Target signal recognition

Abstract

Embodiments of the present disclosure provide a target signal recognition method and apparatus, a computer program product, a computer device, and a storage medium. The target signal recognition method comprises: determining a plurality of initial signals sent by a plurality of signal providing ends, and determining signal feature information of each initial signal; on the basis of the plurality of initial signals and the signal feature information of each initial signal, constructing a signal relationship graph; determining associated information of each initial signal in the signal relationship graph, and on the basis of the associated information, determining, from the signal relationship graph, a signal group to be processed, wherein said signal group comprises a plurality of initial signals having an association relationship; and under the condition that the attribute information of the plurality of initial signals having the association relationship is determined, and a signal recognition condition is met, recognizing said signal group as a target signal.

Description

[0001] Target signal recognition technology field

[0002]

[0001] The present disclosure relates to the field of computer technology, and more particularly to target signal recognition.

[0003]

[0002] With the continuous development of computer technology, data signals are applied to various scenarios to provide services; a data signal source informs its current operating status by sending data signals.

[0004] However, in the related art, a large number of data signal sources transmit a large number of signals, which makes it impossible to timely and accurately determine the more important signals from the large number of signals. This further makes it impossible to accurately determine the current operating status of the data signal source. Therefore, how to determine the more important signals from the large number of signals has become an urgent problem to be solved.

[0005] In view of this, embodiments of the present disclosure provide a target signal recognition method. One or more embodiments of the present disclosure also relate to a target signal recognition apparatus, a computing device, a computer-readable storage medium, and a computer program product to address technical deficiencies in related technologies.

[0006]

[0005] According to a first aspect of an embodiment of the present disclosure, a target signal identification method is provided, comprising: determining multiple initial signals sent by multiple signal providing terminals, and determining signal characteristic information of each initial signal; constructing a signal relationship diagram based on the multiple initial signals and the signal characteristic information of each initial signal; determining association information of each initial signal in the signal relationship diagram, and based on the association information, determining a signal group to be processed from the signal relationship diagram, wherein the signal group to be processed includes multiple initial signals with an associated relationship; and when the attribute information of the multiple initial signals with an associated relationship is determined and a signal identification condition is met, identifying the signal group to be processed as a target signal.

[0007]

[0006] According to the second aspect of the embodiment of the present disclosure, a target signal identification device is provided, comprising: an initial signal determination module, configured to determine multiple initial signals sent by multiple signal providing terminals, and determine the signal characteristic information of each initial signal; a relationship diagram construction module, configured to construct a signal relationship diagram based on the multiple initial signals and the signal characteristic information of each initial signal; a signal group determination module, configured to determine the association information of each initial signal in the signal relationship diagram, and based on the association information, determine a signal group to be processed from the signal relationship diagram, wherein the signal group to be processed includes multiple initial signals with an associated relationship; a target signal identification module, configured to identify the signal group to be processed as a target signal when the attribute information of the multiple initial signals with an associated relationship is determined and a signal identification condition is met.

[0008]

[0007] According to a third aspect of an embodiment of the present disclosure, a computing device is provided, comprising: a memory and a processor; the memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, and when the computer programs / instructions are executed by the processor, the steps of the above-mentioned target signal recognition method are implemented.

[0009]

[0008] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program / instructions. When the computer program / instructions are executed by a processor, the computer program / instructions implement the steps of the target signal recognition method described above.

[0009] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, including a computer program / instructions. When the computer program / instructions are executed by a processor, the computer program / instructions implement the steps of the target signal recognition method described above.

[0010]

[0010] A target signal identification method provided by an embodiment of the present disclosure includes: determining multiple initial signals sent by multiple signal providing terminals, and determining signal characteristic information of each initial signal; constructing a signal relationship diagram based on the multiple initial signals and the signal characteristic information of each initial signal; determining association information of each initial signal in the signal relationship diagram, and based on the association information, determining a signal group to be processed from the signal relationship diagram, wherein the signal group to be processed includes multiple initial signals with an associated relationship; when the attribute information of the multiple initial signals with an associated relationship is determined and the signal identification condition is met, the signal group to be processed is identified as a target signal.

[0011]

[0011] Specifically, the target signal identification method provided by the present disclosure utilizes multiple initial signals sent by multiple signal providing terminals and the signal characteristic information of each initial signal to construct a signal relationship diagram, and determines a signal group to be processed containing multiple initial signals with associated relationships based on the signal relationship diagram; by grouping multiple initial signals into a signal group to be processed for processing, the efficiency of determining the target signal is improved; then, when the attribute information of the multiple initial signals with associated relationships is determined to meet the signal identification conditions, the signal group to be processed is identified as the target signal; thereby accurately identifying a more important target signal from the multiple initial signals, facilitating accurate determination of the current state of the signal providing terminal based on the target signal.

[0012] FIG1 is a schematic diagram of an application of a target signal recognition method provided by one embodiment of the present disclosure;

[0013] FIG2 is a flow chart of a target signal recognition method provided by one embodiment of the present disclosure;

[0014] FIG3 is a schematic diagram of a signal relationship diagram in a target signal recognition method provided by one embodiment of the present disclosure;

[0015] FIG4 is a schematic diagram of a preset signal type in a target signal recognition method provided by one embodiment of the present disclosure;

[0016] FIG5 is a schematic diagram of another signal relationship diagram in a target signal identification method provided by one embodiment of the present disclosure;

[0017] FIG6 is a flowchart of a target signal recognition method according to an embodiment of the present disclosure;

[0018] FIG7 is a schematic diagram of the structure of a target signal recognition device provided by one embodiment of the present disclosure;

[0019]

[0019] FIG8 is a block diagram of a computing device provided by an embodiment of the present disclosure.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure. However, the present disclosure can be implemented in many other ways different from those described herein, and those skilled in the art can make similar promotions without violating the connotation of the present disclosure. Therefore, the present disclosure is not limited by the specific implementation disclosed below.

[0021]

[0021] The terms used in one or more embodiments of the present disclosure are intended only to describe specific embodiments and are not intended to limit one or more embodiments of the present disclosure. The singular forms "a," "an," "the," and "the" used in one or more embodiments of the present disclosure and the appended claims are intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of the present disclosure refers to and includes any or all possible combinations of one or more associated listed items.

[0022] It should be understood that although the terms "first," "second," and so on may be used to describe various information in one or more embodiments of the present disclosure, such information should not be limited to these terms. These terms are merely used to distinguish information of the same type from one another. For example, "first" may also be referred to as "second," and similarly, "second" may also be referred to as "first," without departing from the scope of one or more embodiments of the present disclosure. Depending on the context, the term "if" as used herein may be interpreted as "at the time," "when," or "in response to a determination."

[0023]

[0023] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of the present disclosure are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0024]

[0024] First, the terms involved in one or more embodiments of the present disclosure are explained.

[0025]

[0025] GDB: Graph database, a data management system that uses nodes and edges as basic storage units and is designed with the principle of efficient storage and query of graph data.

[0026] Ping: It is a network tool used to test network connections and delays between hosts.

[0027]

[0027] With the continuous development of computer technology, data signals are being applied to various scenarios to provide services. Data signal sources transmit data signals to communicate their current operating status. However, a large number of data signal sources transmit a large number of signals, making it difficult to promptly and accurately identify the more important signals from the large number of signals. For example, in the event of a major fault, the alarm system easily generates a large number of alarm messages from different data sources, thereby affecting the efficiency of finding effective information. For another example, when a flood alarm occurs, the flood alarm will drown out the alarms corresponding to other concurrent fault sources, making it impossible to effectively identify important alarm information. Furthermore, flood alarm signals may be out of order or partially lost due to a fault. In such cases, it is impossible to promptly and effectively determine the root cause sequence.

[0028]

[0028] To address the above-mentioned issues, a signal processing solution has been proposed. This solution converges or suppresses data signals of the same attributes or type (for example, converges or suppresses alarm signals of the same attributes or type). However, it should be noted that this solution has significant technical drawbacks. First, this solution does not have a good way to achieve convergence for multi-source signals, and therefore cannot solve the problem of convergence for multi-source alarm signals. Second, this solution cannot implement operations that do not involve scenario-specific risks. Finally, this solution requires strict targeting of defined data signals. When key data signals are missing, it is impossible to accurately determine the current status of the data signal source. For example, this solution requires strict targeting of defined alarm signals. When key alarm signals are missing, it is impossible to detect significant risks. Therefore, this solution cannot effectively solve the above-mentioned technical problems.

[0029]

[0029] Based on this, the present disclosure provides a target signal recognition method. The present disclosure also relates to a target signal recognition device, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

[0030] Referring to FIG. 1 , FIG. 1 illustrates a schematic diagram of a target signal identification method according to an embodiment of the present disclosure. The target signal identification method is applied to a server 102, which can receive multiple alarm signals sent by multiple signal providers 104. Since alarm systems often generate a large number of alarm signals from diverse data sources, information organization requires considerable time. Therefore, the target signal identification method provided by the present disclosure leverages the capabilities of graph databases to define a signal relationship graph (i.e., a connectivity graph) based on alarm signals and their corresponding signal feature information. Each node represents an alarm signal, and each edge represents a relationship between two alarm signals. Through this relationship (keyword / information), the two alarms can be associated. Signal nodes with associated relationships are searched for in the signal relationship graph. Based on these associated signal nodes, a signal group to be processed is constructed. Attribute information of the multiple associated initial signals is determined. If signal identification conditions are met, the signal group to be processed is identified as a critical alarm signal. This achieves the goals of rapidly discovering risks, achieving rapid convergence, and extracting effective information.

[0031]

[0031] Referring to FIG. 2 , FIG. 2 shows a flow chart of a target signal recognition method provided according to an embodiment of the present disclosure, which specifically includes the following steps.

[0032]

[0032] Step 202: Determine multiple initial signals sent by multiple signal providers, and determine signal characteristic information of each initial signal.

[0033]

[0033] The signal provider can be understood as the signal source that provides the initial signal; the initial signal can be understood as a data signal that represents the current operating status of the signal provider; and the signal provider and initial signal can be configured according to different application scenarios. In one or more embodiments provided herein, when the target signal identification method is applied in a network risk identification scenario, the signal provider can be a log file or a Ping tool on a device. In other words, in the target signal identification method provided herein, the initial signal can be obtained from a log file or received from a Ping tool. Based on this, the initial signal can be understood as a data signal that represents the current operating status of the device. For example, the initial signal can be an alarm signal, including but not limited to a fault event or network interruption event in a log file, or a network interruption alarm signal or network delay alarm signal sent by a Ping tool. When the target signal identification method is applied in a device risk monitoring scenario, the signal provider can be a device, such as a client, various types of IoT devices, or a device sensor. The initial signal can be understood as a data signal representing the current operating status of the device. For example, the initial signal can be an alarm signal, a fault signal, etc. When the target signal identification method is applied to a server risk monitoring scenario, the signal provider can be a server. For example, the signal provider can be a server (physical machine), a cloud server, etc. The initial signal can be understood as a data signal representing the current operating status of the server. For example, the initial signal can be a downtime signal, an alarm signal, a fault signal, etc. Based on this, it can be seen that the initial signal can be set according to the actual application scenario and is not limited here.

[0034] Signal characteristic information can be understood as information carried by the initial signal and used to characterize the signal characteristics of the initial signal. The signal characteristic information can be used to characterize the signal characteristics of the initial signal in the form of a numerical value, a type name, a number, a character, etc., without specific limitation herein. For example, the signal characteristic information can be understood as a specific keyword or a specific logical relationship carried by the initial signal. The keyword can be a common key. Different initial signals have the same characteristic (common key). An edge between the initial signals indicates that the initial signals have a common characteristic (key). For example, the common characteristic (common key) can be regional information. It should be noted that signal characteristic information can be of various different types.

[0035] In one or more embodiments provided in the present disclosure, determining multiple initial signals sent by multiple signal providers includes: receiving multiple to-be-processed signals sent by the multiple signal providers; determining a signal type of each to-be-processed signal, and determining a to-be-processed signal whose signal type is consistent with a target signal type as a to-be-merged signal, wherein there are multiple to-be-merged signals; determining corresponding similar signals for each of the multiple to-be-merged signals from historical signals based on the signal types of the multiple to-be-merged signals; and using each to-be-merged signal in the multiple to-be-merged signals, as well as the similar signal corresponding to each to-be-merged signal, as an initial signal to obtain the multiple initial signals.

[0035]

[0036] The signal to be processed can be understood as a data signal sent by the signal provider that needs to be processed. For example, the signal to be processed can be an alarm signal, a fault signal, etc. Such signal types include, but are not limited to, fiber disconnection, board offline, circuit group disconnection, chassis failure, delay warning, circuit group fullness, packet loss, etc., and are not specifically limited here.

[0036]

[0037] The target signal type can be understood as the type of the initial signal that can be used to construct a signal relationship graph. In the target signal identification method provided herein, a signal type set V is predefined, where the signal type set V = {v1, v2, v3, -vn}, where v1, v2, v3, -vn represent different types of target signal types. For example, target signal types include, but are not limited to, fiber break, board offline, circuit group interruption, chassis failure, latency warning, circuit group full, packet loss, and so on.

[0037]

[0038] The signal to be combined can be understood as a signal to be processed that needs to be combined with previously received signals. In the target signal identification method provided in the present disclosure, to avoid reduced processing efficiency caused by excessive received signals, multiple signals of the same type are combined into a single signal, thereby improving processing efficiency.

[0038]

[0039] It should be noted that the initial signal can be understood as a signal set consisting of multiple to-be-processed signals of the same signal type. Since the initial signal is obtained by merging the to-be-merged signal with corresponding historical signals of the same type, the initial signal can be understood as a signal type and can represent the merged multiple to-be-processed signals of the same signal type. Based on this, the multiple initial signals can be understood as multiple signal types, each of which can represent a type of the merged multiple to-be-processed signals of the same signal type.

[0039]

[0040] The historical signal may be understood as a previously received signal to be processed. The similar signal may be understood as a historical signal of the same signal type as the signal to be processed. It should be noted that whenever a new signal to be processed is received, a historical signal of the same signal type as the signal to be processed is determined, and the two are merged into one signal.

[0040]

[0041] Specifically, the target signal identification method provided in the present disclosure can be applied to a signal processing end, which can be understood as a server end that analyzes and processes received signals. For example, the signal processing end may be an operations platform. Based on this, upon receiving multiple signals to be processed from multiple signal providers, the signal processing end first determines the signal type of each signal to be processed and matches the signal type with a preset target signal type. Signals to be processed that have the same signal type as the target signal are then identified as signals to be merged. There may be multiple signals to be merged. Next, based on the signal type, similar signals of the same signal type are identified from historical signals for the multiple signals to be merged. Each signal to be merged is then merged with its corresponding signal of the same type to obtain multiple initial signals. By merging signals to be processed of the same type, processing efficiency is improved, thereby avoiding the reduction in processing efficiency caused by receiving too many signals.

[0041]

[0042] Taking the application of the target signal identification method provided by the present disclosure in a scenario of significant network risk identification as an example, the method of merging the signal to be processed with historical signals to obtain an initial signal is described. The multiple signal providers include log files, Ping tools, and the like; the signal to be processed is an alarm signal; the target signal type is a predefined signal type set V, where V = {v1, v2, v3, ..., vn}; the historical signals are alarm signals received before the alarm signal; and the signal processing end is the server. Based on this, the target signal identification method provided by the present disclosure can be applied to the server, capable of receiving alarm signals sent by Ping tools on the device side and identifying alarm signals from the device side's log files. When an input alarm signal s arrives, the alarm signal s corresponds to a signal type vx. If vx is determined to be within the signal type set V, an alarm signal of the same type as the alarm signal s is determined and converged into one alarm signal. The total number of input alarm signals of the same type received in the same time period is determined as C = Cx + L. If vx is determined not to be within the signal type set V, the alarm signal s is stored locally.

[0042]

[0043] In one or more embodiments provided in the present disclosure, the receiving of multiple signals to be processed sent by the multiple signal providing terminals includes: receiving multiple signals to be divided sent by the multiple signal providing terminals, and determining a signal reception time of each signal to be divided; determining, based on the signal reception time, a preset time range corresponding to each signal to be divided from multiple preset time ranges; and dividing the multiple signals to be divided in the same preset time range into the multiple signals to be processed.

[0043]

[0044] The signal to be divided can be understood as a data signal that needs to be divided and processed, and the signal to be divided can be understood as a fault signal or an alarm signal. By dividing and processing the signal to be divided, a signal to be processed can be obtained. The signal reception time can be understood as the time when the signal to be divided is received.

[0044]

[0045] The preset time range can be understood as a pre-set signal processing time range. To avoid inefficient signal processing due to an excessive number of received signals, the received signals are divided into multiple preset time ranges. Target signal identification is then performed on each of the signals received within each preset time range, thereby improving processing efficiency by avoiding reduced processing efficiency due to an excessive number of received signals.

[0045]

[0046] Continuing with the above example, the preset time range can be understood as a predefined known time interval T. Based on this, the server predefines the known time interval T. When it determines that an alarm signal falls within the time interval T, it constructs a graph set G = {V(G), K(G), E(G)} based on the set of alarm signals received within the time interval T. This avoids processing efficiency reduction caused by excessive signals being received, thereby improving processing efficiency. It should be noted that the length of the preset time interval T is set based on experience and algorithms and can be fixed or dynamically determined. For example, the time interval T can be 1 minute, 10 minutes, etc. Furthermore, the server performs correlation analysis on the alarm signals within each time interval T and constructs an alarm signal relationship graph. Therefore, each time interval T has a corresponding alarm signal relationship graph.

[0046]

[0047] Step 204: Construct a signal relationship graph according to the multiple initial signals and the signal characteristic information of each initial signal.

[0047]

[0048] The signal relationship graph can be understood as a relationship graph constructed based on initial signals. Specifically, see FIG. 3 , which is a schematic diagram of a signal relationship graph in a target signal recognition method provided by an embodiment of the present disclosure. Each circle in FIG. 3 can be understood as an initial signal, or even a node. Initial signals are connected by edges.

[0048]

[0049] In one or more embodiments provided in the present disclosure, constructing a signal relationship graph based on the multiple initial signals and the signal feature information of each initial signal includes: determining the multiple initial signals as multiple relationship graph nodes, and determining the edges of each relationship graph node based on the signal feature information of each initial signal; and constructing a signal relationship graph based on the relationship graph nodes and the edges of each relationship graph node.

[0049]

[0050] Specifically, after determining multiple initial signals and the signal feature information of each initial signal, the server identifies the multiple initial nodes as multiple relationship graph nodes for constructing a signal relationship graph. The server then analyzes the signal feature information of each initial signal to determine whether the relationship graph nodes can be associated. An edge is then determined between two relationship graph nodes that can be associated, thereby obtaining the edges of each relationship graph node. Finally, a signal relationship graph is constructed based on the relationship graph nodes and the edges of each relationship graph node. Based on this, the target signal identification method provided in this disclosure utilizes the computing power of GDB by constructing a signal relationship graph to address the problem of low processing efficiency caused by an excessive number of initial signals. For example, if the initial signal is an alarm signal, GDB computing power can be used to address the alarm flood problem during a fault, facilitate the extraction of effective information, and can also be used to detect high-risk fault scenarios without any scenarios.

[0050]

[0051] Continuing with the previous example, we perform correlation analysis between the received alarm signal s and the historical signals sprevious received within the time interval T to determine whether they can form an edge e. The specific steps are as follows: First, based on the specific keywords or specific logical relationships carried by the alarm signal s, match the signal s with each historical signal sprevious. If the specific keywords or specific logical relationships carried by the signal s match those of any historical signal in the sprevious, the two signals are considered connected, and an edge e is determined for them.

[0051]

[0052] In one or more embodiments provided in the present disclosure, determining the edges of each relationship graph node based on the signal feature information of each initial signal includes: determining a node to be associated and other nodes from the multiple relationship graph nodes, wherein the node to be associated is any one of the multiple relationship graph nodes, and the other nodes are other relationship graph nodes in the multiple relationship graph nodes except the node to be associated; performing feature matching on the signal feature information of the node to be associated with the signal feature information of the other signals using a feature association function to obtain a feature matching result; and constructing an edge between the node to be associated and the other nodes when it is determined that the feature matching is consistent based on the feature matching result.

[0052]

[0053] The feature association function can be understood as a function used to calculate whether the node to be associated and other nodes can be associated. For example, the feature association function is the mapping function Ky(). The feature matching result can be understood as information used to indicate whether two graph nodes can be connected. The feature matching result can take various forms, including but not limited to numerical values, characters, symbols, labels, and other information indicating whether the two graph nodes can be connected. For example, the feature matching result can be a numerical value of 0 or a numerical value of 0; wherein a numerical value of 1 indicates that the two graph nodes can be connected, and a numerical value of 0 indicates that the two graph nodes cannot be connected.

[0053]

[0054] Continuing with the above example, the received alarm signal s is correlated with the historical signal sprevious received within the time interval T, and it is determined whether an edge e can be formed. The specific steps are as follows: Step 1 to Step 4.

[0054]

[0055] 1. The server has a mapping function Ky(s). Based on this mapping function, Ky(s) extracts specific keywords or specific logical relationships carried in alarm signal s. The keywords or logical relationships are conditions for determining whether an alarm signal s of signal type vx is associated with a historical alarm signal of another signal type vy. (If the two types cannot be associated, the mapping function returns null.)

[0055]

[0056] 2. Based on the specific keyword or specific logical relationship, the alarm signal s is matched with each historical signal sprevious to determine whether the two are associated, thereby forming an edge e.

[0057] 3. If the mapping function returns null, it is determined that the alarm signal s is not associated with the historical signal sprevious. Therefore, it is determined that the neighbor signal of signal s in the relationship graph does not exist, and the server stores signal s locally.

[0056]

[0058] 4> If the mapping function returns a non-null value, the signal s is determined to be associated with the historical signal sprevious, forming an edge e, and the edge set E = E + e is updated. For example, if a pair {si, sj} in the input signal corresponds to signal types vx and vy, respectively, and Ky(si) = Kx(sj) exists between them, then {si, sj} is determined to be connected (associated) with each other, and an edge eij is defined for this pair. The edge set E can be understood as the set of all edges in the relationship graph nodes.

[0057]

[0059] After the historical signal sprevious traversal is completed to determine the associations between all alarm signals within the time interval T, the alarm signal types vx and vy at both ends of each edge e in the edge set E are traversed (here x and y may be equal, meaning that a certain signal type can be associated with itself). Furthermore, information about this edge is recorded in the log files corresponding to the two types vx and vy. Based on this, the target signal identification method provided by the present disclosure utilizes a feature correlation function to rapidly determine and construct edges between the node to be associated and other nodes, thereby improving the efficiency of constructing the signal relationship graph.

[0058]

[0060] It should be noted that during the construction of the signal relationship graph, initially, a graph set G = {V(G), K(G), E(G)} can be formed based on the set V and the set K. At this point, the edge set E is initially empty. After the historical signal sprevious traversal is completed, determining the correlation between all alarm signals within the time interval T, the signal relationship graph is constructed based on V(G) and E(G). V(G) can be understood as the set consisting of all initial signals in the signal relationship graph, and K(G) can be understood as a function mapping, or in other words, a relationship function.

[0059]

[0061] Step 206: Determine correlation information of the initial signals in the signal relationship diagram, and determine a signal group to be processed from the signal relationship diagram based on the correlation information.

[0060]

[0062] The signal group to be processed includes a plurality of initial signals having an associated relationship.

[0061]

[0063] The correlation information can be understood as information that determines whether there is a correlation between the initial signals. For example, the correlation information can be the edges between the initial signals, or the correlation information can be understood as the service feature information corresponding to each initial signal. The service feature information can be understood as information that characterizes the application service scenario corresponding to each initial signal. For example, the service feature information can be feature information that characterizes application service scenarios such as network transmission services, hardware services, and storage services. Although different initial signals represent different operating status information of the signal provider, the different initial signals belong to the same application service scenario. For example, network delay alarm information and network interruption alarm information can be alarm information within the application service scenario of network transmission services.

[0062]

[0064] The signal group to be processed can be understood as a set of multiple associated initial signals, a sub-relationship graph consisting of multiple associated initial signals, or a set V' of suspected fault types. For example, the initial signal "fiber break" and the initial signal "packet loss" are connected by an edge determined by the same signal characteristics (e.g., the same geographical location "Province A"). Therefore, the sub-graph formed by connecting "fiber break" and "packet loss" can be defined as the signal group to be processed, or as a set V' of suspected fault types.

[0063]

[0065] In an embodiment provided in the present disclosure, determining the association information of each initial signal in the signal relationship diagram, and determining the group of signals to be processed from the signal relationship diagram based on the association information, includes: determining a first signal and multiple adjacent signals adjacent to the first signal from the initial signals in the signal relationship diagram, wherein the first signal is any initial signal in the signal relationship diagram, and the multiple adjacent signals are initial signals connected to the first signal through edges; determining the edges between the first signal and each adjacent signal, and determining edges with the same characteristics based on signal feature information corresponding to the multiple edges; determining the multiple initial signals connected by the edges with the same characteristics as multiple initial signals with an association relationship, and determining the multiple initial signals with an association relationship as the group of signals to be processed.

[0064]

[0066] Continuing with the above example, after traversing edge set E, the graph set G connects some signal types (i.e., initial signals) due to the presence of edge set E. Therefore, the adjacent nodes (i.e., neighbor nodes) corresponding to each graph node are searched. Each graph node must be directly connected to its corresponding adjacent node by a reachable edge. If an alarm signal for a neighbor node already exists, edges connecting multiple adjacent nodes must be identified, along with the key corresponding to each edge. It should be noted that the edge between each graph node and its corresponding adjacent node is identified by the key (common key) carried by the graph node.

[0065]

[0067] After determining the keywords corresponding to the edges, multiple edges with the same keywords are identified to determine whether the keywords on the edges (relationships) can link multiple graph nodes. If multiple edges with the same keywords exist, they are recorded, and the graph nodes connected by these edges are defined as a signal group to be processed. By grouping multiple initial signals into signal groups for processing, the efficiency of identifying target signals is improved.

[0066]

[0068] In one or more embodiments provided in the present disclosure, determining the association information of each initial signal in the signal relationship diagram and determining the signal group to be processed from the signal relationship diagram based on the association information includes: determining the service feature information of each initial signal in the signal relationship diagram, and determining multiple initial information with the same service feature information as multiple initial signals with an associated relationship; and determining the multiple initial signals with an associated relationship as the signal group to be processed.

[0067]

[0069] Continuing with the above example, Figure 4 is a schematic diagram illustrating signal convergence in a target signal identification method provided by an embodiment of the present disclosure. Based on Figure 4 , it can be seen that the target signal identification method provided by the present disclosure can define three risk scenarios in a graph database: square nodes, triangular nodes, and circular nodes in Figure 4 . Each graph node of each shape represents a risk scenario, and each risk scenario includes several alarm types. For example, the risk scenario can be an application service scenario corresponding to an alarm signal, such as a network transmission service, hardware service, or storage service. Each graph node in Figure 4 represents an alarm signal type. The intersection of two risk scenarios indicates that the graph node satisfies both risk scenario definitions.

[0068]

[0070] Based on this, when the server receives several alarm signals generated in a real environment, it determines whether the signal type of the alarm signal belongs to one of the predefined alarm types in the graph database. It then obtains alarm signals with the same signal type as the predefined alarm types and constructs a signal relationship graph based on the alarm signals. Specifically, the server categorizes the alarm signals generated in the real environment into four square nodes, two triangular nodes, and two circular nodes according to alarm type. After type sorting and relationship matching, the real signal relationship graph determined by signal type (Figure 5) is obtained based on the signal type relationship graph defined in Figure 4. Figure 5 is a schematic diagram of another signal relationship graph in a target signal identification method provided by one embodiment of the present disclosure.

[0069]

[0071] After constructing the signal relationship diagram in Figure 5, convergence calculations are required. Therefore, based on the risk scenario information of the alarm signals in the signal relationship diagram, multiple alarm signals belonging to the same risk scenario are identified as multiple associated alarm signals. These multiple associated alarm signals are then identified as a set of suspected fault types V', or, more accurately, a signal group to be processed. By grouping multiple initial signals into a signal group to be processed, the efficiency of identifying the target signal is improved.

[0070]

[0072] Step 208: When it is determined that the attribute information of the plurality of associated initial signals meets a signal identification condition, the to-be-processed signal group is identified as a target signal.

[0071]

[0073] The attribute information may be understood as information characterizing a specific attribute of the multiple associated initial signals. The attribute information may include the number of the multiple associated initial signals; the attribute information may include the signal type of each of the multiple associated initial signals; or the attribute information may include the number of similar signals corresponding to each of the multiple associated initial signals. This is not intended to limit the scope of this disclosure.

[0072]

[0074] The target signal can be understood as a signal of significant significance or a relatively important signal. For example, the target signal can be a signal of a significant network risk; or the target signal can be a signal of a server downtime that may cause a serious problem.

[0073]

[0075] In one or more embodiments provided by the present disclosure, when determining that the attribute information of the multiple initially signals with an association relationship satisfies the signal recognition condition, identifying the signal group to be processed as a target signal includes: when determining that the attribute information of the multiple network alarm signals with an association relationship satisfies the signal recognition condition, identifying the signal group of the network alarm signals to be processed as a severe network alarm signal.

[0074]

[0076] Continuing with the above example, determine the attribute information of the multiple alarm signals with an association relationship. This attribute information can be the number of signals of the multiple alarm signals with an association relationship, the signal type of the multiple alarm signals with an association relationship, or the number of the same type of signals of each alarm signal among the multiple alarm signals with an association relationship.

[0075]

[0077] When determining that the attribute information of the multiple alarm signals with an association relationship satisfies the signal recognition condition, identify the signal group of the alarm signals to be processed as a severe alarm signal. Specifically, refer to FIG. 5. When performing convergence calculation, the threshold H for the number of nodes in the scene signal connectivity domain is set to 4. As shown in FIG. 5, the risk scene corresponding to the square node has 4 nodes, which meets the condition of being greater than or equal to the number threshold 4. The risk scene corresponding to the square node will report a convergence signal (target signal). Since both the triangular node and the circular node have only 3 nodes, which is less than H, neither of them will report a convergence signal. Thus, a relatively significant severe alarm signal can be accurately identified from multiple alarm signals, facilitating the accurate determination of the current state of the signal providing end based on this severe alarm signal.

[0076]

[0078] It should be noted that when an unknown scene is discovered, if the defined number threshold H is 5, the scene in FIG. 5 above has a total of 8 nodes and is a connected graph. Therefore, an unknown scene convergence signal can be reported. In one or more embodiments provided by the present disclosure, when determining that the attribute information of the multiple initially signals with an association relationship satisfies the signal recognition condition, identifying the signal group to be processed as a target signal includes: when determining that the number of the multiple initially signals with an association relationship is greater than or equal to a preset number threshold, identifying the signal group to be processed as a target signal.

[0079] Continuing with the above example, the number of signal types (initial signals) in each suspected fault type set V' is CV; where the number of signal types CV refers to the number of signal types included in the suspected fault type set V'. For example, a suspected fault type set V' can include 3 or 5 signal types.

[0077]

[0080] Based on this, when CV reaches a preset threshold, the set of suspected fault types V is used as a logical alarm signal S. This logical alarm signal S can be understood as a significant risk signal, representing all alarm signals belonging to the set of suspected fault types V (in other words, the logical alarm signal aggregates all signal types within the set of suspected fault types V) and highlights the alarm signal closest to the cause of the problem. This allows accurate identification of the more important target signal from multiple initial signals, facilitating accurate determination of the current status of the signal provider based on the target signal. The preset threshold can be set based on actual needs; for example, it can be 4, 10, 20, etc.

[0078]

[0081] In one or more embodiments provided by the present disclosure, the method of identifying the group of signals to be processed as target signals when determining that the attribute information of the multiple initial signals with an associated relationship meets a signal identification condition includes: determining the number of similar signals corresponding to each initial signal among the multiple initial signals with an associated relationship, wherein the number of similar signals is the number of similar signals of the initial signal, and the similar signal is a signal of the same signal type as the initial signal; and identifying the group of signals to be processed as target signals when determining that the number of similar signals is greater than or equal to a preset similar signal threshold.

[0079]

[0082] Continuing with the above example, each signal type in each suspected fault type set V has its own number of similar signals, C. C refers to the total number of signal types included in the suspected fault type set V in the entire graph set G. In other words, C refers to the number of multiple pending signals of the same type corresponding to each signal type. Since each signal type is obtained by merging multiple pending signals of the same type, the number of multiple pending signals of the same type corresponding to each signal type can be determined. The maximum value of C for each suspected fault type set V in the number of similar signals, Cmax, is C.

[0080]

[0083] Based on this, when Cmax reaches the preset threshold for similar signals, the set of suspected fault types V' is used as a logical alarm signal S'. This logical alarm signal S' can be understood as a significant risk signal. It represents all alarm signals belonging to the set of suspected fault types V' (in other words, the logical alarm signal S' aggregates all signal types within the set of suspected fault types V') and highlights the alarm signal closest to the cause of the problem. This allows accurate identification of the more important target signal from multiple initial signals, facilitating accurate determination of the current status of the signal provider based on the target signal. The preset threshold for similar signals can be set based on actual needs. For example, the threshold for similar signals can be 4, 10, or so on.

[0081]

[0084] In one or more embodiments provided in the present disclosure, the method of identifying the group of signals to be processed as a target signal when determining that the attribute information of the multiple initial signals with an associated relationship meets a signal identification condition includes: determining the signal type of each initial signal among the multiple initial signals with an associated relationship; and identifying the group of signals to be processed as a target signal when determining that any one of the signal types of the initial signals is consistent with the target signal type.

[0082]

[0085] Specifically, in the target signal identification method provided by the present disclosure, the rule for identifying the signal group to be processed as the target signal may be: once certain specific signal types appear, their weight is sufficient to generate a logical signal. Based on this rule, the signal types of multiple associated initial signals in the signal group to be processed are matched with the target signal type. If any one of the multiple signal types matches the target signal type, the signal group to be processed is identified as the target signal.

[0083]

[0086] Continuing with the above example, in the target signal identification method provided by the present disclosure, the rule for determining whether to generate a logical convergence signal s can be: once certain specific signal types appear, their weight is sufficient to generate a logical signal. Based on this, each signal type in the set of suspected fault types V' is matched with the target signal type. If a match is found, the set of suspected fault types V' is used as the logical alarm signal S'. For example, a downtime alarm signal is a relatively serious alarm signal. Therefore, if a downtime alarm signal is determined to exist in the set of suspected fault types V', the set of suspected fault types V' can be used as the logical alarm signal S'. This allows accurate identification of the more important target signal from multiple initial signals, facilitating accurate determination of the current status of the signal provider based on the target signal. The target signal type can be set based on the actual application scenario. For example, the target signal can be a downtime alarm signal, a network interruption signal, or the like.

[0084]

[0087] In one or more embodiments provided in the present disclosure, identifying the group of signals to be processed as a target signal includes: identifying the group of signals to be processed including the multiple associated initial signals as the target signal; determining a fault signal providing end from the signal providing ends corresponding to the multiple associated initial signals, and using the initial signal sent by the fault signal providing end as the fault signal.

[0085]

[0088] The fault signal provider can be understood as the signal provider closest to the fault cause. In one or more embodiments provided herein, when a device failure occurs due to certain reasons, it may cause multiple signal providers connected to the faulty device to malfunction or experience service interruptions, further leading to the connected signal providers sending a large number of initial signals. Therefore, to accurately identify the actual conditions of multiple signal providers and discover the root cause of the large number of initial signals, it is necessary to identify the faulty signal provider from among the multiple signal providers and use the initial signal sent by the faulty signal provider as the fault signal. This facilitates the subsequent timely and accurate identification of the actual conditions of the faulty device based on the fault signal, facilitates subsequent rapid processing of the faulty device, and reduces the number of initial signals sent by the signal providers. It should be noted that in one or more embodiments provided herein, the faulty device may be a fault information provider.

[0086]

[0089] The fault signal may be understood as a signal sent by the fault signal provider that represents the actual operating status of the faulty device, for example, a signal that represents the actual cause of the fault of the faulty device.

[0087]

[0090] Continuing with the previous example, after converging the set of suspected fault types V' into a logical alarm signal S', we need to identify the fault signal source within the set of suspected fault types V' that is closest to the cause of the problem. The alarm signal sent by this fault information source is then identified as the alarm signal closest to the cause of the problem. For example, a fault event involving a faulty device typically generates multiple alarm signals, originating from different systems or data sources connected to the faulty device, such as logs or pings. Therefore, a particular alarm signal is likely closest to the cause of the fault (i.e., the faulty device). For example: Alarm Signal 1: Link 1 detection failure; Alarm Signal 2: Device a card down, device a connected to Link 1. Therefore, we can conclude that the information provided by Alarm Signal 2 is closest to the fault source (Device a card down).

[0088]

[0091] In one or more embodiments provided herein, the target signal identification method provided herein may also adjust the format of a received signal to improve target signal identification efficiency, thereby adjusting signals of different formats sent by multiple signal providers to signals of the same format, thereby facilitating signal processing. Specifically, the target signal identification method provided herein may adjust the format of multiple initial information, multiple signals to be merged, or multiple signals to be divided, thereby obtaining multiple initial information, multiple signals to be merged, or multiple signals to be divided having the same format. Subsequently, corresponding or corresponding steps in the above-described embodiments are performed on the multiple initial information, multiple signals to be merged, or multiple signals to be divided, and no further details are provided herein.

[0089]

[0092] The target signal identification method provided by the embodiment of the present disclosure utilizes multiple initial signals sent by multiple signal providing terminals and signal feature information of each initial signal to construct a signal relationship diagram, and determines a signal group to be processed that includes multiple initial signals with associated relationships based on the signal relationship diagram. By grouping the multiple initial signals into a signal group to be processed for processing, the efficiency of determining the target signal is improved. Then, when the attribute information of the multiple initial signals with associated relationships is determined and the signal identification conditions are met, the signal group to be processed is identified as a target signal. In this way, a more important target signal can be accurately identified from the multiple initial signals, facilitating accurate determination of the current state of the signal providing terminal based on the target signal.

[0090]

[0093] The target signal identification method provided by the present disclosure will be further described below, using its application in a significant network risk identification scenario as an example, with reference to FIG6 . FIG6 illustrates a flowchart of the target signal identification method according to one embodiment of the present disclosure, specifically including the following steps.

[0091]

[0094] Step 602: Generate an alarm signal.

[0092]

[0095] Specifically, the target signal identification method provided in this disclosure can be applied to a server, capable of receiving alarm signals sent by a Ping tool on a device side and identifying the alarm signals from a log file on the device side. The device side can be understood as an IoT device, client, or server.

[0093]

[0096] Based on this, first, when a certain input alarm signal s arrives, it is determined that the alarm signal s corresponds to a signal type of VX.

[0094]

[0097] Step 604: Determine whether the relationship graph node exists; if so, execute step 606, if not, execute step 618.

[0095]

[0098] Specifically, it is determined whether the signal type vx exists in the signal type set V={v1, v2, v3, -vn).

[0096]

[0099] If not, execute step 618 to store the signal type vx locally.

[0097]

[0100] If so, first, determine the historical alarm signal corresponding to the signal type vx from the signals sprevious received within a predefined known time interval T.

[0098]

[0101] Secondly, the alarm signal s and the historical alarm signal are converged into a signal type s, thereby completing the convergence operation within the alarm node and determining the total number of alarm signals of the same type received in the time interval T: Cx = Cx + L

[0099]

[0102] Next, based on all the alarm signals received within the time interval T and the keywords or specific logical relationships carried by all the alarm signals, a signal relationship graph is constructed. The specific steps are steps 1 to 5 below.

[0100]

[0103] 1. The server has a mapping function Ky(s), based on which specific keywords or specific logical relationships carried by all alarm signals are extracted.

[0101]

[0104] The keyword or logical relationship is the condition for determining whether the signal type vx of the alarm signal is connected to another signal type vy (if the two types are not related, the mapping function returns empty).

[0102]

[0105] 2. Based on the specific keyword or specific logical relationship, each signal type within the time interval T is matched pairwise to determine whether the two can be connected to form an edge e.

[0103]

[0106] 3. If the mapping function returns empty, it is determined that the signal type cannot be connected to other signal types.

[0104]

[0107] 4. If the mapping function returns a non-null value, it is determined that the signal type can be connected to other signal types to form an edge e, and the edge set E = E + e is updated.

[0105]

[0108] For example, a pair {si,sj} in the input signal corresponds to signal types vx and vy, respectively. If Ky(si)=Kx(sj) exists between them, then {si,sj} can be connected to each other, and an edge eij is defined for this pair.

[0106]

[0109] 5. Use the signal type as a signal node and construct a signal relationship graph based on the signal node and the edges between the signal types.

[0107]

[0110] Step 606: Check neighbor alarm signals.

[0108]

[0111] Specifically, using GDB computing capabilities, the edges between signal nodes (i.e., signal types) in the signal relationship graph are queried. Based on these edges, the neighboring alarm signals (i.e., adjacent signal types) of the signal node are determined. A reachable edge directly connects the signal node to its neighboring alarm signals.

[0109]

[0112] Step 608: Determine whether a neighbor signal exists; if so, execute step 610; if not, execute step 618.

[0110]

[0113] Specifically, for a signal node with a neighbor signal, step 610 is performed. For a signal node with no neighbor signal, step 618 is performed.

[0111]

[0114] Step 610: Determine whether convergence is possible; if so, execute step 612; if not, execute step 618.

[0112]

[0115] The specific steps are as follows.

[0113]

[0116] First, if a neighbor node's alarm signal already exists, we need to identify the edges connecting the signaling node and multiple neighboring nodes and determine the keywords corresponding to these edges. It should be noted that the edges between each graph node and its corresponding neighboring node are identified by the keywords (public keys) carried by the graph nodes.

[0114]

[0117] Next, after determining the keyword, it is determined whether there are multiple edges with the same keyword. If so, step 612 is executed to perform a convergence operation; if not, step 618 is executed.

[0115]

[0118] Step 612: Check the connected domain.

[0116]

[0119] Specifically, check the number of signal nodes connected by multiple edges of the same keyword. The number of signal nodes is the connected domain.

[0117]

[0120] Alternatively, the number of neighboring signal nodes of a certain signal node is determined, and the neighboring signal nodes are the connected domain.

[0118]

[0121] Step 614: Determine whether the number of connected nodes reaches a threshold; if so, execute step 616; if not, execute step 618.

[0119]

[0122] The specific steps are as follows: First, define the subgraph consisting of the nodes in the connected domain as a set of suspected fault types V. Second, determine whether the connected nodes have reached a threshold. There are three specific methods.

[0123] The first method is to determine the number of signal types CV in each suspected fault type set V, where CV refers to the number of signal types contained in the suspected fault type set V. For example, a suspected fault type set V may contain 3, 5, or 5 signal types.

[0120]

[0124] When the CV is greater than or equal to a preset number threshold, it is determined that the connected nodes reach the threshold.

[0121]

[0125] When the CV is less than the preset number threshold, it is determined that the connected nodes do not reach the threshold, and step 618 is executed.

[0122]

[0126] The second method is to determine a set of suspected fault types V, where the number C of similar signals for each signal type is determined, with the maximum value of C being Cmax. It should be noted that since signal types are obtained by converging alarm signals of the same type, each signal type has a corresponding number of similar signals, which indicates how many alarm signals of the same type the server has received.

[0123]

[0127] When it is determined that Cmax is greater than or equal to the preset threshold value of the number of similar signals, it is determined that the connected nodes reach the threshold value.

[0124]

[0128] When it is determined that Cmax is less than the preset threshold value of the number of similar signals, it is determined that the number of connected nodes does not reach the threshold, and step 618 is executed.

[0125]

[0129] In the third approach, once certain specific signal types appear, their weight is sufficient to generate a logical signal. Based on this, each signal type in the set of suspected fault types V, is matched with the target signal type.

[0126]

[0130] In the case of consistent matching, it is determined that the connected nodes have reached the threshold.

[0127]

[0131] In the case of inconsistent matching, it is determined that the connected nodes do not reach the threshold, and step 618 is executed.

[0128]

[0132] Step 616: Generate a risk convergence event.

[0129]

[0133] Specifically, first, the suspected fault type set V is converged into a logical alarm signal S, which represents all signals belonging to the suspected fault type set V' (in other words, the logical alarm signal S, aggregates all signals in the suspected fault type set V).

[0130]

[0134] Next, the alarm signal sent by the signal source closest to the cause of the problem in the set of suspected fault types V is identified as the alarm signal closest to the cause of the problem. For example, a fault event typically generates multiple alarm signals, each coming from different systems or data sources, such as logs or pings. Therefore, a particular alarm signal is likely closest to the cause of the fault. For example: Alarm Signal 1: Link 1 detection failure; Alarm Signal 2: Device a board down, connected to Link 1. Therefore, Alarm Signal 2 provides the information closest to the fault.

[0131]

[0135] Step 618: Signal storage.

[0132]

[0136] Specifically, the alarm signal received by the server is stored.

[0133]

[0137] Step 620: End.

[0134]

[0138] The target signal identification method provided in this disclosure provides a method for identifying major network risks. This method focuses on the convergence of multi-source alarm signals. In major fault scenarios, the alarm system is prone to generating a large number of alarm signals from different data sources, leading to an alarm flood. In such cases, it takes a long time to organize the information and extract the signal closest to the root cause.

[0135]

[0139] Based on this, this method leverages graph database capabilities in flood warning scenarios to define a connected graph. Each node represents an alarm signal, and each edge represents a relationship between two alarm signals. Through this relationship (keywords / information), the two alarms can be linked. Different alarm signals are also classified to identify which ones are closest to the root cause. In the case of a flood warning, when an alarm signal is generated, the node in the relationship graph is searched to see if it matches a node. If not, the alarm signal is stored. If a match is found, the node's neighboring nodes (directly connected by reachable edges) are searched. If an alarm signal for a neighboring node already exists, the algorithm checks whether it can be linked using the relationship keywords. If so, the connection is recorded and the node is triggered to check the number of connected nodes. When the number of connected nodes reaches a definable threshold, a high-risk signal with convergence is proactively issued. A signal close to the root cause is also issued, indicating the alarm signal closest to the cause of the problem. This allows different alarm signals to converge based on defined graph relationships and extracts effective information. This achieves the goal of rapidly discovering risks, achieving rapid convergence, and extracting effective information. Furthermore, the high convergence speed solves the risk assessment problem with minimal overhead.

[0136]

[0140] Corresponding to the above method embodiments, the present disclosure further provides an embodiment of a target signal recognition device. FIG7 shows a schematic structural diagram of a target signal recognition device provided by an embodiment of the present disclosure.

[0137]

[0141] As shown in Figure 7, the device includes: an initial signal determination module 702, configured to determine multiple initial signals sent by multiple signal providers, and determine signal feature information of each initial signal; a relationship graph construction module 704, configured to construct a signal relationship graph based on the multiple initial signals and the signal feature information of each initial signal; a signal group determination module 706, configured to determine association information of each initial signal in the signal relationship graph, and based on the association information, determine a signal group to be processed from the signal relationship graph, wherein the signal group to be processed includes multiple initial signals with an associated relationship; a target signal identification module 708, configured to identify the signal group to be processed as a target signal when the attribute information of the multiple initial signals with an associated relationship is determined and a signal identification condition is met.

[0138]

[0142] Optionally, the signal group determination module 706 is further configured to: determine a first signal and multiple adjacent signals adjacent to the first signal from the initial signals in the signal relationship diagram, wherein the first signal is any initial signal in the signal relationship diagram, and the multiple adjacent signals are initial signals connected to the first signal through edges; determine the edges between the first signal and each adjacent signal, and based on the signal feature information corresponding to the multiple edges, determine edges with the same features; determine the multiple initial signals connected by the edges with the same features as multiple initial signals with an associated relationship, and determine the multiple initial signals with an associated relationship as the signal group to be processed.

[0139]

[0143] Optionally, the signal group determination module 706 is further configured to: determine the service feature information of each initial signal in the signal relationship diagram, and determine multiple initial information with the same service feature information as multiple initial signals with associated relationships; and determine the multiple initial signals with associated relationships as the signal group to be processed.

[0140]

[0144] Optionally, the relationship graph construction module 704 is further configured to: determine the multiple initial signals as multiple relationship graph nodes, and determine the edges of each relationship graph node based on the signal feature information of each initial signal; and construct a signal relationship graph based on the relationship graph nodes and the edges of each relationship graph node.

[0141]

[0145] Optionally, the relationship graph construction module 704 is further configured to: determine the node to be associated and other nodes from the multiple relationship graph nodes, wherein the node to be associated is any one of the multiple relationship graph nodes, and the other nodes are other relationship graph nodes in the multiple relationship graph nodes except the node to be associated; perform feature matching on the signal feature information of the node to be associated with the signal feature information of the other signals using a feature association function to obtain a feature matching result; and when it is determined that the feature matching is consistent according to the feature matching result, construct an edge between the node to be associated and the other nodes.

[0142]

[0146] Optionally, the target signal identification module 708 is further configured to: identify the to-be-processed signal group as a target signal when it is determined that the number of the multiple associated initial signals is greater than or equal to a preset number threshold.

[0143]

[0147] Optionally, the target signal identification module 708 is further configured to: determine the number of similar signals corresponding to each initial signal among the multiple associated initial signals, wherein the number of similar signals is the number of similar signals of the initial signal, and the similar signals are signals of the same signal type as the initial signal.

[0144]

[0148] When it is determined that the number of the same type of signals is greater than or equal to a preset same type of signal threshold, the group of signals to be processed is identified as target signals.

[0145]

[0149] Optionally, the target signal identification module 708 is further configured to: determine a signal type of each initial signal among the multiple associated initial signals; and if it is determined that the signal type is consistent with the target signal type, identify the to-be-processed signal group as a target signal.

[0146]

[0150] Optionally, the initial signal determination module 702 is further configured to: receive multiple to-be-processed signals sent by the multiple signal providers; determine a signal type of each to-be-processed signal, and determine a to-be-processed signal whose signal type is consistent with a target signal type as a to-be-merged signal, where there are multiple to-be-merged signals; determine corresponding similar signals for the multiple to-be-merged signals from historical signals according to the signal types of the multiple to-be-merged signals; and merge the multiple to-be-merged signals with corresponding similar historical signals, respectively, to obtain multiple initial signals.

[0147]

[0151] Optionally, the initial signal determination module 702 is further configured to: receive multiple signals to be divided sent by the multiple signal providing terminals, and determine a signal reception time of each signal to be divided; determine a preset time range corresponding to each signal to be divided from multiple preset time ranges based on the signal reception time; and divide the multiple signals to be divided in the same preset time range into the multiple signals to be processed.

[0148]

[0152] Optionally, the target signal identification module 708 is further configured to: if it is determined that the attribute information of the multiple alarm signals having an associated relationship meets a signal identification condition, identify the alarm signal group to be processed as a serious alarm signal.

[0149]

[0153] The target signal identification device provided by the present disclosure utilizes multiple initial signals sent by multiple signal providing terminals and signal feature information of each initial signal to construct a signal relationship graph, and determines a signal group to be processed comprising multiple initial signals with associated relationships based on the signal relationship graph. By grouping the multiple initial signals into a signal group to be processed for processing, the efficiency of determining the target signal is improved. Then, when the attribute information of the multiple initial signals with associated relationships is determined and the signal identification conditions are met, the signal group to be processed is identified as a target signal. This allows accurate identification of a more important target signal from the multiple initial signals, facilitating accurate determination of the current state of the signal providing terminal based on the target signal.

[0150]

[0154] The above is a schematic diagram of a target signal recognition device according to this embodiment. It should be noted that the technical solution of this target signal recognition device and the technical solution of the target signal recognition method described above share the same concept. For details not described in detail in the technical solution of the target signal recognition device, please refer to the description of the technical solution of the target signal recognition method described above.

[0151]

[0155] 8 shows a block diagram of a computing device 800 according to an embodiment of the present disclosure. Components of the computing device 800 include, but are not limited to, a memory 810 and a processor 820. The processor 820 is connected to the memory 810 via a bus 830. A database 850 is used to store data.

[0152]

[0156] Computing device 800 also includes an access device 840 that enables computing device 800 to communicate via one or more networks 860. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. Access device 840 may include one or more of any type of wired or wireless network interface (e.g., a network interface card (NIC)), such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, or a near field communication (NFC) interface.

[0153]

[0157] In one embodiment of the present disclosure, the aforementioned components of computing device 800 and other components not shown in FIG. 8 may also be connected to one another, for example, via a bus. It should be understood that the computing device structure block diagram shown in FIG. 8 is for illustrative purposes only and does not limit the scope of the present disclosure. Those skilled in the art may add or replace other components as needed.

[0154]

[0158] Computing device 800 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, personal digital assistant, laptop computer, notebook computer, netbook computer, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smartwatch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or personal computer (PC). Computing device 800 may also be a mobile or stationary server.

[0155]

[0159] The processor 820 is configured to execute the following computer-executable instructions, which implement the steps of the above-mentioned target signal recognition method when executed by the processor.

[0156]

[0160] The various embodiments of this disclosure are described in a progressive manner. Similar or identical parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the computing device embodiment is generally similar to the target signal identification method embodiment, so its description is relatively simple. For relevant parts, refer to the description of the target signal identification method embodiment.

[0157]

[0161] An embodiment of the present disclosure further provides a computer-readable storage medium storing a computer program / instruction, which implements the steps of the above-mentioned target signal recognition method when executed by a processor.

[0158]

[0162] The various embodiments of this disclosure are described in a progressive manner. Similar or identical parts between the various embodiments may be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the computer-readable storage medium embodiment is generally similar to the target signal identification method embodiment, so its description is relatively simple. For relevant details, refer to the description of the target signal identification method embodiment.

[0159]

[0163] An embodiment of the present disclosure further provides a computer program product, including a computer program / instruction, which implements the steps of the above-mentioned target signal recognition method when executed by a processor.

[0160]

[0164] The above is an illustrative embodiment of a computer program product according to this embodiment. It should be noted that the technical solution of this computer program product shares the same concept as the technical solution of the aforementioned target signal recognition method. For details not described in detail in the technical solution of the computer program product, please refer to the description of the technical solution of the aforementioned target signal recognition method.

[0161]

[0165] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0162]

[0166] The computer instructions include computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signals, telecommunication signals, and software distribution media. It should be noted that the content of the computer-readable medium may be appropriately increased or decreased based on the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0163]

[0167] It should be noted that, for ease of description, the aforementioned method embodiments are described as a series of actions. However, those skilled in the art should be aware that the embodiments of the present disclosure are not limited by the order of the actions described, as certain steps may be performed in other orders or simultaneously according to the embodiments of the present disclosure. Furthermore, those skilled in the art should also be aware that the embodiments described in this specification are preferred embodiments, and the actions and modules involved are not necessarily required for the embodiments of the present disclosure.

[0164]

[0168] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0165]

[0169] The preferred embodiments disclosed above are intended only to illustrate the present disclosure. The alternative embodiments do not exhaustively describe all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of the embodiments disclosed. These embodiments are selected and described in detail to better explain the principles and practical applications of the embodiments, thereby enabling those skilled in the art to better understand and utilize the present disclosure. The present disclosure is limited only by the claims and their full scope and equivalents.

Claims

Claims 1. A target signal recognition method, comprising: determining a plurality of initial signals sent by a plurality of signal providers, and determining signal characteristic information of each initial signal; Constructing a signal relationship graph according to the multiple initial signals and signal characteristic information of each initial signal; Determine association information of each initial signal in the signal relationship diagram, and based on the association information, determine a signal group to be processed from the signal relationship diagram, wherein the signal group to be processed includes multiple initial signals with an associated relationship; and when attribute information of the multiple initial signals with an associated relationship is determined and a signal identification condition is satisfied, identify the signal group to be processed as a target signal.

2. The target signal recognition method according to claim 1, wherein constructing a signal relationship graph based on the multiple initial signals and the signal feature information of each initial signal comprises: Determining the multiple initial signals as multiple relationship graph nodes, and determining an edge of each relationship graph node based on signal feature information of each initial signal; A signal relationship graph is constructed based on the relationship graph nodes and the edges of the relationship graph nodes.

3. The target signal recognition method according to claim 2, wherein determining the edges of each relationship graph node based on the signal feature information of each initial signal comprises: Determine the node to be associated and other nodes from the multiple relationship graph nodes, wherein the node to be associated is any one of the multiple relationship graph nodes, and the other nodes are other relationship graph nodes in the multiple relationship graph nodes except the node to be associated; perform feature matching on the signal feature information of the node to be associated with the signal feature information of the other signals using a feature association function to obtain a feature matching result; and construct an edge between the node to be associated and the other nodes when it is determined that the feature matching is consistent according to the feature matching result.

4. The target signal identification method according to claim 1, wherein determining association information of each initial signal in the signal relationship diagram and determining a signal group to be processed from the signal relationship diagram based on the association information comprises: determining a first signal and a plurality of adjacent signals adjacent to the first signal from each initial signal in the signal relationship diagram, wherein: The first signal is any initial signal in the signal relationship graph, and the multiple adjacent signals are initial signals connected to the first signal through edges; edges between the first signal and each adjacent signal are determined, and based on signal feature information corresponding to the multiple edges, edges with the same features are determined; the multiple initial signals connected by the edges with the same features are determined as multiple initial signals with an associated relationship, and the multiple initial signals with an associated relationship are determined as the signal group to be processed.

5. The target signal identification method according to claim 1, wherein determining association information of each initial signal in the signal relationship graph and determining a signal group to be processed from the signal relationship graph based on the association information comprises: determining service feature information of each initial signal in the signal relationship graph, and determining multiple initial signals having the same service feature information as multiple associated initial signals; The multiple initial signals having an associated relationship are determined as the signal group to be processed.

6. The target signal identification method according to any one of claims 1 to 5, wherein, if it is determined that the attribute information of the plurality of associated initial signals meets a signal identification condition, identifying the to-be-processed signal group as a target signal comprises: When it is determined that the number of the plurality of associated initial signals is greater than or equal to a preset number threshold, the to-be-processed signal group is identified as a target signal.

7. The target signal identification method according to any one of claims 1 to 5, wherein, if it is determined that the attribute information of the plurality of associated initial signals meets a signal identification condition, identifying the to-be-processed signal group as a target signal comprises: Determine the number of similar signals corresponding to each initial signal among the multiple initial signals that have an associated relationship, wherein the number of similar signals is the number of similar signals of the initial signal, and the similar signals are signals of the same signal type as the initial signal; when it is determined that the number of similar signals is greater than or equal to a preset similar signal threshold, identify the signal group to be processed as a target signal.

8. The target signal identification method according to any one of claims 1 to 5, wherein, if it is determined that the attribute information of the plurality of associated initial signals meets a signal identification condition, identifying the to-be-processed signal group as a target signal comprises: determining a signal type of each of the multiple associated initial signals; When it is determined that any one of the signal types of the initial signals is consistent with the target signal type, the to-be-processed signal group is identified as the target signal.

9. The target signal identification method according to claim 1, wherein determining a plurality of initial signals sent by a plurality of signal providers comprises: Receive multiple signals to be processed sent by the multiple signal providers; determine the signal type of each signal to be processed, and determine the signal to be processed whose signal type is consistent with the target signal type as a signal to be merged, wherein there are multiple signals to be merged; according to the signal types of the multiple signals to be merged, determine corresponding similar signals for the multiple signals to be merged from historical signals; use each signal to be merged in the multiple signals to be merged, and the similar signal corresponding to each signal to be merged as an initial signal, to obtain the multiple initial signals.

10. The target signal identification method according to claim 9, wherein the step of receiving a plurality of signals to be processed sent by the plurality of signal providers comprises: receiving a plurality of signals to be divided sent by the plurality of signal providers, and determining a signal receiving time of each signal to be divided; According to the signal reception time, a preset time range corresponding to each of the to-be-divided signals is determined from a plurality of preset time ranges; and a plurality of to-be-divided signals in the same preset time range are divided into the plurality of to-be-processed signals.

11. The target signal identification method according to claim 1, wherein identifying the group of signals to be processed as target signals comprises: Identifying the signal group to be processed, including the plurality of associated initial signals, as the target signal; A fault signal providing end is determined from the signal providing ends corresponding to the multiple associated initial signals, and the initial signal sent by the fault signal providing end is used as a fault signal.

12. The target signal identification method according to claim 1, wherein, if it is determined that the attribute information of the plurality of associated initial signals meets a signal identification condition, identifying the to-be-processed signal group as a target signal comprises: When it is determined that the attribute information of the plurality of associated network alarm signals meets a signal identification condition, the to-be-processed network alarm signal group is identified as a serious network alarm signal.

13. A target signal recognition device, comprising: an initial signal determination module, configured to determine a plurality of initial signals sent by a plurality of signal providers, and determine signal characteristic information of each initial signal; a relationship graph construction module, configured to construct a signal relationship graph according to the multiple initial signals and signal feature information of each initial signal; a signal group determination module configured to determine association information of the initial signals in the signal relationship graph, and determine a signal group to be processed from the signal relationship graph based on the association information, wherein the signal group to be processed includes a plurality of initial signals having an associated relationship; The target signal identification module is configured to identify the group of signals to be processed as target signals when it is determined that the attribute information of the multiple initial signals having an associated relationship meets a signal identification condition.

14. A computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the target signal recognition method according to any one of claims 1 to 12.

15. A computing device comprising: memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer programs / instructions are executed by the processor, the steps of the target signal recognition method according to any one of claims 1 to 12 are implemented.

16. A computer-readable storage medium storing a computer program / instruction, wherein the computer program / instruction, when executed by a processor, implements the steps of the target signal recognition method according to any one of claims 1 to 12.

Images