CN111835830A

CN111835830A - Data perception system, method and device

Info

Publication number: CN111835830A
Application number: CN202010547545.XA
Authority: CN
Inventors: 彭木根; 武文斌; 闫实
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2020-06-16
Filing date: 2020-06-16
Publication date: 2020-10-27
Anticipated expiration: 2040-06-16
Also published as: CN111835830B

Abstract

The embodiment of the invention provides a data perception system, a method and a device, wherein data acquisition equipment acquires data as data to be perceived and sends the data to be perceived to edge computing equipment; the edge computing equipment performs data perception on the data to be perceived under the condition that the data to be perceived is determined to be data of a local network according to the acquisition cycle of the data to be perceived, and a first perception result is obtained; sending a first sensing result to data acquisition equipment; under the condition that the data to be perceived is determined to be data of a global network according to the acquisition period of the data to be perceived, sending the data to be perceived to a cloud server; the cloud server performs data perception on data to be perceived to obtain a second perception result; sending the second perception result to the edge computing device; the edge computing equipment sends a second sensing result to the data acquisition equipment; and the data acquisition equipment receives the first sensing result or the second sensing result. By applying the scheme provided by the embodiment of the invention, the data sensing efficiency can be improved.

Description

Data perception system, method and device

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to a data sensing system, method, and apparatus.

Background

In the prior art, after data acquisition equipment such as a mobile phone, a sensor and special measuring equipment acquires data, the acquired data is sent to edge computing equipment for data sensing, and the edge computing equipment feeds back a data sensing result to the data acquisition equipment. The computing power of the edge computing device is often weak, and under the condition that the received data is complex, more computing time is needed in the process of data perception, so that the efficiency of the edge computing device in data perception is low.

Disclosure of Invention

Embodiments of the present invention provide a data sensing system, method and apparatus to improve the processing efficiency of data to be sensed. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a data-aware system, where the system includes: the system comprises data acquisition equipment, edge computing equipment and a cloud server, wherein the data acquisition equipment, the edge computing equipment and the cloud server are positioned in a global network; wherein the content of the first and second substances,

the data acquisition equipment is used for acquiring data as data to be perceived and sending the data to be perceived to the edge computing equipment in the global network and the local network to which the data acquisition equipment belongs;

the edge computing equipment is used for receiving the data to be sensed sent by the data acquisition equipment; under the condition that the data to be perceived is determined to be data of a local network according to the acquisition cycle of the data to be perceived, carrying out data perception on the data to be perceived to obtain a first perception result; sending the first sensing result to the data acquisition equipment; under the condition that the data to be perceived is determined to be data of a global network according to the acquisition cycle of the data to be perceived, sending the data to be perceived to the cloud server;

the cloud server is used for receiving the data to be perceived sent by the edge computing equipment, performing data perception on the data to be perceived and obtaining a second perception result; sending the second perception result to the edge computing device;

the edge computing equipment is further used for receiving a second sensing result sent by the cloud server and sending the second sensing result to the data acquisition equipment;

the data acquisition device is further configured to receive the first perception result or the second perception result sent by the edge computing device.

In an embodiment of the present invention, the edge computing device is specifically configured to determine whether an acquisition period of the data to be perceived is smaller than a preset period, and if so, determine that the data to be perceived is data of a local network, otherwise, determine that the data to be perceived is data of a global network.

In an embodiment of the present invention, the edge computing device is further configured to determine a sensing accuracy of a first sensing result of data to be verified, and send the data to be verified to the cloud server if the sensing accuracy is lower than a preset accuracy; the data to be verified is: the data acquisition equipment acquires data to be sensed of the local network within a preset time length;

the cloud server is used for receiving data to be verified sent by the edge computing equipment, performing data perception on the data to be verified to obtain a third perception result, and sending the third perception result to the edge computing equipment;

the edge computing device is configured to receive a third sensing result sent by the cloud server, and send the third sensing result to the data acquisition device;

and the data acquisition equipment is used for receiving the third perception result sent by the edge computing equipment.

In an embodiment of the present invention, the edge computing device is specifically configured to perform data sensing on data of a local network in the data to be sensed by using a data sensing model deployed by the edge computing device; the cloud server is further configured to send first sample data in the data to be perceived to the cloud server, where the first sample data is: the edge computing device receives data of a local network in the data sent by the data acquisition device;

the cloud server is specifically configured to perform data perception on data of a global network in the data to be perceived by using a data perception model deployed by the cloud server; the edge computing device is further configured to receive first sample data sent by the edge computing device, determine a first training sample and a first test sample in the first sample data, train a data perception model in the cloud server according to the first training sample, and test the trained data perception model according to the first test sample; sending the tested model parameters of the data perception model to the edge computing equipment;

the edge computing device is further configured to receive the model parameters sent by the cloud server; determining second sample data in the data to be sensed; training a data perception model configured by the model parameters according to a second training sample in the second sample data; testing the trained data perception model configured by the model parameters according to a second test sample in the second sample data; updating the data perception model deployed by the edge computing device by using the tested data perception model configured by the model parameters, wherein the second sample data is: and the data of the local network in the data which is received by the edge computing device and sent by the data acquisition device is different from the first sample data.

In an embodiment of the present invention, the data acquisition device is specifically configured to determine whether the data to be perceived is structured data, if so, pre-process the data to be perceived, and send the pre-processed data to be perceived to the edge computing device, otherwise, directly send the data to be perceived to the edge computing device.

In a second aspect, an embodiment of the present invention provides a data-aware method applied to an edge computing device, the method including:

receiving data sent by data acquisition equipment as data to be sensed;

under the condition that the data to be perceived is determined to be data of a local network according to the acquisition cycle of the data to be perceived, performing data perception on the data to be perceived to obtain a first perception result, and sending the first perception result to the data acquisition equipment, wherein the local network is as follows: a sub-network to which the data acquisition device belongs in a global network, the global network being: a network consisting of data acquisition equipment, edge computing equipment and a cloud server;

under the condition that the data to be perceived is determined to be data of a global network according to the acquisition cycle of the data to be perceived, sending the data to be perceived to a cloud server, and enabling the cloud server to perform data perception on the data to be perceived to obtain a second perception result;

and receiving the second sensing result sent by the cloud server, and sending the second sensing result to the data acquisition equipment.

In an embodiment of the invention, whether the acquisition cycle of the data to be perceived is smaller than a preset cycle is judged, if so, the data to be perceived is determined to be data of a local network, otherwise, the data to be perceived is determined to be data of a global network.

In one embodiment of the invention, the method further comprises: determining the perception accuracy of a first perception result of data to be verified, and if the perception accuracy is lower than a preset accuracy, sending the data to be verified to the cloud server, so that the cloud server performs data perception on the data to be verified to obtain a third perception result, wherein the data to be verified is as follows: the data acquisition equipment acquires data to be sensed of the local network within a preset time length;

and receiving a third perception result sent by the cloud server, and sending the third perception result to the data acquisition equipment.

In an embodiment of the present invention, in a case that the edge computing device performs data awareness on data of a local network in the data to be perceived by using a data awareness model deployed by the edge computing device, the method further includes:

sending first sample data in data to be perceived to the cloud server, so that the cloud server trains and tests a data perception model deployed in the cloud server according to the first sample data, wherein the first sample data is as follows: the edge computing device receives data of a local network in the data sent by the data acquisition device;

receiving model parameters of the trained and tested data perception model sent by the cloud server;

determining second sample data in the data to be perceived, and training the data perception model configured by the model parameters according to a second training sample in the second sample data, wherein the second sample data is as follows: the data of the local network in the data which is different from the first sample data and is received by the edge computing equipment and sent by the data acquisition equipment;

testing the trained data perception model configured by the model parameters according to a second test sample in the second sample data;

updating the data-aware model deployed by the edge computing device using the tested data-aware model configured with the model parameters.

In a third aspect, an embodiment of the present invention provides a data sensing method, which is applied to a cloud server, and the method includes:

receiving global data sent by an edge computing device, wherein the global data is: determining data which belongs to a global network in the data to be sensed according to the period of the data to be sensed, wherein the data to be sensed is data acquired by data acquisition equipment;

carrying out data perception on the global data to obtain a second perception result;

sending the second perception result to the edge computing device.

In one embodiment of the invention, the method further comprises:

receiving data to be verified sent by the edge computing device, and performing data perception on the data to be verified to obtain a third perception result, wherein the data to be verified is as follows: the data acquisition equipment acquires data to be sensed of the local network within a preset time length;

sending the third perception result to the edge computing device.

In an embodiment of the present invention, in a case that the cloud server performs data awareness on data of a global network in the data to be perceived by using a data awareness model deployed by the cloud server, the method further includes:

receiving first sample data sent by the edge computing device, and determining a first training sample and a first test sample in the first sample data, where the first sample data is: the edge computing device receives data of a local network in the data sent by the data acquisition device;

training a data perception model in the cloud server according to the first training sample;

testing the trained data perception model according to the first test sample;

and sending the tested model parameters of the data perception model to the edge computing equipment.

In a fourth aspect, an embodiment of the present invention provides a data sensing apparatus, which is applied to an edge computing device, and the apparatus includes:

the first data receiving module is used for receiving data sent by the data acquisition equipment as data to be sensed;

the first data sensing module is configured to perform data sensing on the data to be sensed to obtain a first sensing result and send the first sensing result to the data acquisition device when it is determined that the data to be sensed is data of a local network according to the acquisition cycle of the data to be sensed, where the local network is: a sub-network to which the data acquisition device belongs in a global network, the global network being: a network consisting of data acquisition equipment, edge computing equipment and a cloud server;

the first data sending module is used for sending the data to be perceived to a cloud server under the condition that the data to be perceived is determined to be data of a global network according to the acquisition cycle of the data to be perceived, so that the cloud server conducts data perception on the data to be perceived to obtain a second perception result;

and the first result sending module is used for receiving the second perception result sent by the cloud server and sending the second perception result to the data acquisition equipment.

In one embodiment of the present invention, the apparatus further comprises:

the second data sending module is configured to determine a sensing accuracy of a first sensing result of data to be verified, and if the sensing accuracy is lower than a preset accuracy, send the data to be verified to the cloud server, so that the cloud server performs data sensing on the data to be verified to obtain a third sensing result, where the data to be verified is: the data acquisition equipment acquires data to be sensed of the local network within a preset time length;

and the result receiving module is used for receiving a third sensing result sent by the cloud server and sending the third sensing result to the data acquisition equipment.

In an embodiment of the present invention, in a case that the edge computing device performs data awareness on data of a local network in the data to be perceived by using a data awareness model deployed by the edge computing device, the apparatus further includes:

the sample data sending module is used for sending first sample data in data to be perceived to the cloud server, so that the cloud server trains and tests a data perception model deployed in the cloud server according to the first sample data, wherein the first sample data is as follows: the edge computing device receives data of a local network in the data sent by the data acquisition device;

the parameter receiving module is used for receiving model parameters of the trained and tested data perception model sent by the cloud server;

the first model training module is used for determining second sample data in the data to be perceived, and training the data perception model configured by the model parameters according to the second training sample in the second sample data, wherein the second sample data is: the data of the local network in the data which is different from the first sample data and is received by the edge computing equipment and sent by the data acquisition equipment;

the second model testing module is used for testing the trained data perception model configured by the model parameters according to a second test sample in the second sample data;

and the model updating module is used for updating the data perception model deployed by the edge computing equipment by using the tested data perception model configured by the model parameters.

In a fifth aspect, an embodiment of the present invention provides a data sensing apparatus, which is applied to a cloud server, and the apparatus includes:

a second data receiving module, configured to receive global data sent by the edge computing device, where the global data is: determining data which belongs to a global network in the data to be sensed according to the period of the data to be sensed, wherein the data to be sensed is data acquired by data acquisition equipment;

the second data perception module is used for carrying out data perception on the global data to obtain a second perception result;

and the second result sending module is used for sending the second sensing result to the edge computing equipment.

In one embodiment of the present invention, the apparatus further comprises:

the third data perception module is configured to receive data to be verified sent by the edge computing device, perform data perception on the data to be verified to obtain a third perception result, where the data to be verified is: the data acquisition equipment acquires data to be sensed of the local network within a preset time length;

and the third result sending module is used for sending the third perception result to the edge computing equipment.

In an embodiment of the present invention, in a case that the cloud server performs data awareness on data of a global network in the data to be perceived by using a data awareness model deployed by the cloud server, the apparatus further includes:

a sample data receiving module, configured to receive first sample data sent by the edge computing device, and determine a first training sample and a first test sample in the first sample data, where the first sample data is: the edge computing device receives data of a local network in the data sent by the data acquisition device;

the second model training module is used for training the data perception model in the cloud server according to the first training sample;

the second model testing module is used for testing the trained data perception model according to the first testing sample;

and the parameter sending module is used for sending the tested model parameters of the data perception model to the edge computing equipment.

In a sixth aspect, an embodiment of the present invention provides an electronic device, which is an edge computing device and includes a processor, a communication interface, a memory, and a communication bus, where the processor and the communication interface complete communication between the memory and the processor through the communication bus;

a memory for storing a computer program;

a processor for implementing the method steps of the second aspect when executing the program stored in the memory.

A seventh aspect of the present invention provides an electronic device, which is a cloud server and includes a processor, a communication interface, a memory, and a communication bus, where the processor and the communication interface complete communication between the memory and the processor through the communication bus;

a memory for storing a computer program;

a processor for implementing the method steps of the third aspect when executing the program stored in the memory.

In an eighth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the method steps of any one of the second aspects.

In a ninth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the method steps in any one of the third aspects.

In a tenth aspect, embodiments of the present invention also provide a computer program product including instructions, which when executed on a computer, cause the computer to perform the method steps of any one of the second aspects.

In an eleventh aspect, embodiments of the present invention also provide a computer program product including instructions, which when run on a computer, cause the computer to perform the method steps of any one of the above third aspects.

The embodiment of the invention has the following beneficial effects:

after receiving the data to be sensed sent by the data acquisition equipment, the edge computing equipment determines the data acquisition period of the data to be sensed, and determines whether the data to be sensed is data of a local network or data of a global network according to the data acquisition period of the data to be sensed. And if the data to be sensed is data of the local network, performing data sensing by the edge computing equipment which can control the local network and is close to the data acquisition equipment. If the data to be sensed is data of a global network, the data of the global network is often complex, the edge computing device sends the data to be sensed to the cloud server, and the cloud server which can control the global network and has strong computing capability senses the data. Therefore, the system provided by the embodiment of the invention can respectively sense the data of the local network and the data of the global network by using different devices, respectively meet different data sensing requirements of the data of the local network and the data of the global network, and improve the processing efficiency when the data to be sensed is processed.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic structural diagram of a data sensing system according to an embodiment of the present invention;

fig. 2 is a signaling flow diagram of a data-aware system according to an embodiment of the present invention;

fig. 3 is a schematic flowchart of a data sensing method applied to an edge computing device according to an embodiment of the present invention;

fig. 4 is a schematic flowchart of a method for verifying a sensing result applied to an edge computing device according to an embodiment of the present invention;

FIG. 5 is a schematic flowchart of a data-aware model training method applied to an edge computing device according to an embodiment of the present invention;

fig. 6 is a schematic flowchart of a data sensing method applied to a cloud server according to an embodiment of the present invention;

fig. 7 is a schematic flowchart of a method for verifying a sensing result applied to a cloud server according to an embodiment of the present invention;

fig. 8 is a schematic flowchart of a data perception model training method applied to a cloud server according to an embodiment of the present invention;

FIG. 9 is a schematic structural diagram of a data sensing apparatus applied to an edge computing device according to an embodiment of the present invention;

fig. 10 is a schematic structural diagram of a sensing result verification apparatus applied to an edge computing device according to an embodiment of the present invention;

FIG. 11 is a schematic structural diagram of a data perception model training apparatus applied to an edge computing device according to an embodiment of the present invention;

fig. 12 is a schematic structural diagram of a data sensing apparatus applied to a cloud server according to an embodiment of the present invention;

fig. 13 is a schematic structural diagram of a sensing result verification apparatus applied to a cloud server according to an embodiment of the present invention;

fig. 14 is a schematic structural diagram of a data perception model training apparatus applied to a cloud server according to an embodiment of the present invention;

fig. 15 is a schematic structural diagram of an electronic device according to an embodiment of the present invention;

fig. 16 is a schematic structural diagram of another electronic device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Because the prior art has the problem of low data sensing efficiency, embodiments of the present invention provide a data sensing system, method, and apparatus to solve the problem.

In one embodiment of the present invention, a data awareness system is provided, the system comprising: the system comprises data acquisition equipment, edge computing equipment and a cloud server, wherein the data acquisition equipment, the edge computing equipment and the cloud server are positioned in a global network; wherein the content of the first and second substances,

the data acquisition device is configured to acquire data as data to be sensed, and send the data to be sensed to the edge computing device in the local network to which the data acquisition device belongs in the global network.

The edge computing device is configured to receive the data to be sensed, which is sent by the data acquisition device; under the condition that the data to be perceived is determined to be data of a local network according to the acquisition cycle of the data to be perceived, carrying out data perception on the data to be perceived to obtain a first perception result; sending the first sensing result to the data acquisition equipment; and sending the data to be perceived to the cloud server under the condition that the data to be perceived is determined to be data of a global network according to the acquisition period of the data to be perceived.

The cloud server is used for receiving the data to be perceived sent by the edge computing device, performing data perception on the data to be perceived, and obtaining a second perception result; and sending the second perception result to the edge computing device.

The edge computing device is further configured to receive a second sensing result sent by the cloud server, and send the second sensing result to the data acquisition device.

The data acquisition device is further configured to receive the first sensing result or the second sensing result sent by the edge computing device.

As can be seen from the above, if the data to be sensed is data of a local network, the data is sensed by the edge computing device which can control the local network and is close to the data acquisition device. If the data to be sensed is data of a global network, the data of the global network is often complex, the edge computing device sends the data to be sensed to the cloud server, and the cloud server which can control the global network and has strong computing capability senses the data. Therefore, the system provided by the embodiment of the invention can respectively sense the data of the local network and the data of the global network by using different devices, respectively meet different data sensing requirements of the data of the local network and the data of the global network, and improve the processing efficiency when the data to be sensed is processed.

A data sensing system, a method and an apparatus provided in the embodiments of the present invention are described below with specific embodiments.

Referring to fig. 1, there is provided a schematic structural diagram of a data perception system, the system including: data acquisition device 101, edge computing device 102, and cloud server 103.

The data acquisition device 101, the edge computing device 102, and the cloud server 103 are located in a global network.

The global network includes the data acquisition device 101, the edge computing device 102, and the cloud server 103, and since the network formed by the data acquisition device 101, the edge computing device 102, and the cloud server 103 includes many devices and has a large network, the network may be referred to as a global network.

Corresponding to the global network, the local network is a part of the global network, the local network includes the edge computing device 102 and the data acquisition device 101, and the coverage area of the edge computing device 102 is limited, so that compared with the global network, the number of the data acquisition devices 101 included in the local network is smaller, and the network formed by the edge computing device 102 and the data acquisition device 101 connected to the edge computing device 102 is smaller, so that the local network may be referred to as the local network.

Specifically, the edge computing device may be deployed in a single base station, may be shared by multiple base stations, or may be deployed independently from a base station and connected to the base station through a network interface.

The system described above is described below with reference to fig. 2, and with reference to fig. 2, a signaling flow diagram of a data-aware system is provided.

S201: the data acquisition device 101 acquires data as data to be sensed.

S202: the data collecting device 101 sends the data to be sensed to the edge computing device 102 in the local network to which the data collecting device 101 belongs in the global network.

The data acquisition device 101 may be a mobile phone, a tablet computer, a sensing device, a professional measurement device, or the like.

The data to be sensed may be data obtained by direct measurement by the data acquisition device 101, such as temperature data, humidity data, image data, battery attribute data of the data acquisition device 101, and the like, which are directly acquired by the data acquisition device 101. The data to be sensed may also be data obtained by calculating the acquired data by the data acquisition device 101, such as vehicle speed data obtained by calculating according to the vehicle driving distance and the driving time.

The data to be sensed can be structured data such as temperature data, vehicle geographical position data and the like, and the data to be sensed can also be unstructured data such as picture data, video data and the like.

The local network may include an edge computing device 102 and a plurality of data collecting devices 101, and the global network may include a cloud server 103 and a plurality of local networks.

S203: and under the condition that the data to be perceived is determined to be data of a local network according to the acquisition period of the data to be perceived, carrying out data perception on the data to be perceived to obtain a first perception result.

Specifically, data of the local network may be subjected to data perception in the local network, and the local network includes the edge computing device 102 and the data acquisition device 101, so that the edge computing device 102 may perform data perception on the data to be perceived.

In an embodiment of the present invention, the edge computing device 102 may first perform feature extraction on the received data to be perceived and the stored historical data to be perceived that the data acquisition device 101 sends before sending the data to be perceived, so as to reduce the dimensionality of the data to be perceived and obtain the main information in the data to be perceived.

Specifically, under the condition that the data to be perceived is structured data, the correlation among different features in the data to be perceived can be calculated, and the features with higher correlation are combined, so that the dimensionality of the data to be perceived is reduced. The feature extraction of the data to be perceived, which are structured data, can be performed by methods such as a principal component analysis method and a linear discriminant analysis method.

In addition, under the condition that the data to be perceived is unstructured, the data to be perceived can be scanned to obtain information of the data to be perceived in different dimensions, and specifically, the data to be perceived, which is unstructured data, can be subjected to feature extraction through a convolutional neural network.

After the feature extraction is performed on the data to be perceived, the data to be perceived after the feature extraction can be subjected to data perception, so that a first perception result is obtained.

The data to be perceived after feature extraction may be subjected to data perception through the data perception model deployed in the edge computing device 102. The data perception model can be a model obtained through supervised learning, unsupervised learning or reinforcement learning training.

Specifically, the first sensing result may be predicted characteristic data, such as vehicle peripheral information, or an operation instruction for guiding the data acquisition device 101, such as an alarm instruction.

In addition, the edge computing device 102 may store the received data to be sensed and a first sensing result obtained through data sensing.

S204: the edge computing device 102 sends the first sensing result to the data collecting device 101.

S205: when it is determined that the data to be perceived is data of a global network according to the acquisition cycle of the data to be perceived, the edge computing device 102 sends the data to be perceived to the cloud server 103.

Specifically, data of the global network can be subjected to data perception in the global network, which is often complex, the global network includes the cloud server 103 and the local network, and the cloud server has strong computing capability, so that the cloud server 103 can be used for performing data perception on the complex data to be perceived.

In an embodiment of the present invention, the edge computing device 102 may determine, through the following step a, whether the data to be perceived is data of a local network or data of a global network.

Step A: and judging whether the acquisition period of the data to be perceived is smaller than a preset period, if so, determining that the data to be perceived is data of a local network, and otherwise, determining that the data to be perceived is data of a global network.

Specifically, the preset period may be 1s, 3s, and the like.

Specifically, if the acquisition period of the data to be sensed is short, it is indicated that the data to be sensed is frequently updated data, and the data to be updated needs to be sensed more quickly. Therefore, the data to be sensed is used as the data of the local network, and the edge computing device 102 which is closer to the data acquisition device 101 performs data sensing, so that the data transmission time can be shortened, and the data sensing time is shortened.

On the contrary, if the acquisition period of the data to be sensed is long, it indicates that the data to be sensed is not updated frequently. Therefore, the data to be sensed is used as the data of the global network, and the cloud server 103 with strong computing power senses the data.

S206: the cloud server 103 performs data sensing on the data to be sensed to obtain a second sensing result.

The data sensing manner of the cloud server 103 is the same as that of the edge computing device 102, and is not described herein again.

S207: the cloud server 103 sends the second sensing result to the edge computing device 102.

In an embodiment of the present invention, the cloud server 103 may perform data fusion on the data representing the features in the received data to be sensed.

And performing feature extraction on the data to be perceived after feature fusion, reducing the dimensionality of the data to be perceived, and acquiring main information in the data to be perceived.

After the feature extraction is performed on the data to be perceived, the data to be perceived after the feature extraction can be subjected to data perception, and a second perception result is obtained.

The data to be perceived after the feature extraction can be subjected to data perception through the data perception model deployed in the cloud server 103. The data perception model can be a model obtained through supervised learning, unsupervised learning or reinforcement learning training.

Specifically, the second sensing result may be predicted characteristic data, such as vehicle peripheral information, or an operation instruction for guiding the data acquisition device 101, such as an alarm instruction.

In addition, the cloud server 103 may store the received data to be sensed and a second sensing result obtained through data sensing.

S208: the edge computing device 102 sends the second sensing result to the data collecting device 101.

As can be seen from the above, the data sensing system includes the data acquisition device 101, the edge computing device 102, and the cloud server 103 and the edge computing device 102 respectively perform data sensing on different data to be sensed. Corresponding to cloud computing, a fog computing process is formed in which the edge computing device 102 cooperates with the cloud server 103.

In an embodiment of the present invention, the edge computing device 102 is further configured to determine a sensing accuracy of a first sensing result of data to be verified, and send the data to be verified to the cloud server 103 if the sensing accuracy is lower than a preset accuracy.

Wherein, the data to be verified is: the data acquisition device 101 acquires data to be sensed of the local network within a preset time period.

The preset time period may be 5 minutes, 10 minutes, or the like.

Specifically, the first sensing result is: under the condition of a prediction result obtained by predicting the data to be perceived, which is acquired by the data acquisition device 101, the perception accuracy may be determined according to the data to be perceived, which is acquired by the data acquisition device 101.

For example, in the first sensing result: under the condition of a prediction result obtained by predicting the data to be perceived, which is acquired by the data acquisition device 101 at the next acquisition time, if 6 first perception results obtained by calculation in the preset time period are the same as the data to be perceived, which is acquired by the data acquisition device 101 at the next acquisition time, the perception accuracy is 60%.

In addition, the first sensing result is: under the condition of the result of early warning on the data acquisition device 101, the sensing accuracy can be determined according to the result whether the early warning fed back by the user is correct or not.

For example, 10 first sensing results are obtained by calculation in a preset time, the 10 first sensing results are sent to a data acquisition device, after the data acquisition device is early-warned, results whether early warning fed back by 10 users is correct or not are received, and if the first sensing results include 7 early-warned correct first sensing results, the sensing accuracy is 70%.

Moreover, the above-mentioned perceptual accuracy may also be calculated in other ways.

The cloud server 103 is configured to receive data to be verified sent by the edge computing device 102, perform data sensing on the data to be verified to obtain a third sensing result, and send the third sensing result to the edge computing device 102.

The edge computing device 102 is configured to receive a third sensing result sent by the cloud server 103, and send the third sensing result to the data acquisition device 101.

The data collecting device 101 is configured to receive the third sensing result sent by the edge computing device 102.

As can be seen from the above, under the condition that the accuracy of the first sensing result obtained by the edge computing device performing data sensing on the data to be verified is low, the data to be verified is sent to the cloud server, and the cloud server performs data sensing on the data to be verified again, so that a third sensing result of the data to be verified is obtained, and the third sensing result is sent to the data acquisition device through the edge computing server. The data acquisition equipment can receive the data sensing result obtained by carrying out data sensing on the data to be verified again, so that the accuracy of the data sensing result is improved.

In an embodiment of the present invention, the edge computing device 102 is specifically configured to perform data sensing on data of a local network in the data to be sensed by using a data sensing model deployed by the edge computing device 102.

Because the data acquisition environment in which the data acquisition device 101 is located may change greatly with time, for example, when the data acquisition device 101 is installed in a vehicle, along with the driving of the vehicle, the data acquisition environment in which the data acquisition device 101 is located may change greatly with time, so the data to be sensed acquired by the data acquisition device 101 also changes greatly, and if the same data sensing model is used to perform data sensing on the data of the local network in the data to be sensed all the time, the accuracy of the first sensing result is low. Therefore, the data perception model can be trained every time a preset time period passes, so that the data perception model is matched with a data acquisition environment.

For example, the preset time period may be a time period having a duration of 5 minutes, a time period having a duration of 10 minutes, or the like.

In addition, under the condition that the sensing accuracy of the first sensing result of the data to be verified is lower than the preset accuracy, it is indicated that the accuracy of the data sensing model is low, and if the data sensing model is continuously used for carrying out data sensing on the data to be sensed, the accuracy of the obtained first sensing result is still low, so that the data sensing model can be retrained.

Specifically, the process of training the data perception model is as follows:

the edge computing device 102 is further configured to send the first sample data in the data to be perceived to the cloud server 103.

The first sample data is: and the edge computing equipment receives the data of the local network in the data sent by the data acquisition equipment.

Specifically, the first sample data includes a first training sample and a first test sample.

The first sample data may be: and determining data of the local network in the data to be sensed, which is acquired by the data acquisition device 101, within a first preset time before the first sample determination time of the first sample data.

For example, the first preset time period may be 5 minutes or 10 minutes.

In addition, the first sample data may be: and sampling the data of the local network and the data to be sensed, which are acquired by the data acquisition equipment 101, in a first preset time before the first sample determination time for determining the first sample data.

Where the data may be sampled by random sampling or other means.

The cloud server 103 is specifically configured to perform data awareness on data of the global network in the data to be perceived by using the data awareness model deployed by the cloud server.

Specifically, the data awareness model deployed in the cloud server 103 is the same as the data awareness model deployed in the edge computing device 102.

In the case where the global network includes a plurality of local networks, each edge computing device 102 includes a data awareness model, and thus the cloud server 103 includes a plurality of data awareness models that are respectively the same as the data awareness models included in each edge computing device 102.

The cloud server 103 is further configured to train the data perception model through the following steps B to E.

And B: and receiving the first sample data sent by the edge computing device 102, and determining a first training sample and a first test sample in the first sample data.

Specifically, the cloud server 103 may sample the first sample data by random sampling or other methods after receiving the first sample data.

And C: and training the data perception model in the cloud server 103 according to the first training sample.

Specifically, when the global network includes a plurality of local networks, the cloud server 103 may receive first sample data sent by each edge computing device 102, and train the data perception model together with first training samples in the first sample data sent by each edge computing device 102. The trained data perception model can meet the data perception requirements of the data to be perceived, which are collected by each data collection device 101 in the global network.

The data perception model can be trained through methods such as supervised learning, unsupervised learning and reinforcement learning.

Step D: and testing the trained data perception model according to the first test sample.

Specifically, in a case where the global network includes a plurality of local networks, the cloud server 103 may collectively test the data awareness model with a first test sample in the first sample data sent by each edge computing device 102.

And in the case that the test accuracy of the test result of the first test sample is higher than the preset test accuracy, the data perception model is considered to pass the test.

Step E: sending the tested model parameters of the data perception model to the edge computing device 102.

In the case that the cloud server 103 includes a plurality of data perception models, and each data perception model is the same as the data perception model included in the edge computing device, the model parameters of the data perception model with the highest accuracy of the data perception result output after training and testing in the data perception models with the same function may be determined, and the determined model parameters are sent to the edge computing device corresponding to each data perception model with the same function.

Specifically, the data perception model with the highest accuracy can be determined and obtained by an Adaboost method.

The edge computing device 102 is further configured to receive the model parameter sent by the cloud server 103. And determining second sample data in the data to be sensed. And training the data perception model configured by the model parameters according to a second training sample in the second sample data. And testing the trained data perception model configured by the model parameters according to a second test sample in the second sample data. And updating the data perception model deployed by the edge computing equipment by using the tested data perception model configured by the model parameters.

Wherein the second sample data is: and the data of the local network in the data which is received by the edge computing device and sent by the data acquisition device is different from the first sample data.

Specifically, the second sample data is data of a local network received by the edge computing device 102, so that the second sample data is only data in the local network to which the edge computing device belongs, and the data sensing model is trained and tested by using the second sample data, so that the trained and tested data sensing model can meet the data sensing requirement of the data to be sensed, which is acquired by the data acquisition device in the local network, and is matched with the data acquisition environment of the local network to which the edge computing device belongs.

Therefore, the data perception model is retrained and tested, and the data perception model obtained after retraining and testing is matched with the data to be perceived newly acquired by the data acquisition equipment, so that the accuracy of the data perception result output by the data perception model is improved.

In another embodiment of the present invention, the data collecting device 101 is specifically configured to determine whether the data to be perceived is structured data, if so, pre-process the data to be perceived, and send the pre-processed data to be perceived to the edge computing device 102, otherwise, directly send the data to be perceived to the edge computing device 102.

In particular, since the complexity of the structured data is often low, the structured data may be preprocessed using a data acquisition device with weak computing power.

In an embodiment of the present invention, the data acquisition device 101 may perform preprocessing on the structured data through one or more of the following steps F-I.

Step F: and removing abnormal data in the data to be sensed.

Specifically, outlier data having a large difference from data values of other data to be sensed in the data to be sensed may be calculated, the outlier data may be used as abnormal data, and the abnormal data may be removed.

Specifically, outlier data can be identified by descriptive statistics, boxplot, three sigma, etc.

The three-sigma method is that for data to be perceived which obeys normal distribution, the data to be perceived, of which the absolute value of the difference value with the average value is more than three times of the standard deviation, is regarded as outlier data.

Step G: and correcting the incomplete data in the data to be sensed.

The incomplete data are: the data which are not successfully collected in the data to be sensed.

Specifically, the correction value for correcting the incomplete data may be calculated according to first data to be sensed acquired within a second preset time period before the acquisition time of the incomplete data and/or second data to be sensed acquired within a third preset time period after the acquisition time of the incomplete data.

The correction value may be an average value, a maximum value, a minimum value, or the like of the first data to be sensed and the second data to be sensed. The data to be sensed with the highest occurrence probability in the first data to be sensed and the second data to be sensed may also be used as the correction value.

In addition, the incomplete data in the data to be sensed can be corrected by methods such as cluster filling and multiple interpolation. Since the cluster filling method and the multiple interpolation method are prior art, they are not described herein again.

Step H: and identifying data with similar characteristics in the data to be sensed, and reserving one of the data to be sensed with similar characteristics.

Specifically, a correlation coefficient between any two pieces of feature data may be calculated, and one of the two pieces of feature data having a correlation coefficient larger than a preset coefficient may be retained.

The correlation coefficient may be a pearson coefficient, and then the data to be perceived with similar characteristics may be identified by a correlation filtering method based on the pearson coefficient, and one of the data to be perceived with similar characteristics is retained.

Step I: and normalizing the data to be sensed to the same dimension.

Specifically, the data to be sensed can be normalized to the same dimension by a zero-mean normalization method, a maximum-minimum normalization method and the like.

In another embodiment of the present invention, the edge computing device 102 with higher computing power than the data acquisition device may be used to pre-process the unstructured data with higher complexity, and perform data sensing on the pre-processed data to be sensed.

Therefore, abnormal data in the data to be perceived can be removed or incomplete data in the data to be perceived can be corrected by preprocessing the data to be perceived, and data with inaccurate numerical values in the data to be perceived is reduced, so that the accuracy of a perception result obtained by perceiving the data to be perceived is improved. The data with similar characteristics in the data to be perceived are removed, so that the data volume of the data to be perceived is reduced, or the data to be perceived is normalized to the same dimension, so that the numerical value of the data to be perceived is relatively neat, and the perception efficiency of data perception of the data to be perceived is improved.

Corresponding to the data perception system, the embodiment of the invention also provides a data perception method applied to the edge computing equipment.

Referring to fig. 3, there is provided a flow chart of a data-aware method applied to an edge computing device, the method including:

s301: and receiving data sent by the data acquisition equipment as data to be sensed.

S302: and under the condition that the data to be perceived is determined to be data of a local network according to the acquisition cycle of the data to be perceived, carrying out data perception on the data to be perceived to obtain a first perception result, and sending the first perception result to the data acquisition equipment.

Wherein, the local network is: a sub-network to which the data acquisition device belongs in a global network, wherein the global network is: the system comprises a network consisting of data acquisition equipment, edge computing equipment and a cloud server.

S303: and under the condition that the data to be perceived is determined to be data of a global network according to the acquisition period of the data to be perceived, sending the data to be perceived to a cloud server, so that the cloud server performs data perception on the data to be perceived to obtain a second perception result.

In an embodiment of the present invention, it may be determined whether an acquisition period of the data to be sensed is less than a preset period, and if so, the data to be sensed is determined to be data of a local network, otherwise, the data to be sensed is determined to be data of a global network.

S304: and receiving the second sensing result sent by the cloud server, and sending the second sensing result to the data acquisition equipment.

The data sensing method applied to the edge computing device is the same as the method for the edge computing device to sense data in the data sensing system, and is not described herein again.

As can be seen from the above, if the data to be sensed is data of a local network, the data is sensed by the edge computing device which can control the local network and is close to the data acquisition device. And if the data to be perceived is data of the global network, the edge computing equipment sends the data to be perceived to the cloud server. Therefore, the edge computing equipment only carries out data perception on the data of the local network with short feedback time, meets different data perception requirements of the data of the local network, and improves the processing efficiency when the data to be perceived is processed.

In an embodiment of the present invention, referring to fig. 4, a flowchart of a method for verifying a sensing result applied to an edge computing device is provided, where the method includes:

s401: and determining the sensing accuracy of a first sensing result of the data to be verified, and if the sensing accuracy is lower than a preset accuracy, sending the data to be verified to the cloud server, so that the cloud server performs data sensing on the data to be verified, and a third sensing result is obtained.

The data to be verified is as follows: the data acquisition equipment acquires data to be sensed of the local network within a preset time length.

S402: and receiving a third sensing result sent by the cloud server, and sending the third sensing result to the data acquisition equipment.

The method for verifying the sensing result applied to the edge computing device is the same as the method for verifying the sensing result of the edge computing device in the data sensing system, and is not repeated herein.

In one embodiment of the present invention, referring to fig. 5, a flowchart of a data-aware model training method applied to an edge computing device is provided. In a case that the edge computing device performs data awareness on data of a local network in the data to be perceived by using a data awareness model deployed by the edge computing device, the method includes:

s501: and sending first sample data in the data to be perceived to the cloud server, so that the cloud server trains and tests a data perception model deployed in the cloud server according to the first sample data.

Wherein the first sample data is: the edge computing device receives data of a local network in the data sent by the data acquisition device;

s502: and receiving model parameters of the trained and tested data perception model sent by the cloud server.

S503: determining second sample data in the data to be perceived, and training the data perception model configured by the model parameters according to a second training sample in the second sample data, wherein the second sample data is as follows: and the data of the local network in the data which is received by the edge computing device and sent by the data acquisition device is different from the first sample data.

S504: and testing the trained data perception model configured by the model parameters according to a second test sample in the second sample data.

S505: and updating the data perception model deployed by the edge computing equipment by using the tested data perception model configured by the model parameters.

The data perception model training method applied to the edge computing device is the same as the data perception model training method performed by the edge computing device in the data perception system, and is not repeated herein.

Corresponding to the data perception method applied to the edge computing device, the embodiment of the invention also provides a data perception method applied to the cloud server.

Referring to fig. 6, an embodiment of the present invention provides a schematic flow diagram of a data sensing method applied to a cloud server, where the method includes:

s601: receiving global data sent by edge computing equipment, wherein the global data is as follows: and determining data belonging to the global network in the data to be sensed according to the period of the data to be sensed, wherein the data to be sensed is data acquired by data acquisition equipment.

S602: and carrying out data perception on the global data to obtain a second perception result.

S603: and sending the second perception result to the edge computing device.

The data sensing method applied to the cloud server is the same as the method for the cloud server to sense the data in the data sensing system, and is not repeated herein.

As can be seen from the above, if the data to be perceived is data of a global network, the data of the global network is often complex, and the cloud server capable of controlling the global network and having a strong computing power is used to perceive the data of the global network. The data perception requirement of the data of the global network is met, and the processing efficiency of the data to be perceived is improved.

Referring to fig. 7, an embodiment of the present invention provides a schematic flow diagram of a method for verifying a sensing result applied to a cloud server, where the method includes:

s701: and receiving to-be-verified data sent by the edge computing equipment, and performing data perception on the to-be-verified data to obtain a third perception result.

S702: and sending the third perception result to the edge computing device.

The sensing result verification method applied to the cloud server is the same as the method for data sensing of the cloud server in the data sensing system, and is not repeated herein.

Referring to fig. 8, a schematic flow chart of a data awareness model training method applied to a cloud server is provided, where the cloud server performs data awareness on data of a global network in the data to be perceived by using a data awareness model deployed by the cloud server, the method further includes:

s801: and receiving first sample data sent by the edge computing equipment, and determining a first training sample and a first test sample in the first sample data.

S802: and training the data perception model in the cloud server according to the first training sample.

S803: and testing the trained data perception model according to the first test sample.

S804: and sending the tested model parameters of the data perception model to the edge computing equipment.

The data perception model training method applied to the cloud server is the same as the method for performing data perception model training on the cloud server in the data perception system, and is not repeated herein.

Corresponding to the data perception method applied to the edge computing device, the embodiment of the invention also provides a data perception device applied to the edge computing device.

Referring to fig. 9, there is provided a schematic structural diagram of a data perception apparatus applied to an edge computing device, where the apparatus includes:

a first data receiving module 901, configured to receive data sent by a data acquisition device, as data to be sensed;

a first data sensing module 902, configured to perform data sensing on the data to be sensed to obtain a first sensing result and send the first sensing result to the data acquisition device when it is determined that the data to be sensed is data of a local network according to the acquisition cycle of the data to be sensed, where the local network is: a sub-network to which the data acquisition device belongs in a global network, the global network being: a network consisting of data acquisition equipment, edge computing equipment and a cloud server;

a first data sending module 903, configured to send the data to be perceived to a cloud server under the condition that it is determined that the data to be perceived is data of a global network according to an acquisition cycle of the data to be perceived, so that the cloud server performs data perception on the data to be perceived to obtain a second perception result;

a first result sending module 904, configured to receive the second sensing result sent by the cloud server, and send the second sensing result to the data acquisition device.

Corresponding to the sensing result verification method applied to the edge computing device, the embodiment of the invention also provides a sensing result verification device applied to the edge computing device.

Referring to fig. 10, there is provided a schematic structural diagram of a sensing result verification apparatus applied to an edge computing device, where the apparatus includes:

a second data sending module 1001, configured to determine a sensing accuracy of a first sensing result of data to be verified, and if the sensing accuracy is lower than a preset accuracy, send the data to be verified to the cloud server, so that the cloud server performs data sensing on the data to be verified to obtain a third sensing result, where the data to be verified is: the data acquisition equipment acquires data to be sensed of the local network within a preset time length;

the result receiving module 1002 is configured to receive a third sensing result sent by the cloud server, and send the third sensing result to the data acquisition device.

Corresponding to the data perception model training method applied to the edge computing equipment, the embodiment of the invention also provides a data perception model training device applied to the edge computing equipment.

Referring to fig. 11, a schematic structural diagram of a data perception model training apparatus applied to an edge computing device is provided, where the edge computing device performs data perception on data of a local network in the data to be perceived by using a data perception model deployed by the edge computing device, the apparatus further includes:

a sample data sending module 1101, configured to send first sample data in data to be perceived to the cloud server, so that the cloud server trains and tests a data perception model deployed in the cloud server according to the first sample data, where the first sample data is: the edge computing device receives data of a local network in the data sent by the data acquisition device;

a parameter receiving module 1102, configured to receive model parameters of the trained and tested data perception model sent by the cloud server;

a first model training module 1103, configured to determine second sample data in data to be perceived, and train, according to the second training sample in the second sample data, the data perception model configured with the model parameters, where the second sample data is: the data of the local network in the data which is different from the first sample data and is received by the edge computing equipment and sent by the data acquisition equipment;

a second model testing module 1104, configured to test the trained data perception model configured with the model parameters according to a second test sample in the second sample data;

a model update module 1105 configured to update the data-aware model deployed by the edge computing device using the tested data-aware model configured with the model parameters.

Corresponding to the data perception method applied to the cloud server, the embodiment of the invention also provides a data perception device applied to the cloud server.

Referring to fig. 12, there is provided a schematic structural diagram of a data sensing apparatus applied to a cloud server, where the apparatus includes:

a second data receiving module 1201, configured to receive global data sent by an edge computing device, where the global data is: determining data which belongs to a global network in the data to be sensed according to the period of the data to be sensed, wherein the data to be sensed is data acquired by data acquisition equipment;

a second data sensing module 1202, configured to perform data sensing on the global data to obtain a second sensing result;

a second result sending module 1203, configured to send the second sensing result to the edge computing device.

Corresponding to the sensing result verification method applied to the cloud server, the embodiment of the invention also provides a sensing result verification device applied to the cloud server.

Referring to fig. 13, there is provided a schematic structural diagram of a sensing result verifying apparatus applied to a cloud server, where the apparatus includes:

a third data sensing module 1301, configured to receive data to be verified sent by the edge computing device, perform data sensing on the data to be verified to obtain a third sensing result, where the data to be verified is: the data acquisition equipment acquires data to be sensed of the local network within a preset time length;

a third result sending module 1302, configured to send the third sensing result to the edge computing device.

Corresponding to the data perception model training method applied to the cloud server, the embodiment of the invention also provides a data perception model training device applied to the cloud server.

Referring to fig. 14, a schematic structural diagram of a data awareness model training device applied to a cloud server is provided, where in a case that the cloud server performs data awareness on data of a global network in the data to be perceived by using a data awareness model deployed by the cloud server, the device further includes:

a sample data receiving module 1401, configured to receive first sample data sent by the edge computing device, and determine a first training sample and a first test sample in the first sample data, where the first sample data is: the edge computing device receives data of a local network in the data sent by the data acquisition device;

a second model training module 1402, configured to train a data perception model in the cloud server according to the first training sample;

a second model testing module 1403, configured to test the trained data sensing model according to the first test sample;

a parameter sending module 1404, configured to send the tested model parameters of the data perception model to the edge computing device.

An embodiment of the present invention further provides an electronic device, as an edge computing device, as shown in fig. 15, including a processor 1501, a communication interface 1502, a memory 1503, and a communication bus 1504, where the processor 1501, the communication interface 1502, and the memory 1503 complete communication with each other through the communication bus 1504,

a memory 1503 for storing a computer program;

the processor 1501, when executing the program stored in the memory 1503, implements any of the steps of the data sensing method applied to the edge computing device described above.

When the electronic equipment provided by the embodiment of the invention is used as edge computing equipment for data perception, if the data to be perceived is data of a local network, the edge computing equipment which can control the local network and is close to the data acquisition equipment carries out data perception. And if the data to be perceived is data of the global network, the edge computing equipment sends the data to be perceived to the cloud server. Therefore, the edge computing equipment only carries out data perception on the data of the local network with short feedback time, meets different data perception requirements of the data of the local network, and improves the processing efficiency when the data to be perceived is processed.

An embodiment of the present invention further provides another electronic device, which is a cloud server, as shown in fig. 16, and includes a processor 1601, a communication interface 1602, a memory 1603, and a communication bus 1604, where the processor 1601, the communication interface 1602, and the memory 1603 complete communication with each other via the communication bus 1604,

a memory 1603 for storing a computer program;

the processor 1601 is configured to implement any one of the steps of the data sensing method applied to the cloud server when executing the program stored in the memory 1603.

When the electronic device provided by the embodiment of the invention is used as a cloud server to sense data, if the data to be sensed is data of a global network, the data of the global network is often complex, and the cloud server which can control the global network and has strong computing power senses the data of the global network. The data perception requirement of the data of the global network is met, and the processing efficiency of the data to be perceived is improved.

The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the electronic equipment and other equipment.

The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.

The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component.

In another embodiment of the present invention, a computer-readable storage medium applied to an edge computing device is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any of the above data sensing methods applied to the edge computing device.

When the computer program stored in the computer-readable storage medium applied to the edge computing device provided by the embodiment of the invention is executed to perform data perception, if the data to be perceived is data of a local network, the edge computing device which can control the local network and is close to the data acquisition device performs data perception. And if the data to be perceived is data of the global network, the edge computing equipment sends the data to be perceived to the cloud server. Therefore, the edge computing equipment only carries out data perception on the data of the local network with short feedback time, meets different data perception requirements of the data of the local network, and improves the processing efficiency when the data to be perceived is processed.

In another embodiment provided by the present invention, a computer-readable storage medium applied to a cloud server is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements any of the above steps of the data awareness method applied to the cloud server.

When the computer program stored in the computer-readable storage medium applied to the cloud server provided by the embodiment of the invention is executed to perform data perception, if the data to be perceived is data of a global network, the data of the global network is often complex, and the cloud server capable of controlling the global network and having strong computing power performs data perception on the data of the global network. The data perception requirement of the data of the global network is met, and the processing efficiency of the data to be perceived is improved.

In another embodiment of the present invention, there is also provided a computer program product containing instructions for an edge computing device, which when run on a computer, causes the computer to perform any one of the above-mentioned data awareness methods for an edge computing device.

When the computer program product applied to the edge computing device provided by the embodiment of the invention is executed for data perception, if the data to be perceived is data of a local network, the edge computing device which can control the local network and is close to the data acquisition device carries out data perception. And if the data to be perceived is data of the global network, the edge computing equipment sends the data to be perceived to the cloud server. Therefore, the edge computing equipment only carries out data perception on the data of the local network with short feedback time, meets different data perception requirements of the data of the local network, and improves the processing efficiency when the data to be perceived is processed.

In another embodiment provided by the present invention, there is also provided a computer program product containing instructions, which is applied to a cloud server and when running on a computer, causes the computer to execute any one of the above-mentioned data awareness methods applied to the cloud server.

When the computer program product applied to the cloud server provided by the embodiment of the invention is executed for data perception, if the data to be perceived is data of a global network, the data of the global network is often complex, and the cloud server which can control the global network and has strong computing power is used for data perception of the data of the global network. The data perception requirement of the data of the global network is met, and the processing efficiency of the data to be perceived is improved.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may 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 loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the methods, apparatus, electronic devices, computer-readable storage media, and computer program products are substantially similar to the system embodiments, so that the descriptions are simplified, and reference may be made to some descriptions of the system embodiments for relevant points.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

1. A data awareness system, the system comprising: the system comprises data acquisition equipment, edge computing equipment and a cloud server, wherein the data acquisition equipment, the edge computing equipment and the cloud server are positioned in a global network; wherein the content of the first and second substances,

2. The system of claim 1,

the edge computing device is specifically configured to determine whether an acquisition period of the data to be perceived is smaller than a preset period, determine that the data to be perceived is data of a local network if the acquisition period of the data to be perceived is smaller than the preset period, and determine that the data to be perceived is data of a global network if the acquisition period of the data to be perceived is not smaller than the preset period.

3. The system of claim 1,

the edge computing device is further configured to determine a sensing accuracy of a first sensing result of the data to be verified, and if the sensing accuracy is lower than a preset accuracy, send the data to be verified to the cloud server; the data to be verified is: the data acquisition equipment acquires data to be sensed of the local network within a preset time length;

4. The system of claim 1,

the edge computing device is specifically configured to perform data perception on data of a local network in the data to be perceived by using a data perception model deployed by the edge computing device; the cloud server is further configured to send first sample data in the data to be perceived to the cloud server, where the first sample data is: the edge computing device receives data of a local network in the data sent by the data acquisition device;

5. The system according to any one of claims 1-4,

the data acquisition device is specifically configured to determine whether the data to be perceived is structured data, if so, preprocess the data to be perceived, and send the preprocessed data to be perceived to the edge computing device, otherwise, directly send the data to be perceived to the edge computing device.

6. A data-aware method, for application to an edge computing device, the method comprising:

receiving data sent by data acquisition equipment as data to be sensed;

7. The method of claim 6,

and judging whether the acquisition period of the data to be perceived is smaller than a preset period, if so, determining that the data to be perceived is data of a local network, otherwise, determining that the data to be perceived is data of a global network.

8. A data perception method is applied to a cloud server, and the method comprises the following steps:

sending the second perception result to the edge computing device.

9. A data-aware apparatus, for application to an edge computing device, the apparatus comprising:

10. A data perception device is applied to a cloud server, and the device comprises: