A method and device for optical network fault location based on attention mechanism

Abstract

The embodiment of the present invention discloses a method and device for locating optical network faults based on an attention mechanism, including steps of data collection, node and link data characterization, node and link data feature extraction, network-wide data feature extraction, and fault location decision. The method and device of the present invention can filter out the most important data from large-scale monitoring data, greatly improving the efficiency and accuracy of network fault location. In addition, the present invention can accurately locate the network equipment where the fault occurs, which is more accurate than the existing solutions, which can greatly reduce the overhead caused by service rerouting, and meet the needs of future 6G / F6G green network construction and efficient operation and maintenance.

Description

Technical Field

[0001] The present invention relates to the field of communication network technology, and in particular to an optical network fault locating method and device based on an attention mechanism. Background Art

[0002] Due to the large bandwidth and low latency data transmission characteristics, optical networks play an important role in carrying network services such as 4K video transmission. In the upcoming 6G and F6G era, as network services continue to increase, it is foreseeable that the traffic of optical networks will show an explosive growth trend. Fault management is an important part of network operation and maintenance. Faults caused by aging equipment or manual operation errors (such as a link failure) will cause a large number of business transmission interruptions, resulting in gigabytes of data loss, thereby reducing the quality of network services. Therefore, once a network failure occurs, network operators need to quickly and accurately locate the fault and take relevant measures to recover from the fault to reduce economic losses.

[0003] In optical networks, a common fault location method is to find the root fault location by mining alarm text information. In recent years, many scholars have used deep learning technology to analyze alarm text information, including graph neural networks, deep belief networks, etc., to infer the location of faults in the network. However, the formation of alarm text information is highly dependent on manual experience, and when new faults that are not recorded in the historical database appear in the network, the accuracy of the alarm text content collected by the controller needs to be confirmed. Therefore, we have reason to question the ability of this method to locate new faults in the network.

[0004] Currently, a large number of high-precision optical performance monitors (OPMs) and telemetry systems are deployed in optical networks to monitor physical layer status parameters, such as optical power. When a network fault occurs, the network controller uses high-speed, high-precision telemetry technology to obtain sufficient monitoring data from the network for fault analysis. Many studies locate network faults by analyzing the correlation between monitoring data. For example, some scholars use artificial neural networks to extract the relationship between monitoring data to infer the location of network faults. However, existing studies can usually only locate node faults, but cannot more accurately locate a certain device in the node, such as WSS. Or an internal connection, such as the internal connection between WSS. This will cause many rerouting measures to be added unnecessarily during fault recovery, resulting in additional fault recovery costs. On the one hand, due to the large scale of optical networks, the amount of monitoring data collected by the controller is huge. Among these data, only a small part of the data (such as data located near the faulty device) is helpful for fault location, and the remaining large part does not contribute much to fault location. Therefore, when locating a fault, we need to pay more attention to those important data to improve the efficiency and accuracy of fault location. How to filter out the most important data from the large-scale monitoring data has become an important issue. On the other hand, in optical networks, the failure of a network device often causes abnormalities in multiple monitoring data. Since the types of devices that implement optical signal transmission are diverse and the connections between them are intricate, the relationship between these monitoring data is often quite complex. Therefore, how to extract the relationship between monitoring data to locate network failures has become another important issue.

[0005] In summary, the existing fault location solutions have certain limitations and are difficult to effectively adapt to the future low-cost, high-intelligence network operation and maintenance needs. Therefore, in the future 6G / F6G era, a more optimal fault location strategy is urgently needed to improve network reliability and overall operational efficiency. Summary of the invention

[0006] The present invention aims to address the limitations of existing fault location solutions and to provide an optical network fault location method based on an attention mechanism.

[0007] In order to achieve the above object, the present invention provides the following technical solutions:

[0008] The present invention provides an optical network fault location method based on an attention mechanism, comprising the following steps:

[0009] S1. Data collection: Obtain optical power monitoring data from the optoelectronic double-layer network, pre-process the data, and divide the data into node data and link data;

[0010] S2, node and link data representation: each link data is represented as a sequence, and each node data is represented as a matrix;

[0011] S3, node and link data feature extraction: Use the sequence attention mechanism to extract link data features, whose input is the link power sequence and output is the feature sequence of link data; use the channel attention mechanism to extract node data features, whose input is the node power matrix and output is the feature sequence of node data;

[0012] S4, full network data feature extraction: The network topology is represented as an abstract graph, the node features in the abstract graph are initialized using the extracted node and link data features, and the full network data features are comprehensively extracted using the graph attention network;

[0013] S5. Fault location decision: Input the data features of the entire network into a fully connected neural network, and output a set of fault probability values ​​of network devices, and determine the device corresponding to the highest probability value as the faulty device.

[0014] Furthermore, OPM is deployed at the input and output ports of all pre-EDFA, line-EDFA and WSS in the optoelectronic double-layer network of step S1 to obtain the optical power monitoring data of the network equipment, and an OPM is equipped inside Tx and Rx to transmit and receive power monitoring data respectively.

[0015] Furthermore, step S1 adopts an event-triggered detection data collection method.

[0016] Furthermore, the data collection process of step S1 is as follows: when service packet loss occurs due to network equipment failure in the network, the controller collects a set of optical power monitoring data from the network, performs data reduction preprocessing operations on the data, and finally divides the power data into two categories: link data and node data according to the position of the OPM; wherein the node data includes Tx transmission power, Rx receiving power, input and output power of pre-EDFA and input and output power of WSS; and the link data packet passes through the input and output power of line-EDFA.

[0017] Furthermore, the process of extracting link data features using the sequential attention mechanism in step S3 is as follows: the input data is regarded as a KV key-value pair. First, the input data is passed through three different fully connected layers to obtain Q, {K1-K n} and {V1-V n}; Secondly, for Q and {K1-K n} performs a dot product operation and passes through a softmax layer to obtain a set of weight values ​​{a1-a n}; Then, {a1-a n} and {V1-V n} are multiplied and then added, and the result is used as the input of the fully connected layer; finally, FCL outputs a set of link feature sequences {L1-L m}.

[0018] Furthermore, the process of extracting node data features using the channel attention mechanism in step S3 is as follows: first, global pooling is used to compress the number of features of each sequence to 1; second, a set of numerical values ​​{S1-S c}; Next, the score value is multiplied by the corresponding sequence, and the result is used as the input of FCL; Finally, FCL outputs a set of node feature sequences {N0-N m}.

[0019] Furthermore, the process of initializing the node features in the abstract graph using the extracted node and link data features in step S4 is as follows: the hollow nodes in the network abstract graph represent the switching nodes in the network topology, and the node feature sequence {N1-N m} completes its feature initialization, and adds a solid node to the edge of the original network topology graph, using the link feature sequence {L1-L m}Complete its initialization.

[0020] Furthermore, the process of using the graph attention network to comprehensively extract the features of the entire network data in step S4 is as follows: the network abstract graph with completed feature initialization is used as the input of the graph attention network, a multi-head attention mechanism is introduced to help the graph attention network aggregate node information, and the features of each node output by the graph attention network are spliced ​​into a feature vector {Z1-Z n} to complete the feature extraction of the whole network data.

[0021] On the other hand, the present invention also provides an optical network fault location device based on an attention mechanism, comprising the following modules to implement any of the above methods:

[0022] The data acquisition module is used to obtain optical power monitoring data from the optoelectronic double-layer network, pre-process the data, and divide the data into node data and link data;

[0023] The node and link data representation module is used to represent each link data as a sequence and each node data as a matrix;

[0024] The node and link data feature extraction module is used to extract link data features using the sequence attention mechanism and node data features using the channel attention mechanism;

[0025] The whole network data feature extraction module is used to represent the network topology as an abstract graph, use the extracted node and link data features to complete the initialization of the node features in the abstract graph, and use the graph attention network to comprehensively extract the whole network data features;

[0026] The fault location decision module is used to input the data features of the entire network into a fully connected neural network, and output the fault probability values ​​of a group of network devices, and determine the device corresponding to the highest probability value as the faulty device.

[0027] In another aspect, the present invention further provides an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; wherein:

[0028] The memory is used to store computer programs;

[0029] The processor is used to implement any of the above methods when executing the program stored in the memory.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] The optical network fault location method and device based on the attention mechanism provided by the present invention can filter out the most important data from large-scale monitoring data, greatly improving the efficiency and accuracy of network fault location. In addition, this solution can accurately locate the faulty network device, which is more accurate than the existing solution, which can greatly reduce the overhead caused by service rerouting, and meet the needs of future 6G / F6G green network construction and efficient operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and for those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0033] Figure 1 A diagram of the optoelectronic double-layer network structure provided in an embodiment of the present invention.

[0034] Figure 2 A flow chart of an optical network fault location method based on an attention mechanism provided in an embodiment of the present invention.

[0035] Figure 3 A link power sequence is provided in an embodiment of the present invention.

[0036] Figure 4 A node power matrix provided by an embodiment of the present invention.

[0037] Figure 5 A structural diagram of the sequence attention mechanism provided for an embodiment of the present invention.

[0038] Figure 6 A structural diagram of the channel attention mechanism provided for an embodiment of the present invention.

[0039] Figure 7 A flow chart of network data feature extraction provided by an embodiment of the present invention.

[0040] Figure 8 A fault location decision flow chart provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to better understand the technical solution, the method of the present invention is described in detail below with reference to the accompanying drawings.

[0042] The optoelectronic double-layer network structure provided by the embodiment of the present invention is as follows Figure 1 As shown in the figure, one of the optoelectronic hybrid switching nodes consists of an electrical switch and a broadcast and select (B&S) ROADM. On the link, the fiber segment (fiber) is used to achieve long-distance transmission of signals. The line erbium-doped fiber amplifier (line-EDFA) is placed between two fiber spans to compensate for fiber attenuation. In the node, the wavelength signal is sent by a transmitter (Tx) or received by a receiver (Rx). The arrayed waveguide grating (AWG) is used to achieve wavelength multiplexing and demultiplexing. The pre-erbium-doped fiber amplifier (pre-EDFA) is used to compensate for the insertion loss of each device. The splitter (splitter) can disperse the wavelength signal in all directions of the ROADM, and the wavelength selective switch (WSS) can achieve flexible wavelength addition / drop. We refer to the equipment on the link, including fiber and line-EDFA, as link equipment, and the equipment on the node, including Tx, Rx, AWG, splitter and Pre-EDFA, as intra-node equipment.

[0043] The optical network fault location method based on attention mechanism proposed in the present invention is as follows: Figure 2 As shown, the input is a set of optical power monitoring data {P0-P n}, the output is the device that has failed in the network, including the following steps:

[0044] Step (1) Data acquisition: Obtain optical power monitoring data from the optoelectronic double-layer network, pre-process the data, and divide the data into node data and link data.

[0045] Specifically, the present invention deploys OPM at all input and output ports of pre-EDFA, line-EDFA and WSS in the optoelectronic double-layer network to obtain optical power monitoring data of network equipment. At the same time, Tx and Rx are each equipped with an OPM to transmit and receive power monitoring data respectively.

[0046] The present invention adopts an event-triggered detection data collection method. The data collection process is as follows: when a service packet is lost due to a network device failure in the network, the controller collects a set of optical power monitoring data from the network, performs data reduction preprocessing operations on the data, and finally divides the power data into two categories: link data and node data according to the position of the OPM; wherein the node data includes Tx transmission power, Rx receiving power, input and output power of pre-EDFA and input and output power of WSS; and the link data packet passes the input and output power of line-EDFA.

[0047] Step (2) Node and link data representation: each link data is represented as a sequence, and each node data is represented as a matrix.

[0048] Specifically, on the link, this solution regards the power data collected from the source end to the destination end of the link as a power sequence. Figure 3 As shown in the figure, the input and output powers {P0-P n} can be regarded as a link power sequence.

[0049] In a node, when traffic passes through a ROADM, the wavelength signal enters the ROADM from one direction and leaves from the other direction. Figure 1 As shown, the wavelength signal transmitted by Tx1 can enter the ROADM from direction1 through AWG1, and leave from direction2 through pre-EDFA1, plitter1 and WSS2. This scheme regards such a path as a channel. All OPM power data in the channel can be regarded as a power sequence. Due to the multi-dimensional characteristics of ROADM, there are many similar channels in a ROADM. Therefore, we convert the power data in a node into Figure 4 The power matrix shown in FIG. 1 , where each row of the matrix represents a power sequence. If the length of a sequence is less than 5, for example, the sequence from direction 2 to direction 3, this scheme will use -200 dBm to fill it.

[0050] Step (3) Node and link data feature extraction: Use the sequence attention mechanism to extract link data features, whose input is the link power sequence and output is the feature sequence of link data; use the channel attention mechanism to extract node data features, whose input is the node power matrix and output is the feature sequence of node data.

[0051] Specifically, the present invention uses a sequence attention mechanism to extract link data features, whose input is a link power sequence and output is a feature sequence of link data. The structure of the sequence attention mechanism is as follows: Figure 5 As shown. The present invention regards the input data as a KV key-value pair. The process is as follows: First, the input data is passed through three different fully connected layers to obtain Q, {K1-K n} and {V1-V n}; Secondly, for Q and {K1-K n} performs a dot product operation and passes through a softmax layer to obtain a set of weight values ​​{a1-a n}; Then, {a1-a n} and {V1-V n} are multiplied and then added, and the result is used as the input of the fully connected layer; finally, FCL outputs a set of link feature sequences {L1-L m}.

[0052] The present invention uses the channel attention mechanism to extract node data features, whose input is the node power matrix and output is the feature sequence of node data. The structure of the channel attention mechanism is as follows Figure 6 As shown in Figure 1, the process is as follows: first, global pooling is used to compress the number of features of each sequence to 1; second, a set of numerical values ​​{S1-S c}; Next, the score value is multiplied by the corresponding sequence, and the result is used as the input of FCL; Finally, FCL outputs a set of node feature sequences {N0-N m}.

[0053] Step (4) Extraction of network-wide data features: The network topology is represented as an abstract graph, the extracted node and link data features are used to initialize the node features in the abstract graph, and the graph attention network is used to comprehensively extract network-wide data features.

[0054] Specifically, Figure 7 As shown, first, this scheme represents the network topology as a network abstract graph. The hollow nodes in the network abstract graph represent the switching nodes in the network topology. The node feature sequence {N1-N m} completes its feature initialization, and adds a solid node to the edge of the original network topology graph, using the link feature sequence {L1-L m} to complete its initialization. Secondly, the network abstract graph with completed feature initialization is used as the input of the graph attention network (GAT), and a multi-head attention mechanism is introduced to help the graph attention network aggregate node information. Finally, the features of each node output by the graph attention network are concatenated into a feature vector {Z1-Z n} to complete the feature extraction of the whole network data.

[0055] Step (5) Fault location decision: The whole network data features {Z1-Z n} Input a fully connected neural network and output a set of network device failure probability values ​​{y1-y t}, the device corresponding to the highest probability value is determined as a faulty device. Figure 8 shown.

[0056] Corresponding to the optical network fault location method based on the attention mechanism provided in the above-mentioned embodiment of the present invention, the embodiment of the present invention also provides an optical network fault location device based on the attention mechanism.

[0057] The device includes:

[0058] The data acquisition module is used to obtain optical power monitoring data from the optoelectronic double-layer network, pre-process the data, and divide the data into node data and link data;

[0059] The node and link data representation module is used to represent each link data as a sequence and each node data as a matrix;

[0060] The node and link data feature extraction module is used to extract link data features using the sequence attention mechanism and node data features using the channel attention mechanism;

[0061] The whole network data feature extraction module is used to represent the network topology as an abstract graph, use the extracted node and link data features to complete the initialization of the node features in the abstract graph, and use the graph attention network to comprehensively extract the whole network data features;

[0062] The fault location decision module is used to input the data features of the entire network into a fully connected neural network, and output the fault probability values ​​of a group of network devices, and determine the device corresponding to the highest probability value as the faulty device.

[0063] Corresponding to the optical network fault location method based on the attention mechanism provided in the above embodiment of the present invention, the embodiment of the present invention further provides an electronic device, including: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus;

[0064] Memory, used to store computer programs;

[0065] The processor is used to implement the semi-federated learning method process provided by the above-mentioned embodiment of the present invention when executing the program stored in the memory, and implement the steps of the optical network fault location method based on the attention mechanism provided by the above-mentioned embodiment of the present invention.

[0066] The communication bus mentioned in the above control device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0067] The communication interface is used for communication between the above electronic device and other devices.

[0068] 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 storage device located away from the aforementioned processor.

[0069] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0070] In another embodiment provided by the present invention, a computer-readable storage medium is also provided, which stores a computer program. When the computer program is executed by a processor, the steps of any optical network fault location method based on the attention mechanism provided in the above-mentioned embodiments of the present invention are implemented.

[0071] In another embodiment provided by the present invention, a computer program product comprising instructions is also provided, which, when executed on a computer, enables the computer to execute the steps of any of the optical network fault location methods based on the attention mechanism provided by the above-mentioned embodiments of the present invention.

[0072] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by 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 described in the embodiment of the present invention 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 device. 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. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital terminal equipment line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.

[0073] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0074] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, electronic device embodiment, computer-readable storage medium embodiment, and computer program product embodiment, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. An optical network fault location method based on attention mechanism, characterized in that: The following steps are involved: S1. Data collection: Obtain optical power monitoring data from the optoelectronic double-layer network, pre-process the data, and divide the data into node data and link data; S2, node and link data representation: each link data is represented as a sequence, and each node data is represented as a matrix; S3, node and link data feature extraction: Use the sequence attention mechanism to extract link data features, whose input is the link power sequence and output is the feature sequence of link data; use the channel attention mechanism to extract node data features, whose input is the node power matrix and output is the feature sequence of node data; Step S3 uses the sequential attention mechanism to extract link data features. The process is as follows: the input data is regarded as a KV key-value pair. First, the input data is passed through three different fully connected layers to obtain Q, {K1-K n } and {V1-V n }; Secondly, for Q and {K1-K n } performs a dot product operation and passes through a softmax layer to obtain a set of weight values ​​{a1-a n }; Then, {a1-a n } and {V1-V n } are multiplied and then added, and the result is used as the input of the fully connected layer; finally, FCL outputs a set of link feature sequences {L1-L m }; The process of extracting node data features using the channel attention mechanism in step S3 is as follows: first, global pooling is used to compress the number of features of each sequence to 1; second, a set of numerical values ​​{S1-S c }; Next, the score value is multiplied by the corresponding sequence, and the result is used as the input of FCL; Finally, FCL outputs a set of node feature sequences {N0-N m }; S4, full network data feature extraction: The network topology is represented as an abstract graph, the node features in the abstract graph are initialized using the extracted node and link data features, and the full network data features are comprehensively extracted using the graph attention network; Step S4 uses the extracted node and link data features to complete the initialization of the node features in the abstract graph. The hollow nodes in the network abstract graph represent the switching nodes in the network topology. The node feature sequence {N1-N m } completes its feature initialization, and adds a solid node to the edge of the original network topology graph, using the link feature sequence {L1-L m Complete its initialization; Step S4 uses the graph attention network to comprehensively extract the features of the entire network data. The process is as follows: the network abstract graph with feature initialization is used as the input of the graph attention network, a multi-head attention mechanism is introduced to help the graph attention network aggregate node information, and the features of each node output by the graph attention network are spliced ​​into a feature vector {Z1-Z n }, to complete the feature extraction of the whole network data; S5. Fault location decision: Input the data features of the entire network into a fully connected neural network, and output the fault probability values ​​of a group of network devices, and determine the device corresponding to the highest probability value as the faulty device.

2. The optical network fault location method based on the attention mechanism according to claim 1 is characterized in that: In step S1, all pre-EDFA, line-EDFA and WSS input and output ports in the optoelectronic double-layer network are deployed with OPM to obtain the optical power monitoring data of the network equipment. Both Tx and Rx are equipped with an OPM to transmit and receive power monitoring data respectively.

3. The optical network fault location method based on the attention mechanism according to claim 2 is characterized in that: Step S1 adopts an event-triggered detection data collection method.

4. The optical network fault location method based on the attention mechanism according to claim 1 is characterized in that: The data collection process in step S1 is as follows: when service packet loss occurs due to network equipment failure in the network, the controller collects a set of optical power monitoring data from the network, performs data reduction preprocessing operations on the data, and finally divides the power data into two categories: link data and node data according to the position of the OPM; wherein the node data includes Tx transmission power, Rx receiving power, input and output power of pre-EDFA and input and output power of WSS; the link data packet passes through the input and output power of line-EDFA.

5. An optical network fault location device based on attention mechanism, characterized in that: The method comprises the following modules to implement any one of claims 1 to 4: The data acquisition module is used to obtain optical power monitoring data from the optoelectronic double-layer network, pre-process the data, and divide the data into node data and link data; The node and link data representation module is used to represent each link data as a sequence and each node data as a matrix; The node and link data feature extraction module is used to extract link data features using the sequence attention mechanism and node data features using the channel attention mechanism; The whole network data feature extraction module is used to represent the network topology as an abstract graph, use the extracted node and link data features to complete the initialization of the node features in the abstract graph, and use the graph attention network to comprehensively extract the whole network data features; The fault location decision module is used to input the data features of the entire network into a fully connected neural network, and output the fault probability values ​​of a group of network devices, and determine the device corresponding to the highest probability value as the faulty device.

6. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; and is characterized in that: The memory is used to store computer programs; The processor is used to implement the method described in any one of claims 1 to 4 when executing the program stored in the memory.

Images