nodes to the cloud processing center through scheduled task scheduling;
[0081] Step 2: Receive the message data of the node, complete the matching calculation, placeholder replacement, paging sending and statistical information recording of the message data, calculate the comprehensive matching score of the network topology structure of different nodes of the current edge computing end, calculate the weight correction coefficient of different nodes of the current edge computing end and analyze the membership of different nodes of the edge end. The edge end forms the push message information analysis result according to the comprehensive matching score of the network topology structure of different nodes;
[0082] Step three, respond to the analysis results of the push message information, and push the message information in batches or in a single form after matching the corresponding nodes according to the analysis results of the push message information, that is, calculate whether the conditions of the configured trigger event meet the edge node batch or single message push trigger requirements for push work.
[0083] Beneficial effects of the present invention: The present invention configures message push to the edge end for analysis and then matches the corresponding edge end and node, and then pushes node messages one by one, so that the message information is closer to the data source, thereby reducing the delay in data transmission, improving the response speed, and reducing the burden on the central server. In addition, by configuring the matching algorithm at the edge end, the matching degree of customer message push can be increased and the delay can be reduced, making the target reach more accurate and faster.
[0084] In summary, the present invention configures the message push to the edge end for analysis and then matches the corresponding edge end and node, and then pushes the node message one by one, so that the message information is closer to the data source, thereby reducing the delay in data transmission, improving the response speed, and reducing the burden on the central server. In addition, by configuring the matching algorithm at the edge end, the matching degree of the customer message push can be increased and the latency can be reduced, making the target reach more accurate and faster.
[0085] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
[0086] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.
[0087] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0088] Any process or method description in the flow chart or otherwise described herein can be understood to represent a module, fragment or portion of a code including one or more executable instructions for implementing the steps of a specific logical function or process. And the scope of the preferred embodiment of the present application includes other implementations, in which the functions may not be performed in the order shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the functions involved.
[0089] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium for use by an instruction execution system, apparatus or device such as a computer-based system, a system including a processor or other system that can fetch instructions from an instruction execution system, apparatus or device and execute instructions, or used in conjunction with these instruction execution systems, apparatuses or devices.
[0090] It should be understood that the various parts of the present application can be implemented with hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented with software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above embodiment method can be completed by instructing the relevant hardware through a program, which can be stored in a computer-readable storage medium, and when the program is executed, it includes one of the steps of the method embodiment or a combination thereof.
[0091] In addition, each functional unit in each embodiment of the present application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a disk or an optical disk, etc.
[0092] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A low-latency data push system for edge devices. It is characterized in that It includes a cloud processing center and an edge computing processing and push system, and the edge computing processing and push system includes: An edge data collection module (10) is used for the distributed edge end to collect, push requests, aggregate, and transmit node message data to a cloud processing center through scheduled task scheduling; An edge matching push calculation module (20) is used to receive message data of the node of the edge data acquisition module (10), complete matching calculation, placeholder replacement, paging transmission and statistical information recording of the message data, and the matching push calculation module includes an edge matching analysis unit (21) for calculating the comprehensive matching score of the network topology structure of different nodes of the current edge computing end, an edge node analysis unit (22) for calculating the weight correction coefficient of different nodes of the current edge computing end, and an edge node membership analysis unit (23) for analyzing the membership of different nodes of the edge end; wherein the edge end forms a push message information analysis result according to the comprehensive matching score of the network topology structure of different nodes calculated; The edge configuration module (30) is used to respond to the push message information analysis result of the edge matching push calculation module (20), and after matching the corresponding node according to the push message information analysis result, push message information in batch or single form. The edge configuration module (30) includes a batch event processing unit (31) and a single event processing unit (32); that is, the edge configuration module (30) is used to receive a message trigger event and a corresponding push message information analysis result, and calculate whether the edge node batch or single message push trigger requirement is met according to the condition of the configuration trigger event to perform push work.
2. A low-latency data push system for edge terminals as claimed in claim 1, It is characterized in that The push request of the edge data collection module (10) includes a user identifier, a service identifier, a push scenario and message information. The edge matching push calculation module (20) configures a corresponding edge push strategy according to different push scenarios. The push strategy configures an edge push channel, the number of edge notification devices of the push channel and the range to execute push according to the push scenario, that is, according to the protocols of different push edge channels, assembles the relevant parameters required for edge push of each channel, and calls each push edge channel service to execute push.
3. A low-latency data push system for edge terminals as claimed in claim 1, It is characterized in that The message push conditions of the batch event processing unit (31) are based on batch event attributes, support complex Boolean expressions of event attributes, and the batch push conditions are freely configured, and the freely configured conditions are composed of multiple sub-items; the message push conditions of the single event processing unit (32) are based on single event attributes, support complex Boolean expressions of event attributes, and the single push conditions are freely configured, and the freely configured conditions are composed of multiple sub-items.
4. A low-latency data push system for edge terminals as claimed in claim 1, It is characterized in that The edge configuration module (30) also includes an edge message queue matching unit (33), which divides the topic of the message queue into two dimensions, the two dimensions being a topic identifier and a topic priority, the topic priority being preset as a high priority message, a medium priority message and a low priority message by default, the edge message queue matching unit (33) being composed of identifier information and priority, a low priority message queue being used for messages that do not match the identifier information, and a high priority message queue being used for messages that do not match the identifier information priority.
5. A low-latency data push system for edge terminals as claimed in claim 1, It is characterized in that The cloud processing center uses a distributed SQL database to record the triggering status of each node message configuration at the edge and perform data partition caching.
6. A low-latency data push system for edge terminals as claimed in claim 1, It is characterized in that The edge computing processing push system also includes a message push record query module (40) and a log management module (50), which are connected to the edge matching push computing module (20) and are used to query the message push event processing status, message push status, message push content and time of the edge end.
7. A low-latency data push system for edge terminals as claimed in claim 1, It is characterized in that The edge end matching analysis unit (21) is used to determine the network topology structure of different nodes of the current edge computing end, and calculate the comprehensive matching score of the network topology structure of different nodes of the current edge computing end according to the prediction scores of different nodes of the current edge computing end and the matched attribute value information, as shown in formula (1): Where m is the number of nodes in the k-th domain determined by the edge computing end, score(Q) k To determine the comprehensive score of nodes in the domain, α k is the weight correction coefficient of the kth node, β k is the attribute value of the kth node, w k is the weight coefficient of the kth node, score(q) k Score the prediction of the kth node; The node analysis unit (22) is used to calculate the weight correction coefficients of different nodes of the current edge computing end, as shown in formula (2) to formula (4): Among them, A and B represent the identification data subsets in the edge topology chain, Ai∩B i Represents the i-th subset A of the identified data in the edge topology chain i With identification data i Subset B i The intersection of k i ), p(B k i ) represent the matching probability corresponding to the k-th node identification data subset of the edge topology chain; P(A) represents the corrected probability of identifying a data subset by the edge topology chain, The empty set symbol.
8. A low-latency data push system for edge terminals as claimed in claim 7, It is characterized in that The edge node membership analysis unit (23) is used to analyze the membership of different edge nodes to reflect the attribute values of different nodes, and calculate the proportion p of different edge nodes according to different preset indicators through formula (5). k,ij ,as follows: Where n is the number of the ith subset of the identified data in the edge topology chain, y k,ij Identify the value of the data in the i-th subset under the j-th index for the k-th node in the edge topology chain, p k,ij The weight of the i-th subset of the data identified by the k-th node in the edge topology chain under the j-th indicator; The entropy value E of the jth index is calculated by formula (6): j ,as follows: Among them, ln(p ij ) is p ij The natural logarithm of ln(n) is the natural logarithm of n. lnp ij =0; Calculate the weight coefficient w of the kth node at the edge under the jth index through formula (7) k , and the membership R of the edge corresponding to the kth node under the jth index is calculated by formula (8) ij ,as follows: Among them, E k,j is the entropy value of the jth indicator of the kth node at the edge.
9. A low-latency data push system for edge terminals as claimed in claim 7, It is characterized in that After presetting the events triggered by messages and forming expressions, the prediction scores of different nodes are analyzed according to the number of triggers, the number of messages and the trigger time interval, as shown in formula (9): Among them, X represents the message data with the most occurrences in the kth node, K represents the change value of the message data collected by the current node as the trigger time interval changes, ω i represents the weight value of the message data collected by the current node for the i-th time, and N represents the mixed weight, which represents the weight of the message data with the largest number of occurrences of the event triggered by the message at the current node.
10. A method for a low-latency data push system for edge terminals as claimed in claim 1, It is characterized in that The following steps are involved: Step 1: The distributed edge end collects, pushes requests, aggregates, and transmits message data from nodes to the cloud processing center through scheduled task scheduling; Step 2: Receive the message data of the node, complete the matching calculation, placeholder replacement, paging sending and statistical information recording of the message data, calculate the comprehensive matching score of the network topology structure of different nodes of the current edge computing end, calculate the weight correction coefficient of different nodes of the current edge computing end and analyze the membership of different nodes of the edge end. The edge end forms the push message information analysis result according to the comprehensive matching score of the network topology structure of different nodes; Step three, respond to the analysis results of the push message information, and push the message information in batches or in a single form after matching the corresponding nodes according to the analysis results of the push message information, that is, calculate whether the conditions of the configured trigger event meet the edge node batch or single message push trigger requirements for push work.
Citation Information
Patent Citations
Edge cloud system, data transmission method and device and storage medium
CN113726843A
Task pushing method based on edge-end data reasoning mode
CN114997399A
Internet of Things advertisement system, method and device based on edge computing
CN116485475A
Equipment data processing method and device based on edge computing and distributed computing
CN117376346A
Internet of Things middleware system design method supporting edge heterogeneous device access
CN118368323A
