A Cell Outage Detection Method and Device Based on the Change of Adjacent Visibility Hypergraph

Abstract

The present invention discloses a cell outage detection method and apparatus based on the change of an adjacency visibility hypergraph, which receives measurement information sent by active users within the coverage range of the current base station and constructs a list of neighboring cell units; constructs an adjacency visibility hypergraph according to the list of neighboring cell units; calculates a change vector of the adjacency visibility hypergraph based on the adjacency visibility hypergraphs at different times; uses the change vector as input information for a network outage classifier, and determines the network status of the cell to be detected according to the classifier; by constructing an adjacency visibility hypergraph through the list of neighboring cell units of active users, the present invention can combine the detection information of multiple users to improve the detection accuracy, and by calculating the change vector of the visibility hypergraph at different times, more objective signal detection data is obtained, further improving the network signal detection accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of network detection, and particularly to a cell interruption detection method and device based on the change of adjacent visibility hypergraph. Background Art

[0002] In order to meet the needs of different customers and application scenarios for the differentiated service capabilities of communication networks, NFV technology is introduced into the core network, virtualization technology of SDN is introduced into the transmission network, and factors such as ultra-large-scale antenna systems are introduced into the radio access network in 5G, which have caused the scale and complexity of 5G networks to increase sharply. At the same time, the service management method of 5G slices makes network deployment more dynamic and complex. These have greatly increased the difficulty of network operation and management.

[0003] Cell interruption detection is a way to perceive the network state in real time and automatically locate network faults. Traditional cell interruption detection is mainly based on signal measurement at the user cell level, and it is judged whether the cell is interrupted by comparing the RSRP value and RSRQ value received by the user with the corresponding threshold values respectively. However, the threshold values of various indicators in the traditional method are usually specified manually. With the increase of network coverage, the wireless signal changes greatly, and it is difficult to determine the threshold values, which will lead to the problem of low detection accuracy. Summary of the Invention

[0004] The purpose of the present invention is to provide a cell interruption detection method and device based on the change of adjacent visibility hypergraph, establish an adjacent visibility hypergraph, and use its change amount as detection data. Taking objective data as the detection quantity, there is no need to set threshold values, which can improve the detection accuracy.

[0005] The present invention adopts the following technical solutions: A cell interruption detection method based on the change of adjacent visibility hypergraph, comprising the following steps:

[0006] Receiving measurement information sent by active users within the coverage range of the current base station and constructing a list of adjacent cell units;

[0007] Constructing an adjacent visibility hypergraph according to the list of adjacent cell units;

[0008] Calculating the change vector of the adjacent visibility hypergraph based on the adjacent visibility hypergraphs at different times;

[0009] Taking the change vector as the input information of the network interruption classifier, and determining the network state of the cell to be detected according to the classifier.

[0010] Further, receiving measurement information sent by active users within the coverage range of the current base station and constructing a list of adjacent cell units includes:

[0011] Receiving the physical layer cell identifier sent by the active user that meets the configuration request;

[0012] Query whether the physical layer cell identifier exists in the neighbor cell list of the current base station;

[0013] When the physical layer cell identifier is not in the neighbor cell list, send a global cell identifier measurement configuration request to the active user;

[0014] Receive the global cell identifier sent by the active user that meets the configuration request;

[0015] Add the global cell identifier to the neighbor cell unit list.

[0016] Further, before receiving the physical layer cell identifier sent by the active user that meets the configuration request includes:

[0017] When the active user detects that the beam signal strength of the serving cell is lower than the first threshold, detect the second signal strength of all other received beams;

[0018] When the second signal strength is greater than the second threshold, use the corresponding beam signal as the signal that meets the configuration request.

[0019] Further, before constructing the adjacency visibility hypergraph according to the neighbor cell unit list includes:

[0020] Receive the neighbor cell unit list sent by the adjacent cells in the neighbor cell unit list.

[0021] Further, constructing the adjacency visibility hypergraph according to the neighbor cell unit list includes:

[0022] Use the beam ID in the neighbor cell unit list as the nodes of the adjacency visibility hypergraph, all visible beams in the cell as the hyperedges of the adjacency visibility hypergraph, and the visible beams of each active user as the regular edges to establish the adjacency visibility hypergraph;

[0023] Express the adjacency visibility hypergraph through a matrix; the rows in the matrix are the nodes of the adjacency visibility hypergraph, and the columns in the matrix are the hyperedges and regular edges of the adjacency visibility hypergraph.

[0024] Further, calculating the change vector of the adjacency visibility hypergraph based on the adjacency visibility hypergraphs at different times includes:

[0025] Calculate the number of changed bits of all columns in the matrix;

[0026] Combine the number of changed bits of all columns to obtain the change vector.

[0027] Further, the training method of the network interruption classifier is:

[0028] Collect the change vectors of the cells with normal network status and label each change vector as the first training data set;

[0029] Collect the change vectors of the cells with network interruption, and label each change vector as the second training data set.

[0030] Train a network interruption classifier based on the first training data set and the second training data set.

[0031] Furthermore, the network interruption classifier is a knowledge-based classifier or a multivariate statistics-based classifier.

[0032] Another technical solution of the present invention is: a cell interruption detection device based on the change of adjacent visibility hypergraph, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned cell interruption detection method based on the change of adjacent visibility hypergraph.

[0033] The beneficial effects of the present invention are: the present invention constructs an adjacent visibility hypergraph through the list of adjacent cell units of active users, which can combine the detection information of multiple users, improve the detection accuracy, and obtain more objective signal detection data by calculating the change vectors of the visibility hypergraph at different times, further improving the network signal detection accuracy. Description of the Drawings

[0034] Figure 1 It is a schematic diagram of the ANR interaction process in the LTE system in the embodiment of the present invention;

[0035] Figure 2 It is a schematic diagram of the adjacent visibility hypergraph in the embodiment of the present invention;

[0036] Figure 3 It is a schematic diagram of the adjacent visibility hypergraph after the cell Cell 2 goes down in the embodiment of the present invention;

[0037] Figure 4 It is a flow chart of the network interruption classifier training and downtime detection in the embodiment of the present invention. Detailed Embodiments

[0038] The present invention will be described in detail below in conjunction with the drawings and specific embodiments.

[0039] The present invention discloses a cell interruption detection method based on the change of adjacent visibility hypergraph, including the following steps: receiving the measurement information sent by active users within the coverage of the current base station and constructing a list of adjacent cell units; constructing an adjacent visibility hypergraph according to the list of adjacent cell units; calculating the change vectors of the adjacent visibility hypergraph based on the adjacent visibility hypergraphs at different times; using the change vectors as the input information of the network interruption classifier, and determining the network status of the cell to be detected according to the classifier.

[0040] In view of the problem of service interruption caused by faults or parameter configuration problems in the cell network facilities, the present invention constructs an adjacency visibility hypergraph through the list of neighboring cell units of active users, which can combine the detection information of multiple users to improve the detection accuracy. And by calculating the change vectors of the visibility hypergraph at different times, more objective signal detection data is obtained, further improving the network signal detection accuracy.

[0041] In one embodiment, as Figure 1 shown, receiving the measurement information sent by active users within the coverage of the current base station and constructing a list of neighboring cell units includes: receiving the physical layer cell identifier that meets the configuration request sent by the active user; querying whether the physical layer cell identifier exists in the neighboring cell list of the current base station; when the physical layer cell identifier is not in the neighboring cell list, sending a global cell identifier measurement configuration request to the active user; receiving the global cell identifier that meets the configuration request sent by the active user; and adding the global cell identifier to the list of neighboring cell units.

[0042] Specifically, before receiving the physical layer cell identifier that meets the configuration request sent by the active user, it includes:

[0043] In the embodiment of the present invention, the source base station (such as the eNodeB in the LTE system) issues UE (user equipment) measurement configuration information, instructing the UE to measure the surrounding cells according to the configuration requirements. The measurement configuration requirements are that when the active user detects that the beam signal strength of the serving cell is lower than the first threshold, detect the second signal strength of all beam signals of other cells received; when the second signal strength is greater than the second threshold, use the corresponding beam signal as the signal that meets the configuration request.

[0044] Generally, the UE defaults to perform co-frequency measurement. When the UE has established a radio bearer, the eNodeB can default to issue co-frequency handover measurement configuration information through the signaling RRC Connection Reconfiguration. When the UE needs to perform inter-frequency measurement, the eNodeB needs to issue inter-frequency handover measurement configuration information. Generally, the UE has only one receiver and can only receive signals at one frequency point at the same time, and it is impossible to receive two frequency point information simultaneously. Therefore, the measurement GAP (inter-frequency measurement interval) is the time period for the UE to leave the current frequency point to measure other frequency points. The measurement GAP is used for inter-frequency measurement and inter-system measurement. In inter-frequency and inter-system measurement, the UE only measures within the measurement GAP. Activate the GAP mode to perform inter-frequency measurement.

[0045] Exemplarily, the UE in Cell A reports the PCI (Physical Layer Cell Identity) of Cell B (i.e., the serving cell) that meets the measurement configuration requirements (i.e., the measurement configuration request) to the source eNodeB Cell A (i.e., Cell A) in the form of a measurement report. The source eNodeB queries whether the PCI of Cell B exists in the in-system NCL (i.e., the neighbor cell list) of Cell A. If it exists, the process exits; if it does not exist, the source eNodeB sends a measurement configuration request to the UE, requesting the UE to read parameters such as the ECGI (E-UTRAN Cell Global Identity), TAC (Tracking Area Code), and PLMN ID list (Public Land Mobile Network Identity list) of Cell B.

[0046] The UE reports the parameter information such as the ECGI, TAC, and PLMN ID list of Cell B that it has read to the source eNodeB. The source eNodeB adds the neighbor cell information of the newly discovered Cell B to the in-system NCL of the source eNodeB, and at the same time adds the neighbor cell relationship to the in-system NRT of Cell A.

[0047] The UE in Cell B constructs the NCL of Cell B in the same way.

[0048] In one embodiment, before constructing the adjacency visibility hypergraph according to the neighbor cell list, it includes: receiving the neighbor cell list sent by adjacent cells in the neighbor cell list. In order to obtain more useful information, within a certain area, base stations will exchange information with each other through standard interfaces such as X2 to better provide services.

[0049] As a specific implementation method, constructing the adjacency visibility hypergraph according to the neighbor cell list includes: using the beam ID in the neighbor cell list as the nodes of the adjacency visibility hypergraph, using all visible beams in the cell as the hyperedges of the adjacency visibility hypergraph, and using the visible beams of each active user as the regular edges to establish the adjacency visibility hypergraph; expressing the adjacency visibility hypergraph through a matrix; the rows in the matrix are the nodes of the adjacency visibility hypergraph, and the columns in the matrix are the hyperedges and regular edges of the adjacency visibility hypergraph.

[0050] The neighbor cell list (NCL) report can create a visibility relationship graph (i.e., the adjacency visibility hypergraph), which is generated based on the NCL report received by the mobile terminal (UE). The elements of the adjacency hypergraph include hyperedges, edge weights, and vertices. Among them, the edge weight represents the number of mobile terminals that have reported a specific neighbor cell relationship, and the vertex represents the beam ID.

[0051] The NCL report is always generated when the UE is in an active connected state. At this time, the UE continuously measures the beam signal strength and quality of the radio channel of the serving cell to which it is currently connected, and also measures the beam signals of some neighboring cells, which constitute the candidate cells for potential handover. These measurement results are sent to the base station of the current serving cell to determine whether a handover must be performed. The terminal in the process of connection can measure up to M base stations within its visible range and send the NCL report at intervals of T, which contains N best measurement results.

[0052] For example, in the GSM system, the terminal in the process of connection can measure up to 16 base stations within its visible range and send the NCL report at intervals of 480 ms, which contains 6 best measurement results. The NCL reports of different terminals are retrieved through a subordinated network entity. The details of the measurement reports vary from GSM to UMTS or LTE, but it is assumed that the list of neighboring cells is reported at regular time intervals, which can be achieved in different radio access technologies.

[0053] Based on the measurement reports of different terminals, a visibility hypergraph is constructed, and observing the changes in the elements in the visibility hypergraph is the key element for detecting the cell outage problem proposed in the present invention. Specifically, the adjacent visibility hypergraph is created at fixed time intervals, and any abnormal change in this adjacent visibility hypergraph may contain hint information about cell outage. As Figure 2 shown, a specific example of such a hypergraph is depicted. Two consecutive visibility hypergraphs G(t1) and visibility hypergraph G(t1 + T) are established through the time interval T. For the convenience of description, hereinafter referred to as the graph, the detection method proposed in this embodiment is sensitive to any node that becomes isolated in G(t1 + T), and in this case, a change pattern of the graph is created.

[0054] In Figure 2 , user A1 is currently served by the beam V11 of cell 1 and can receive the beam signal of the beam v22 of Cell 2. User A2 is currently served by the beam V13 of cell 1 and can receive the beam signal of the beam v21 of Cell 2. And user B1 is currently served by the beam V22 of cell 2 and can receive the beam signal of the beam v14 of Cell 1. User B2 is currently served by the beam V23 of cell 2 and can receive the beam signal of the beam v12 of Cell 1.

[0055] In this example figure, the coverage visibility ranges of two user terminals in cell A are circles around their respective locations. The NCL reports of users A1 and A2 include the beams V11 and V13 of serving cell cell 1 as the current serving beams, and the beams V21 and V22 of adjacent cell cell 2, which are the beam IDs of adjacent cells within their visible ranges. Correspondingly, the NCL reports of users B1 and B2 in serving cell B will also be constructed.

[0056] In this embodiment, cell 1 and cell 2 exchange information with each other, and then construct a visibility hypergraph based on the NCL reports of four users (i.e., A1, A2, B1, and B2).

[0057] The visibility hypergraph consists of nodes, hyperedges, and regular edges. Among them, the nodes are composed of beams under the cell, and the hyperedges are composed of the beams serving users in the cell and the visible beams reported by users in other cells. In addition to hyperedges, the edges connecting nodes are regular edges, which usually connect two nodes and are composed of the serving beam of one user and the visible beam of another cell.

[0058] In the embodiment of the present invention, for the convenience of description and calculation, the adjacent visibility hypergraph is expressed by a matrix; the rows in the matrix are the nodes of the adjacent visibility hypergraph, and the columns in the matrix are the hyperedges and regular edges of the adjacent visibility hypergraph.

[0059] For example, as shown in Table 1 below, the hyperedge He1 (connecting 4 nodes: v11, v12, v13, v14) includes two parts: 1) composed of the serving beam IDs (v11, v13) in the NCL reports of users A1 and A2; 2) composed of the visible beam IDs in the NCL reports of users B1 and B2 (v12, v14). Similarly, the hyperedge He2 (connecting 3 nodes: v21, v22, v23) also includes two parts: 1) composed of the visible beam IDs (v21, v22) in the NCL reports of users A1 and A2; 2) composed of the serving beam IDs (v22, v23) in the NCL reports of users B1 and B2.

[0060] The hyperedge graph can be represented by a matrix. For the nodes included in the hyperedge, the value at the intersection of the column where the hyperedge is located and the row where the included nodes are located is 1, otherwise it is 0. For the two nodes connected by an edge, the positions of the corresponding nodes in the row where they are located and the column where the edge is located are 1, otherwise it is 0.

[0061] Table 1

[0062] He1 He2 E1 E2 E3 E4 V11 1 0 1 0 0 0 V12 1 0 0 1 0 0 V13 1 0 0 0 1 0 V14 1 0 0 0 0 1 V21 0 1 0 0 1 0 V22 0 1 1 0 0 1 V23 0 1 0 1 0 0 V24 0 0 0 0 0 0

[0063] During the normal operation of the network, the visibility hypergraph changes frequently. These changes are caused by the start and end of calls, user mobility, changes in radio propagation (such as millimeter-wave blockage), and changes in the NCL reports themselves.

[0064] The visibility hypergraph will change when the network experiences a downtime. For example, the hypergraph caused by the downtime of cell Cell 2 is as Figure 3 shown. Compared with the original state of each cell (i.e., the non-downtime situation), Figure 3 the NCL reports of user B (i.e., including B1 and B2) in

[0065] are no longer received, resulting in the disappearance of the visible beams of user B (i.e., user B cannot complete the measurement report), such as (v12, v14). At the same time, the serving beams of the users in cell 2 disappear, such as (v22, v23), and the visible beams of the users in Cell A are (v21, v22). Each downtime situation will cause the disappearance of a hyperedge in the visibility hypergraph.

[0066] Table 2

[0067] He1 He2 E1 E2 E3 E4 V11 1 0 0 0 0 0 V12 0 0 0 0 0 0 V13 1 0 0 0 0 0 V14 0 0 0 0 0 0 V21 0 0 0 0 0 0 V22 0 0 0 0 0 0 V23 0 0 0 0 0 0 V24 0 0 0 0 0 0

[0068] Table 3

[0069] He1 He2 E1 E2 E3 E4 V11 0 0 -1 0 0 0 V12 -1 0 0 -1 0 0 V13 0 0 0 0 -1 0 V14 -1 0 0 0 0 -1 V21 0 -1 0 0 -1 0 V22 0 -1 -1 0 0 0 V23 0 -1 0 -1 0 0 V24 0 0 0 0 0 0

[0070] Furthermore, calculating the change vector of the adjacent visibility hypergraph according to the adjacent visibility hypergraphs at different times includes: calculating the number of changed bits in all columns of the matrix; combining the number of changed bits in all columns to obtain the change vector.

[0071] It can be seen from this that the change amount of the degree of hyperedge 1 is 2, the change amount of the degree of hyperedge 2 is 3, the change amount of the degree of edge 1 is 2, the change amount of the degree of edge 2 is 2, the change amount of the degree of edge 3 is 2, and the change amount of the degree of edge 4 is 2. Then the corresponding hypergraph pattern change vector is: X = [2, 3, 2, 2, 2, 2]. Using this change vector as input to the classifier, the network state prediction of the cell can be obtained.

[0072] In the embodiment of the present invention, the training method of the network interruption classifier is: collecting the change vectors of the cells with normal network states and labeling each change vector as a first training data set; collecting the change vectors of the cells with network interruptions and labeling each change vector as a second training data set; training the network interruption classifier based on the first training data set and the second training data set.

[0073] As can be seen from the above content, a classification model is trained using the vector X composed of the above data, and the outage situation of the cell can be judged through the output result. Since the outage of the cell will generate a characteristic change pattern, through the change pattern, it can be distinguished from the normal fluctuation of the visible graph, so the outage detection problem can be transformed into a classification problem of the change pattern of the visible graph.

[0074] The classification of a set of items is to assign similar items to one of several different categories. The task of classifying the change patterns of the visible graph into interrupted and non-interrupted situations can be regarded as a binary classification problem with a set of predefined categories. Classification algorithms are widely used in automatic pattern recognition. In the embodiments of the present invention, the network interruption classifier is a knowledge-based classifier (such as an expert system, a neural network, a genetic algorithm, etc.) or a classifier based on multivariate statistical data (such as cluster analysis, classification and regression trees, etc.).

[0075] In summary, the outage detection algorithm example based on the classifier can adopt any of the above classification techniques. The basic steps are shown in Figure 4 the training and outage detection flowcharts.

[0076] Collect the neighbor cell measurement reports of the cell users in the normal cell, and construct the visibility hypergraph and the change pattern of the visible graph accordingly; in the abnormal outage cell, collect the neighbor cell measurement reports of the cell users, and construct the visibility hypergraph and the change pattern of the visible graph accordingly.

[0077] Label the normal cell hypergraph pattern change vector data set {X m , m = 1, 2,..., M} as normal {Y m = 1, m = 1, 2,..., M}, and label the abnormal outage cell hypergraph pattern change vector data set {X n , n = 1, 2,..., N} as abnormal {Y n = 0, n = 1, 2,..., N}.

[0078] Use {X m , Y m} and {X n , Y n} to train the classifier; finally, use the trained classifier for outage detection: that is, collect the neighbor cell measurement reports of the cell users to be detected, construct the visibility hypergraph and the visible graph change pattern vector X accordingly, input X into the classifier, and the output result is the detection result.

[0079] The present invention also discloses a cell interruption detection device based on the change of the adjacent visibility hypergraph, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned cell interruption detection method based on the change of the adjacent visibility hypergraph.

[0080] The above-mentioned device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the device may include more or fewer components, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0081] The processor can be a Central Processing Unit (CPU), and the processor can also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0082] The memory can be an internal storage unit of the device in some embodiments, such as the hard disk or memory of the device. The memory can also be an external storage device of the device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the device. Further, the memory can also include both the internal storage unit and the external storage device of the device. The memory is used to store an operating system, application programs, a BootLoader, data, and other programs, such as the program code of the computer program, etc. The memory can also be used to temporarily store data that has been output or will be output.

[0083] It should be noted that for the specific content of the above-mentioned device, since it is based on the same concept as the method embodiment of the present invention, its specific functions and the technical effects brought about can be specifically referred to in the method embodiment section, and will not be elaborated here.

Claims

1. A cell outage detection method based on the change of adjacency visibility hypergraph, characterized in that It includes the following steps: Receiving measurement information sent by active users within the coverage of the current base station and constructing a list of neighboring cell units; Constructing an adjacency visibility hypergraph based on the list of neighboring cell units; Calculating a change vector of the adjacency visibility hypergraph based on the adjacency visibility hypergraphs at different times; Taking the change vector as input information of a network outage classifier and determining the network status of the cell to be detected according to the classifier; Constructing an adjacency visibility hypergraph based on the list of neighboring cell units includes: Using the beam IDs in the list of neighboring cell units as the nodes of the adjacency visibility hypergraph, all visible beams within the cell as the hyperedges of the adjacency visibility hypergraph, and the visible beams of each active user as the regular edges to establish the adjacency visibility hypergraph; Expressing the adjacency visibility hypergraph through a matrix; the rows in the matrix are the nodes of the adjacency visibility hypergraph, and the columns in the matrix are the hyperedges and regular edges of the adjacency visibility hypergraph.

2. The cell outage detection method based on the change of the adjacent visibility hypergraph according to claim 1, characterized in that Receiving measurement information sent by active users within the coverage of the current base station and constructing a list of neighboring cell units includes: Receiving the physical layer cell identifier that meets the configuration request sent by the active user; Querying whether the physical layer cell identifier exists in the neighboring cell list of the current base station; When the physical layer cell identifier is not in the neighboring cell list, sending a global cell identifier measurement configuration request to the active user; Receiving the global cell identifier that meets the configuration request sent by the active user; Adding the global cell identifier to the list of neighboring cell units.

3. The cell outage detection method based on the change of the adjacency visibility hypergraph according to claim 2, wherein Before receiving the physical layer cell identifier that meets the configuration request sent by the active user includes: When the active user detects that the beam signal strength of the serving cell is lower than the first threshold, detecting the second signal strength of all beams of other cells received; When the second signal strength is greater than the second threshold, taking the corresponding beam signal as the signal that meets the configuration request.

4. A cell outage detection method based on the change of the adjacency visibility hypergraph according to claim 2 or 3, characterized in that, Before constructing an adjacency visibility hypergraph based on the list of neighboring cell units includes: Receiving the list of neighboring cell units sent by neighboring cells in the list of neighboring cell units.

5. The cell outage detection method based on the change of the adjacency visibility hypergraph according to claim 4, characterized in that, Calculating a change vector of the adjacency visibility hypergraph based on the adjacency visibility hypergraphs at different times includes: Calculating the number of changed bits of all columns in the matrix; Combining the number of changed bits of all columns to obtain the change vector.

6. A cell outage detection method based on the change of adjacency visibility hypergraph according to claim 4 or 5, characterized in that, The training method of the network outage classifier is: Collecting the change vectors of cells with normal network status and labeling each change vector as the first training data set; Collecting the change vectors of cells with network outages and labeling each change vector as the second training data set; Training the network outage classifier based on the first training data set and the second training data set.

7. The cell outage detection method based on the change of the adjacency visibility hypergraph according to claim 6, wherein The network outage classifier is a knowledge-based classifier or a multivariate statistical data-based classifier.

8. A cell outage detection device based on the change of an adjacency visibility hypergraph, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements a cell outage detection method based on the change of adjacency visibility hypergraph as described in any one of claims 1-7.

Images