WO2016198006A1 - 基站故障检测方法及装置 - Google Patents
基站故障检测方法及装置 Download PDFInfo
- Publication number
- WO2016198006A1 WO2016198006A1 PCT/CN2016/086695 CN2016086695W WO2016198006A1 WO 2016198006 A1 WO2016198006 A1 WO 2016198006A1 CN 2016086695 W CN2016086695 W CN 2016086695W WO 2016198006 A1 WO2016198006 A1 WO 2016198006A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- fault
- base station
- cause
- information
- vector
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/08—Testing, supervising or monitoring using real traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/04—Arrangements for maintaining operational condition
Definitions
- This document relates to, but is not limited to, the field of communication technologies, and relates to a method and device for detecting base station faults.
- the embodiment of the invention provides a base station fault detection method and device, which solves the technical problem that the base station fault detection method is not intelligent enough.
- the fault cause corresponding to the fault cause component that satisfies the preset condition in the fault cause vector is used as the base station fault cause.
- the step of acquiring fault information of the base station and converting the fault information into a fault symptom vector includes:
- the acquiring the fault detection matrix corresponding to the fault type according to the fault type of the fault information, performing a matrix operation on the fault symptom vector and the fault detection matrix, and obtaining the fault cause vector includes:
- the method further includes:
- the processing method corresponding to the base station fault cause is obtained according to the base station fault cause, and the processing measure is Sending to a preset display terminal for the display terminal to display the processing measure.
- the fault detection matrix is obtained by weighted average of expert experience fault membership degree and empirical data fault membership degree.
- An embodiment of the present invention further provides a base station fault detection apparatus, where the base station fault detection apparatus includes:
- a first processing module configured to acquire fault information of the base station, and convert the fault information into a fault symptom vector
- a second processing module configured to acquire the fault type according to the fault type of the fault information Corresponding fault detection matrix, performing matrix operation on the fault symptom vector and the fault detection matrix to obtain a fault cause vector;
- the third processing module is configured to use a fault cause corresponding to the fault cause component that satisfies the preset condition in the fault cause vector as a base station fault cause.
- the first processing module includes:
- a processing unit configured to acquire fault information of the base station, and determine a fault type corresponding to the fault information
- the processing unit is configured to acquire a fault information table corresponding to the fault type according to the fault type, and compare the fault information with the fault information table one by one;
- a conversion unit configured to set a corresponding value of the pre-stored fault information matching the fault information in the fault information table as a first preset value, and pre-store the fault information table that does not match the fault information
- the corresponding value of the fault information is set to a second preset value
- the combining unit is configured to combine the first preset value and the second preset value into a fault symptom vector.
- the second processing module includes:
- Obtaining a unit configured to obtain a fault detection matrix corresponding to the determined fault type by using a mapping relationship between the preset fault type and the fault detection matrix;
- the calculating unit is configured to perform a matrix operation on the fault symptom vector and the acquired fault detection matrix to obtain a fault cause vector.
- the base station fault detection apparatus further includes:
- Obtaining a module configured to acquire a processing measure corresponding to the cause of the failure of the base station according to the cause of the base station failure;
- a sending module configured to send the processing measure to the preset display terminal, so that the display terminal displays the processing measure.
- the fault detection matrix is obtained by weighted average of expert experience fault membership degree and empirical data fault membership degree.
- the embodiment of the present invention further provides a computer readable storage medium, where the computer readable storage medium stores computer executable instructions, and when the computer executable instructions are executed, implements a base station fault detection method.
- the method and device for detecting a base station fault first acquires fault information of a base station and converts the fault information into a fault symptom vector, and then acquires a corresponding fault detection matrix according to the fault type of the fault information, and The fault symptom vector is matrix-operated with the fault detection matrix to obtain a fault cause vector, and finally the fault cause corresponding to the fault cause component satisfying the preset condition in the fault cause vector is used as a base station fault cause.
- the foregoing technical solution detects the fault of the base station by means of alarm reporting or log uploading, and manually analyzes and calculates after the detection, and the scheme improves the intelligence of the base station fault detection. Other aspects will be apparent upon reading and understanding the drawings and detailed description.
- FIG. 1 is a schematic flowchart of a method for detecting a base station fault according to Embodiment 1 of the present invention
- FIG. 2 is a schematic flowchart of a preferred embodiment of converting the fault information into a fault symptom vector according to Embodiment 2 of the present invention
- FIG. 3 is a method for acquiring a fault detection matrix corresponding to the fault type according to the fault type of the fault information according to the fault type of the third embodiment of the present invention, and performing a matrix operation on the fault symptom vector and the fault detection matrix to obtain a fault cause vector.
- FIG. 4 is a schematic flowchart of a method for detecting a base station fault according to Embodiment 4 of the present invention.
- FIG. 5 is a schematic diagram of functional blocks of a base station fault detection apparatus according to Embodiment 5 of the present invention.
- FIG. 6 is a schematic diagram of a refinement function module of the first processing module in FIG. 5;
- FIG. 7 is a schematic diagram of a refinement function module of the second processing module in FIG. 5;
- FIG. 8 is a schematic diagram of another functional module of a base station fault detecting apparatus according to Embodiment 5 of the present invention.
- Embodiments of the present invention provide a base station fault detection method.
- FIG. 1 is a schematic flowchart diagram of a method for detecting a base station fault according to Embodiment 1 of the present invention.
- This embodiment provides a base station fault detection method, where the base station fault detection method includes:
- Step S10 Acquire fault information of the base station, and convert the fault information into a fault symptom vector
- the method for acquiring the fault information of the base station includes: collecting the fault information of the base station and acquiring the collected fault information, or collecting the fault information by using a preset fault information collector, and collecting the fault information. And acquiring the fault information collected by the fault information collector.
- the determination occurs or occurs with a certain frequency for a period of time.
- the number of occurrences per unit time can be counted, and the preset threshold value is set, and the statistical probability value is obtained by dividing the number of occurrences by the threshold value, and the range is (0-1).
- Converting the fault information into a fault symptom vector can obtain the fault symptom vector of the fault information by counting the number of occurrences of the fault information in a unit time, and dividing the number of occurrences of the fault information by a preset threshold value.
- the fault feature vector X(x 1 , x 2 , . . . , x m ) is constructed. If x 5 is fault information occurring at a certain frequency, and the statistical probability is 0.9, the corresponding fault symptom vector is (1, 0, 0, 0, 0.9).
- Step S20 Acquire a fault detection matrix corresponding to the fault type according to the fault type of the fault information, perform a matrix operation on the fault symptom vector and the fault detection matrix, and obtain a fault cause vector.
- the fault information table, the rule table, and the conclusion table are preset in advance, and the fault information table includes a fault type, a fault information number, and a fault information feature; the rule table includes a fault type, a fault information number, and a fault.
- the detection matrix includes a fault type, a membership degree, and a fault cause, and further includes a processing measure; the fault information table, the rule table, and the conclusion table are all stored in the fault information base.
- Step S30 the fault cause corresponding to the fault cause component that satisfies the preset condition in the fault cause vector is used as the base station fault cause.
- the fault cause component of the maximum membership degree in the fault cause vector may be selected.
- the cause of the fault is the cause of the base station failure, wherein the fault cause vector includes a plurality of fault cause components.
- the satisfying preset condition may include a maximum membership degree or a minimum membership degree, that is, when the preset condition is the maximum membership degree, the fault cause corresponding to the fault cause component of the maximum membership degree is obtained.
- the preset condition is the minimum membership degree
- the fault cause corresponding to the fault cause component of the minimum membership degree is taken as the base station fault cause.
- the base station fault detection method proposed in this embodiment first acquires fault information of the base station, converts the fault information into a fault symptom vector, and then acquires a fault detection matrix corresponding to the fault information, and the fault symptom vector is The fault detection matrix performs a matrix operation to obtain a fault cause vector, and finally the fault cause corresponding to the fault cause component satisfying the preset condition in the fault cause vector is used as a base station fault cause.
- the foregoing technical solution detects the fault of the base station by means of alarm reporting or log uploading, and manually analyzes and calculates after the detection, and the scheme improves the intelligence of the base station fault detection.
- the base station fault detection method according to the embodiment of the present invention is proposed based on the first embodiment.
- the step S10 includes:
- Step S11 Acquire fault information of the base station, and determine a fault type corresponding to the fault information.
- Step S12 Acquire a fault information table corresponding to the fault type according to the fault type, and compare the fault information with the fault information table one by one;
- Step S13 setting a corresponding value of the pre-stored fault information matching the fault information in the fault information table as a first preset value, and storing pre-stored fault information in the fault information table that does not match the fault information.
- the corresponding value is set to a second preset value
- Step S14 combining the first preset value and the second preset value into a fault symptom vector.
- the determination occurs or occurs with a certain frequency for a period of time.
- the two states can be represented by the occurrence and the non-occurrence, and the occurrence is taken as 1 and 0 is not generated.
- the reasoner is used to convert the fault information into a fault symptom vector
- the manner in which the fault information is converted into a fault symptom vector may be that the reasoner determines whether the fault information occurs, so that the fault symptom vector of the fault information is set to 1 or 0, that is, when the fault information occurs, The reasoner sets the fault symptom vector of the fault information to 1, and when the fault information does not occur, the reasoner sets the fault symptom vector of the fault information to zero.
- the reasoner determines the fault type corresponding to the fault information, and retrieves the fault information table in the fault information base according to the fault type, due to the fault type
- the fault information is included, so that the obtained fault information is compared with the pre-stored fault information table to determine whether the fault information matches the pre-stored fault information table, and the fault information table is
- the corresponding value of the pre-stored fault information matched by the fault information is set to a first preset value
- the corresponding value of the pre-stored fault information in the fault information table that does not match the fault information is set to a second preset value. That is, the first preset value is 1, and the second preset value is 0.
- the first preset value and the second preset value are combined into a fault symptom vector.
- the design principle is that a common fault image cannot uniquely determine the fault type corresponding to the fault type, but must be co-occurred with other faults to confirm the fault.
- a type of fault occurs, that is, when there are two or more fault classes with a common fault phenomenon, if only one fault phenomenon is acquired, and the type of the fault phenomenon cannot be known at this time, then other faults are acquired.
- a phenomenon, and other fault phenomena can distinguish the fault type corresponding to the fault information, and the fault type corresponding to the fault information can be determined at this time.
- the base station fault detection method according to the embodiment of the present invention is proposed based on any of the foregoing embodiments.
- the step S20 includes:
- Step S21 Obtain a fault detection matrix corresponding to the determined fault type by using a mapping relationship between the preset fault type and the fault detection matrix.
- Step S22 performing a matrix operation on the fault symptom vector and the acquired fault detection matrix to obtain a fault cause vector.
- determining the fault type corresponding to the fault type in the preset rule table determining the fault type corresponding to the fault type in the preset rule table, and obtaining the determined fault by using a mapping relationship between the preset fault type and the fault detection matrix.
- a fault detection matrix corresponding to the type that is, in the rule table, the fault type has a corresponding relationship with the fault detection matrix
- the fault detection matrix of the fault symptom vector can be known according to the fault type, and Obtaining the fault detection matrix, and finally performing a matrix operation on the fault symptom vector and the acquired fault detection matrix to obtain a fault cause vector.
- X is the aforementioned fault representation vector
- Y( ⁇ 1 , ⁇ 2 , . . . , ⁇ n ) is the fault cause vector
- ⁇ i is the membership degree of the analysis object corresponding to the fault cause y i , that is, the degree of matching.
- ⁇ It is a fuzzy operator, and the actual application can be simplified to a general matrix operation.
- R is a fuzzy diagnostic matrix that stores the fault membership.
- the fault detection matrix is obtained by weighted average of expert experience fault membership degree and empirical data fault membership degree, that is, obtained by weighted operation of artificial data and empirical data, and the calculation formula is as follows:
- ⁇ 1 represents the expert experience weight
- ⁇ 2 represents the empirical data weight
- ⁇ 1 + ⁇ 2 1.
- S ij is the expert experience fault membership degree
- v ij is the empirical data fault membership degree
- r ij refers to the degree to which the fault representation matches the cause of the fault, and its value range is [0, 1].
- the downlink baseband power is too low
- the downstream digital attenuator is out of the normal range
- the downlink air interface power is too low
- the cause of the fault corresponding to Y is: the baseband signal is abnormal, the downlink is abnormal, and the uplink is abnormal.
- the fault of the current system is the baseband signal abnormality.
- the base station fault detection method further includes:
- step A when the update command of the fault detection matrix is received, the parameter corresponding to the fault detection matrix is displayed by the display terminal, so that the user can modify the parameter displayed by the display terminal; step B, receiving When the parameter completes the instruction, the fault detection matrix is updated according to the changed parameter.
- the rule table in the fault information database may be preset or online feedback update, that is, the user may perform online revision on the rule table after obtaining certain permissions, that is, receiving an update of the fault detection matrix.
- the parameter corresponding to the fault detection matrix is displayed by the display terminal for the user to modify the parameter displayed by the display terminal, where the parameter includes an expert experience weight ⁇ 1 and an empirical data weight ⁇ 2 .
- the expert experience fault membership degree S ij , the empirical data fault membership degree v ij and other parameters when receiving the parameter change completion instruction, update the fault detection matrix according to the changed parameter.
- the base station fault detection method further includes:
- Step S40 after step S30, acquiring processing measures corresponding to the cause of the failure of the base station according to the cause of the base station failure, and transmitting the processing measure to the preset display terminal, so that the display terminal displays the processing measure.
- the processing action corresponding to the base station fault cause is obtained according to the base station fault cause, and the processing measure is sent to the preset display terminal, so that the display terminal displays the
- the processing measure optionally, may also initiate a fault information query request through the human machine interface, after receiving the notification.
- the conclusion table, the rule table, and the fault representation table in the fault information database are successively retrieved, and the corresponding process data is printed out for the user to confirm.
- the embodiment of the invention further provides a base station fault detecting device.
- FIG. 5 is a schematic diagram of functional modules of a base station fault detecting apparatus according to an embodiment of the present invention.
- FIG. 5 is merely an exemplary diagram of an alternative embodiment, and those skilled in the art will surround the functional modules of the base station fault detecting apparatus shown in FIG.
- the function modules of the present application can be easily supplemented; At the heart of the solution is the functionality that each functional module of the custom name has to achieve.
- the embodiment of the present invention provides a base station fault detection apparatus, where the base station fault detection apparatus includes:
- the first processing module 10 is configured to acquire fault information of the base station, and convert the fault information into a fault symptom vector;
- the manner in which the first processing module 10 acquires the fault information of the base station includes: collecting the fault information of the base station, acquiring the collected fault information, or collecting the fault information by using a preset fault information collector. After the fault information is collected, the fault information collected by the fault information collector is obtained.
- the determination occurs or occurs with a certain frequency for a period of time.
- the number of occurrences per unit time can be counted, and the preset threshold value is set, and the statistical probability value is obtained by dividing the number of occurrences by the threshold value, and the range is (0-1).
- Converting the fault information into a fault symptom vector can obtain the fault symptom vector of the fault information by counting the number of occurrences of the fault information in a unit time, and dividing the number of occurrences of the fault information by a preset threshold value.
- the fault feature vector X(x 1 , x 2 , . . . , x m ) is constructed, and if x 5 is fault information occurring at a certain frequency, And the statistical probability is 0.9, then the corresponding fault symptom vector is (1, 0, 0, 0, 0.9).
- the second processing module 20 is configured to acquire a fault detection matrix corresponding to the fault type according to the fault type of the fault information, perform a matrix operation on the fault symptom vector and the fault detection matrix, and obtain a fault cause vector;
- the fault information table, the rule table, and the conclusion table are preset in advance, and the fault information table includes a fault type, a fault information number, and a fault information feature; the rule table includes a fault type, a fault information number, and a fault.
- the detection matrix includes a fault type, a membership degree, and a fault cause, and further includes a processing measure; the fault information table, the rule table, and the conclusion table are all stored in the fault information base.
- the second processing module 20 first acquires the fault type and the fault information number of the fault information in the fault representation table according to the fault information feature of the fault information, and then searches for the fault type and the fault information number according to the fault type and the fault information number.
- the reasoner indexes from the fault information database according to the input fault information number and the fault type to the corresponding rule table, and the second processing module 20 obtains the corresponding fault detection matrix, and then the fault detection matrix.
- a matrix operation is performed with the fault symptom vector to obtain a fault cause vector matrix.
- the third processing module 30 is configured to use a fault cause corresponding to the fault cause component that satisfies the preset condition in the fault cause vector as a base station fault cause.
- the third processing module 30 may select the maximum cause of the fault cause vector according to the fault type and the retrieved preset conclusion table to obtain the fault cause vector in the conclusion table.
- the cause of the fault corresponding to the fault cause component is the cause of the base station fault, wherein the fault cause vector includes a plurality of fault cause components.
- the satisfying the preset condition may include the maximum membership degree or the minimum membership degree, that is, when the preset condition is the maximum membership degree, the third processing module 30 acquires the maximum membership degree.
- the fault cause corresponding to the fault cause component is the base station fault cause.
- the third processing module 30 acquires the fault cause corresponding to the fault cause component of the minimum membership degree as the base station fault cause.
- the base station fault detection apparatus provided in this embodiment first acquires fault information of the base station, converts the fault information into a fault symptom vector, and then acquires a fault detection matrix corresponding to the fault information, and the fault symptom vector and the The fault detection matrix performs matrix operation to obtain a fault cause vector, and finally the fault source corresponding to the fault cause component satisfying the preset condition in the fault cause vector.
- the above technical solution detects the fault of the base station by means of alarm reporting or log uploading, and manually analyzes and calculates it after the detection. This solution improves the intelligence of the base station fault detection.
- the first processing module 10 includes:
- the processing unit 11 is configured to acquire fault information of the base station, and determine a fault type corresponding to the fault information.
- the processing unit 11 is configured to acquire a fault information table corresponding to the fault type according to the fault type, and compare the fault information with the fault information table one by one;
- the converting unit 12 is configured to set a corresponding value of the pre-stored fault information matching the fault information in the fault information table as a first preset value, and the fault information table does not match the fault information.
- the corresponding value of the pre-stored fault information is set to a second preset value;
- the combining unit 13 is configured to combine the first preset value and the second preset value into a fault symptom vector.
- the determination occurs or occurs with a certain frequency for a period of time.
- the two states can be represented by the occurrence and the non-occurrence, and the occurrence is taken as 1 and 0 is not generated.
- the reasoner is used to convert the fault information into a fault symptom vector
- the manner in which the fault information is converted into a fault symptom vector may be that the reasoner determines whether the fault information occurs, so that the fault symptom vector of the fault information is set to 1 or 0, that is, when the fault information occurs, The reasoner sets the fault symptom vector of the fault information to 1, and when the fault information does not occur, the reasoner sets the fault symptom vector of the fault information to zero. That is, after the processing unit 11 obtains the fault information, the preset reasoner is triggered to determine the fault type corresponding to the fault information, and the fault information table in the fault information database is retrieved according to the fault type.
- the faulty type generally includes a plurality of fault information
- the processing unit 11 first compares the obtained fault information with a pre-stored fault information table to determine whether the fault information matches the pre-stored fault information table.
- the converting unit 12 sets a corresponding value of the pre-stored fault information matching the fault information in the fault information table as a first preset value, and does not match the fault information in the fault information table.
- the corresponding value of the pre-stored fault information is set to a second preset value, that is, the first preset value is 1, and the first preset value is 0.
- the combining unit 13 is configured according to the first preset value.
- the second preset value is combined into a fault symptom vector.
- a fault-type fault occurs, that is, when there are two or more fault classes with a common fault phenomenon, if the processing unit 11 only acquires a fault phenomenon, and the type of the fault phenomenon cannot be known at this time, then The processing unit 11 can determine the fault type corresponding to the fault information by acquiring other fault phenomena, and other fault phenomena can distinguish the fault type corresponding to the fault information.
- the second processing module 20 includes:
- the obtaining unit 21 is configured to obtain a fault detection matrix corresponding to the determined fault type by using a mapping relationship between the preset fault type and the fault detection matrix;
- the calculating unit 22 is configured to perform a matrix operation on the fault symptom vector and the acquired fault detection matrix to obtain a fault cause vector.
- the fault type of the fault type is determined according to the fault type of the fault information, and the fault type is determined by a preset fault type and a fault detection matrix. Obtaining a fault detection matrix corresponding to the determined fault type, that is, in the rule table, if there is a corresponding relationship between the fault type and the fault detection matrix, the fault symptom vector can be known according to the fault type.
- the fault detection matrix, and the acquiring unit 21 acquires the fault detection matrix.
- the calculating unit 22 performs a matrix operation on the fault symptom vector and the acquired fault detection matrix to obtain a fault cause vector.
- X is the aforementioned fault representation vector
- Y( ⁇ 1 , ⁇ 2 , . . . , ⁇ n ) is the fault cause vector
- ⁇ i is the membership degree of the analysis object corresponding to the fault cause y i , that is, matching degree.
- ⁇ It is a fuzzy operator, and the actual application can be simplified to a general matrix operation.
- R is a fuzzy diagnostic matrix that stores the fault membership.
- the fault detection matrix is obtained by weighted average of expert experience fault membership degree and empirical data fault membership degree, that is, obtained by weighted operation of artificial data and empirical data, and the calculation formula is as follows:
- i corresponds to the fault representation
- j corresponds to the cause of the fault
- ⁇ 1 represents the expert experience weight
- ⁇ 2 represents the empirical data weight
- ⁇ 1 + ⁇ 2 1.
- S ij is the expert experience fault membership degree
- v ij is the empirical data fault membership degree
- r ij refers to the degree to which the fault representation matches the cause of the fault, and its value range is [0, 1].
- the downlink baseband power is too low
- the downstream digital attenuator is out of the normal range
- the downlink air interface power is too low
- the cause of the fault corresponding to Y is: the baseband signal is abnormal, the downlink is abnormal, and the uplink is abnormal.
- the fault of the current system is the baseband signal abnormality.
- the base station fault detection apparatus further includes: a display module, configured to display, by the display terminal, the fault detection matrix correspondingly when receiving an update instruction of the fault detection matrix And a parameter for modifying, by the user, the parameter displayed by the display terminal; and the updating module is configured to update the fault detection matrix according to the changed parameter when receiving the change completion instruction of the parameter.
- the rule table in the fault information database may be preset or online feedback update, that is, the user may perform online revision on the rule table after obtaining certain permissions, that is, receiving an update of the fault detection matrix.
- the display module displays the parameter corresponding to the fault detection matrix by the display terminal, so that the user can modify the parameter displayed by the display terminal, where the parameter includes the expert experience weight ⁇ 1 , and the empirical data.
- the base station fault detection apparatus further includes:
- the obtaining module 40 is configured to acquire, according to a base station fault reason, a processing measure corresponding to the cause of the base station fault;
- the sending module 50 is configured to send the processing measure to the preset display terminal, so that the display terminal displays the processing measure.
- the acquiring module 40 acquires the processing measure corresponding to the base station fault cause according to the base station fault reason, and the sending module 50 sends the processing measure to the preset display terminal.
- the fault information query request may also be initiated through the human machine interface, after receiving the notification.
- the conclusion table, the rule table, and the fault representation table in the fault information database are successively retrieved, and the corresponding process data is printed out for the user to confirm.
- the embodiment of the present invention further provides a computer readable storage medium, where the computer readable storage medium stores computer executable instructions, and when the computer executable instructions are executed, implements a base station fault detection method.
- the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is better.
- Implementation Based on such understanding, the technical solution of the present application, which is essential or contributes to the related art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk, CD-ROM).
- the instructions include a number of instructions for causing a terminal device (which may be a cell phone, computer, server, air conditioner, or network device, etc.) to perform the methods described in each embodiment of the present application.
- each module/unit in the above embodiment may be implemented in the form of hardware, for example, by implementing an integrated circuit to implement its corresponding function, or may be implemented in the form of a software function module, for example, executing a program stored in the memory by a processor. / instruction to achieve its corresponding function.
- This application is not limited to any specific combination of hardware and software.
- the above technical solution improves the intelligence of base station fault detection.
Abstract
一种基站故障检测方法，所述方法包括：获取基站的故障信息，将所述故障信息转化为故障征兆向量；根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量；将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。本发明实施例的技术方案提高了基站故障检测的智能性。
Description
本文涉及但不限于通信技术领域，涉及一种基站故障检测方法及装置。
随着通信技术的快速发展，通信领域的技术随之发展，现在大部分的无线通信都是通过基站进行无线信号的传递，因此，基站在无线通信中的作用越来越大，而通常在基站出现故障时，由于基站的复杂性以及故障出现的随机性等等，对基站的检测通常都是通过故障定位方式，如告警上报、黑盒子记录、日志上传等方式进行检测的，而这些检测方式往往只能获取片面局部的信息，即使专业人员参与进来也多需要大量繁复的分析过程，导致基站故障的不够智能。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本发明实施例提出一种基站故障检测方法及装置，解决了基站故障检测的方式不够智能的技术问题。
本发明实施例提供的一种基站故障检测方法，所述基站故障检测方法包括以下步骤：
获取基站的故障信息，将所述故障信息转化为故障征兆向量；
根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量；
将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。
可选地，所述获取基站的故障信息，将所述故障信息转化为故障征兆向量的步骤包括：
获取基站的故障信息，确定所述故障信息对应的故障种类；
根据所述故障种类获取所述故障种类对应的故障信息表，将所述故障信息与所述故障信息表进行一一比对；
将所述故障信息表中与所述故障信息匹配的预存故障信息的对应值设为第一预设值，并将所述故障信息表中与所述故障信息不匹配的预存故障信息的对应值设为第二预设值；
根据所述第一预设值和所述第二预设值组合成故障征兆向量。
可选地，所述根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量的步骤包括：
通过预设的故障种类与故障检测矩阵的映射关系，获取确定的故障种类对应的故障检测矩阵；
对所述故障征兆向量与获取的所述故障检测矩阵进行矩阵运算，得到故障原因向量。
可选地，所述方法还包括：
所述将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因的步骤之后，根据基站故障原因获取所述基站故障原因对应的处理措施，并将所述处理措施发送给预设的显示终端，以供所述显示终端显示所述处理措施。
可选地，所述故障检测矩阵由专家经验故障隶属度和经验数据故障隶属度加权平均得到。
本发明实施例还提供一种基站故障检测装置，所述基站故障检测装置包括：
第一处理模块，设置为获取基站的故障信息，将所述故障信息转化为故障征兆向量；
第二处理模块，设置为根据所述故障信息的故障种类获取所述故障种类
对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量；
第三处理模块，设置为将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。
可选地，所述第一处理模块包括：
处理单元，设置为获取基站的故障信息，确定所述故障信息对应的故障种类；
所述处理单元，设置为根据所述故障种类获取所述故障种类对应的故障信息表，将所述故障信息与所述故障信息表进行一一比对；
转化单元，设置为将所述故障信息表中与所述故障信息匹配的预存故障信息的对应值设为第一预设值，并将所述故障信息表中与所述故障信息不匹配的预存故障信息的对应值设为第二预设值；
组合单元，设置为根据所述第一预设值和所述第二预设值组合成故障征兆向量。
可选地，所述第二处理模块包括：
获取单元，设置为通过预设的故障种类与故障检测矩阵的映射关系，获取确定的故障种类对应的故障检测矩阵；
计算单元，设置为对所述故障征兆向量与获取的所述故障检测矩阵进行矩阵运算，得到故障原因向量。
可选地，所述基站故障检测装置还包括：
获取模块，设置为根据基站故障原因获取所述基站故障原因对应的处理措施；
发送模块，设置为将所述处理措施发送给预设的显示终端，以供所述显示终端显示所述处理措施。
可选地，所述故障检测矩阵由专家经验故障隶属度和经验数据故障隶属度加权平均得到。
本发明实施例还提供一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机可执行指令，所述计算机可执行指令被执行时实现基站故障检测方法。
本发明实施例提出的基站故障检测方法及装置，先获取基站的故障信息并将所述故障信息转化为故障征兆向量，然后根据所述故障信息的故障种类获取对应的故障检测矩阵，并将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量，最后将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。上述技术方案通过告警上报或日志上传等方式对基站的故障进行检测，并在检测后还要人工进行分析与计算，本方案提高了基站故障检测的智能性。在阅读并理解了附图和详细描述后，可以明白其它方面。
图1为本发明实施例一的基站故障检测方法的流程示意图；
图2为本发明实施例二的将所述故障信息转化为故障征兆向量较佳实施例的流程示意图；
图3为本发明实施例三的根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量较佳实施例的流程示意图；
图4为本发明实施例四的基站故障检测方法的流程示意图；
图5为本发明实施例五的基站故障检测装置的功能模块示意图；
图6为图5中第一处理模块的细化功能模块示意图；
图7为图5中第二处理模块的细化功能模块示意图；
图8为本发明实施例五的基站故障检测装置的另一功能模块示意图。
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限
定本申请。
实施例一
本发明实施例提供一种基站故障检测方法。
参照图1，图1为本发明实施例一的基站故障检测方法的流程示意图。
本实施例提出一种基站故障检测方法，所述基站故障检测方法包括：
步骤S10，获取基站的故障信息，将所述故障信息转化为故障征兆向量；
在本实施例中，所述获取基站的故障信息的方式包括：采集基站的故障信息并获取采集的所述故障信息，或者是通过预设的故障信息采集器采集故障信息，并在故障信息采集后，获取所述故障信息采集器采集的所述故障信息。
由于在基站中，故障信息的采集和发生存在两种情况：确定发生或在一段时间内以一定频度发生。对于以一定频度发生的故障信息，可统计单位时间内的发生次数，再设定预设的门限值，用发生次数除以门限值获得统计概率值，范围为(0-1)，即将所述故障信息转化为故障征兆向量可通过统计单位时间内所述故障信息的发生次数，再用所述故障信息的发生次数除以预设地门限值得到所述故障信息的故障征兆向量，例如，当前基站获取的故障信息存在m个，则构建故障特征向量X(x1,x2,...,xm)，如果x5为一定频度发生的故障信息，且统计概率为0.9，则对应的故障征兆向量为(1,0,0,0,0.9)。
步骤S20，根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量；
在本实施例中，会事先预置故障信息表、规则表以及结论表，所述故障信息表包括故障种类、故障信息号和故障信息特征；所述规则表包括故障种类、故障信息号，故障检测矩阵；所述结论表则包括故障种类、隶属度和故障原因，进一步还包括处理措施；所述故障信息表，规则表以及结论表都存储在故障信息库中。根据所述故障信息的故障信息特征先获取所述故障信息在所述故障表象表中的故障种类以及故障信息号，然后根据所述故障种类以及所述故障信息号寻找对应的规则表，并获取所述规则表中所述故障种类对应的故障检测矩阵，最后将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量。即所述推理器从故障信息库中根据输入的故障信
息号以及故障种类，索引到对应的规则表，获取到对应的故障检测矩阵后，再将所述故障检测矩阵同故障征兆向量进行矩阵运算，以得到故障原因向量矩阵。
步骤S30，将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。
在本实施例中，根据故障种类和检索预设的结论表，以获取所述结论表中的所述故障原因向量后，可选将所述故障原因向量中最大隶属度的故障原因分量对应的故障原因作为基站故障原因，其中，所述故障原因向量中包括多个故障原因分量。
在本实施例中，所述满足预设条件可选包括最大隶属度或是最小隶属度，即所述预设条件为最大的隶属度时，则获取最大隶属度的故障原因分量对应的故障原因作为基站故障原因，所述预设条件为最小的隶属度时，则获取最小隶属度的故障原因分量对应的故障原因作为基站故障原因。
本实施例提出的基站故障检测方法，先获取基站的故障信息，将所述故障信息转化为故障征兆向量，然后获取所述故障信息对应的故障检测矩阵，并将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量，最后将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。上述技术方案通过告警上报或日志上传等方式对基站的故障进行检测，并在检测后还要人工进行分析与计算，本方案提高了基站故障检测的智能性。
实施例二
进一步地，为了提高基站故障检测的灵活性，基于实施例一提出本发明实施例的基站故障检测方法，在本实施例中，参照图2，所述步骤S10包括：
步骤S11，获取基站的故障信息，确定所述故障信息对应的故障种类；
步骤S12，根据所述故障种类获取所述故障种类对应的故障信息表，将所述故障信息与所述故障信息表进行一一比对；
步骤S13，将所述故障信息表中与所述故障信息匹配的预存故障信息的对应值设为第一预设值，并将所述故障信息表中与所述故障信息不匹配的预存故障信息的对应值设为第二预设值；
步骤S14，根据所述第一预设值和所述第二预设值组合成故障征兆向量。
由于在基站中，故障信息的采集和发生存在两种情况：确定发生或在一段时间内以一定频度发生。对于确定发生的故障信息可以用发生和未发生两种状态表示，发生取为1，未发生取0，本实施例中，所述推理器用于将所述故障信息转化为故障征兆向量，而将所述故障信息转化为故障征兆向量的方式可以为所述推理器确定故障信息是否发生，从而将所述故障信息的故障征兆向量置于为1或0，即当所述故障信息发生时，所述推理器将所述故障信息的故障征兆向量置于为1，当所述故障信息不发生时，所述推理器将所述故障信息的故障征兆向量置于为0。
也就是说，在获取到故障信息后，先触发预设的推理器，推理器确定所述故障信息对应的故障种类，并根据所述故障种类检索故障信息库中的故障信息表，由于故障种类一般包括多个故障信息，因此先将获取到的所述故障信息与预存的故障信息表进行比对，以确定所述故障信息，是否与预存故障信息表匹配，将所述故障信息表中与所述故障信息匹配的预存故障信息的对应值设为第一预设值，并将所述故障信息表中与所述故障信息不匹配的预存故障信息的对应值设为第二预设值，即第一预设值为1，第二预设值为0，最后根据所述第一预设值和所述第二预设值组合成故障征兆向量。为更好理解本实施例，举例如下：以射频拉远单元下行功率类故障为例，假定当前基站设备存在m个故障现象同这类故障相关，则需要构建故障表象向量X(x1,x2,...,xm)，假定m＝5，x1,x5对应故障现象确定发生，则对应的向量为(1,0,0,0,1)。同时，判断是否将所有的故障信息处理完，如果没有处理完，继续检索下一个故障信息。
进一步地，对于两个或多个故障类之间可以存在共同的故障现象，设计原则为：共同的故障表象不能唯一确定其所对应故障类故障发生，而必须同其它故障共同发生才能确认该故障类故障发生，即当有两个或多个故障类存在共同的故障现象时，如果仅仅获取到一个故障现象，此时是无法得知所述故障现象的类别的，那么要通过获取其他的故障现象，且其它的故障现象可以区分出所述故障信息对应的故障种类，此时才可确定所述故障信息对应的故障种类。
实施例三
进一步地，为了提高基站故障检测的灵活性，基于前述任一实施例提出本发明实施例的基站故障检测方法，在本实施例中，参照图3，所述步骤S20包括：
步骤S21，通过预设的故障种类与故障检测矩阵的映射关系，获取确定的故障种类对应的故障检测矩阵；
步骤S22，对所述故障征兆向量与获取的所述故障检测矩阵进行矩阵运算，得到故障原因向量。
在本实施例中，根据所述故障信息的故障种类，确定所述故障种类在预设的规则表中对应的故障种类，通过预设的故障种类与故障检测矩阵的映射关系，获取确定的故障种类对应的故障检测矩阵，也就是说在所述规则表中，故障种类与故障检测矩阵之间有对应关系的，根据所述故障种类即可得知所述故障征兆向量的故障检测矩阵，并获取所述故障检测矩阵，最后，将所述故障征兆向量与获取的所述故障检测矩阵进行矩阵运算，得到故障原因向量。为更好理解本实施例。举例如下：公式表示如下：
Y＝X·R
式中，X是前述的故障表象向量，Y(θ1,θ2,...,θn)是故障原因向量，θi是分析对象对应故障原因yi的隶属度，即匹配程度。·是模糊算子，实际应用可简化为一般的矩阵运算。
R是模糊诊断矩阵，存储的是故障隶属度。在本实施例中，所述故障检测矩阵由专家经验故障隶属度和经验数据故障隶属度加权平均得到，即通过人工数据及经验数据的加权运算获得，计算公式如下：
rij＝ω1Sij+ω2vij
式中，i对应故障表象，j对应故障原因。ω1代表专家经验权重，ω2代表经验数据权重，ω1+ω2＝1。Sij为专家经验故障隶属度，vij为经验数据故障隶属度。rij指故障表象同故障原因匹配的程度，其取值范围为[0,1]。
举例来说明这个推理过程：
故障表象输入向量为X＝(1,0,0,0,1)，其中：
x1,下行基带功率过低；
x2，下行数控衰减器超出正常范围；
x3，数模转换器状态异常；
x4，上行底噪值过大；
x5，下行空口功率过低；
由前述可知故障检测矩阵是一个5×3的矩阵，经过前述矩阵运算，及归一化后例如得到：Y＝(1,0.5,0.1)
那么，由于Y所对应的故障原因为：基带信号异常，下行链路异常，上行链路异常，而此时若是根据最大隶属度原则，可知当前系统的故障原因为基带信号异常。
可选地，为了提高基站故障检测的灵活性，所述基站故障检测方法还包括：
步骤A，在接收到故障检测矩阵的更新指令时，通过所述显示终端显示所述故障检测矩阵对应的参数，以供用户对所述显示终端显示的所述参数进行修改；步骤B，在接收到参数的更改完成指令时，根据更改后的所述参数更新所述故障检测矩阵。
在本实施例中，所述故障信息库中的规则表，可以预置也可以在线反馈更新，即用户可以在获取一定权限后，对规则表进行在线修订，即在接收到故障检测矩阵的更新指令时，通过所述显示终端显示所述故障检测矩阵对应的参数，以供用户对所述显示终端显示的所述参数进行修改，所述参数包括专家经验权重ω1，经验数据权重ω2，专家经验故障隶属度Sij，经验数据故障隶属度vij等参数，在接收到参数的更改完成指令时，根据更改后的所述参数更新所述故障检测矩阵。通过设定相应权限，以满足不同用户的需求，且增加系统的安全程度。
实施例四
进一步地，为了提高基站故障检测的灵活性，基于前述任一实施例提出本发明实施例的基站故障检测方法，在本实施例中，参照图4，所述基站故障检测方法还包括：
步骤S40，步骤S30之后，根据基站故障原因获取所述基站故障原因对应的处理措施，并将所述处理措施发送给预设的显示终端，以供所述显示终端显示所述处理措施。
在本实施例中，得到故障原因向量后，根据基站故障原因获取所述基站故障原因对应的处理措施，并将所述处理措施发送给预设的显示终端，以供所述显示终端显示所述处理措施，可选地，还可通过人机接口发起故障信息查询请求，在收到通知后。逐次检索故障信息库中的结论表、规则表、故障表象表，将相应的过程数据打印出来，供用户进行确认。
实施例五
本发明实施例还提供一种基站故障检测装置。
参照图5，图5为本发明实施例的基站故障检测装置的功能模块示意图。
需要强调的是，对本领域的技术人员来说，图5所示功能模块图仅仅是一个可选实施例的示例图，本领域的技术人员围绕图5所示的基站故障检测装置的功能模块，可轻易进行新的功能模块的补充；每个功能模块的名称是自定义名称，仅用于辅助理解该基站故障检测装置的每个程序功能块，不用于限定本申请的技术方案，本申请技术方案的核心是，每个自定义名称的功能模块所要达成的功能。
本实施例提出一种基站故障检测装置，所述基站故障检测装置包括：
第一处理模块10，设置为获取基站的故障信息，将所述故障信息转化为故障征兆向量；
在本实施例中，所述第一处理模块10获取基站的故障信息的方式包括：采集基站的故障信息并获取采集的所述故障信息，或者是通过预设的故障信息采集器采集故障信息，并在故障信息采集后，获取所述故障信息采集器采集的所述故障信息。
由于在基站中，故障信息的采集和发生存在两种情况：确定发生或在一段时间内以一定频度发生。对于以一定频度发生的故障信息，可统计单位时间内的发生次数，再设定预设的门限值，用发生次数除以门限值获得统计概率值，范围为(0-1)，即将所述故障信息转化为故障征兆向量可通过统计单位时间内所述故障信息的发生次数，再用所述故障信息的发生次数除以预设地门限值得到所述故障信息的故障征兆向量，例如，所述第一处理模块10获取的故障信息存在m个，则构建故障特征向量X(x1,x2,...,xm)，如果x5为一定频度发生的故障信息，且统计概率为0.9，则对应的故障征兆向量为(1,0,0,0,0.9)。
第二处理模块20，设置为根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量；
在本实施例中，会事先预置故障信息表、规则表以及结论表，所述故障信息表包括故障种类、故障信息号和故障信息特征；所述规则表包括故障种类、故障信息号，故障检测矩阵；所述结论表则包括故障种类、隶属度和故障原因，进一步还包括处理措施；所述故障信息表，规则表以及结论表都存储在故障信息库中。所述第二处理模块20根据所述故障信息的故障信息特征先获取所述故障信息在所述故障表象表中的故障种类以及故障信息号，然后根据所述故障种类以及所述故障信息号寻找对应的规则表，并获取所述规则表中所述故障种类对应的故障检测矩阵，最后将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量。即所述推理器从故障信息库中根据输入的故障信息号以及故障种类，索引到对应的规则表，所述第二处理模块20获取到对应的故障检测矩阵后，再将所述故障检测矩阵同故障征兆向量进行矩阵运算，以得到故障原因向量矩阵。
第三处理模块30，设置为将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。
在本实施例中，所述第三处理模块30根据故障种类和检索预设的结论表，以获取所述结论表中的所述故障原因向量后，可选将所述故障原因向量中最大隶属度的故障原因分量对应的故障原因作为基站故障原因，其中，所述故障原因向量中包括多个故障原因分量。
在本实施例中，所述满足预设条件可选包括最大隶属度或是最小隶属度，即所述预设条件为最大的隶属度时，则所述第三处理模块30获取最大隶属度的故障原因分量对应的故障原因作为基站故障原因，所述预设条件为最小的隶属度时，则第三处理模块30获取最小隶属度的故障原因分量对应的故障原因作为基站故障原因。
本实施例提出的基站故障检测装置，先获取基站的故障信息，将所述故障信息转化为故障征兆向量，然后获取所述故障信息对应的故障检测矩阵，并将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量，最后将所述故障原因向量中满足预设条件的故障原因分量对应的故障原
因作为基站故障原因，上述技术方案通过告警上报或日志上传等方式对基站的故障进行检测，并在检测后还要人工进行分析与计算，本方案提高了基站故障检测的智能性。
可选地，为了提高基站故障检测的灵活性，在本实施例中，参照图6，所述第一处理模块10包括：
处理单元11，设置为获取基站的故障信息，确定所述故障信息对应的故障种类；
所述处理单元11，设置为根据所述故障种类获取所述故障种类对应的故障信息表，将所述故障信息与所述故障信息表进行一一比对；
转化单元12，设置为将所述故障信息表中与所述故障信息匹配的预存故障信息的对应值设为第一预设值，并将所述故障信息表中与所述故障信息不匹配的预存故障信息的对应值设为第二预设值；
组合单元13，设置为根据所述第一预设值和所述第二预设值组合成故障征兆向量。
由于在基站中，故障信息的采集和发生存在两种情况：确定发生或在一段时间内以一定频度发生。对于确定发生的故障信息可以用发生和未发生两种状态表示，发生取为1，未发生取0，本实施例中，所述推理器用于将所述故障信息转化为故障征兆向量，而将所述故障信息转化为故障征兆向量的方式可以为所述推理器确定故障信息是否发生，从而将所述故障信息的故障征兆向量置于为1或0，即当所述故障信息发生时，所述推理器将所述故障信息的故障征兆向量置于为1，当所述故障信息不发生时，所述推理器将所述故障信息的故障征兆向量置于为0。也就是说，在所述处理单元11获取到故障信息后，先触发预设的推理器，确定所述故障信息对应的故障种类，并根据所述故障种类检索故障信息库中的故障信息表，由于故障种类一般包括多个故障信息，因此所述处理单元11先将获取到的所述故障信息与预存的故障信息表进行比对，以确定所述故障信息，是否与预存故障信息表匹配，并且所述转化单元12将所述故障信息表中与所述故障信息匹配的预存故障信息的对应值设为第一预设值，并将所述故障信息表中与所述故障信息不匹配的预存故障信息的对应值设为第二预设值，即所述第一预设值为1，所述第一
预设值为0，最后所述组合单元13根据所述第一预设值和所述第二预设值组合成故障征兆向量。为更好理解本实施例，举例如下：以射频拉远单元下行功率类故障为例，假定当前基站设备存在m个故障现象同这类故障相关，则需要构建故障表象向量X(x1,x2,...,xm)，假定m＝5，x1,x5对应故障现象确定发生，则对应的向量为(1,0,0,0,1)。同时，判断是否将所有的故障信息处理完，如果没有处理完，继续检索下一个故障信息。
进一步地，对于两个或多个故障类之间可以存在共同的故障现象，但设计原则为：共同的故障表象不能唯一确定其所对应故障类故障发生，而必须同其它故障共同发生才能确认该故障类故障发生，即当有两个或多个故障类存在共同的故障现象时，如果所述处理单元11仅仅获取到一个故障现象，此时是无法得知所述故障现象的类别的，那么所述处理单元11要通过获取其他的故障现象，且其它的故障现象可以区分出所述故障信息对应的故障种类，此时才可确定所述故障信息对应的故障种类。
可选地，为了提高基站故障检测的灵活性，在本实施例中，参照图7，所述第二处理模块20包括：
获取单元21，设置为通过预设的故障种类与故障检测矩阵的映射关系，获取确定的故障种类对应的故障检测矩阵；
计算单元22，设置为对所述故障征兆向量与获取的所述故障检测矩阵进行矩阵运算，得到故障原因向量。
在本实施例中，根据所述故障信息的故障种类，确定所述故障种类在预设的规则表中对应的故障种类，通过预设的故障种类与故障检测矩阵的映射关系，所述获取单元21获取确定的故障种类对应的故障检测矩阵，也就是说在所述规则表中，故障种类与故障检测矩阵之间有对应关系的，根据所述故障种类即可得知所述故障征兆向量的故障检测矩阵，并且所述获取单元21获取所述故障检测矩阵，最后，所述计算单元22将所述故障征兆向量与获取的所述故障检测矩阵进行矩阵运算，得到故障原因向量。为更好理解本实施例，举例如下：公式表示如下：
Y＝X·R (1)
(1)式中，X是前述的故障表象向量，Y(θ1,θ2,...,θn)是故障原因向量，
θi是分析对象对应故障原因yi的隶属度，即匹配程度。·是模糊算子，实际应用可简化为一般的矩阵运算。
R是模糊诊断矩阵，存储的是故障隶属度。在本实施例中，所述故障检测矩阵由专家经验故障隶属度和经验数据故障隶属度加权平均得到，即通过人工数据及经验数据的加权运算获得，计算公式如下：
rij＝ω1Sij+ω2vij (2)
(2)式中，i对应故障表象，j对应故障原因。ω1代表专家经验权重，ω2代表经验数据权重，ω1+ω2＝1。Sij为专家经验故障隶属度，vij为经验数据故障隶属度。rij指故障表象同故障原因匹配的程度，其取值范围为[0,1]。
举例来说明这个推理过程：
故障表象输入向量为X＝(1,0,0,0,1)，其中：
x1,下行基带功率过低；
x2，下行数控衰减器超出正常范围；
x3，数模转换器状态异常；
x4，上行底噪值过大；
x5，下行空口功率过低；
由前述可知故障检测矩阵是一个5×3的矩阵，经过前述矩阵运算，及归一化后例如得到：Y＝(1,0.5,0.1)
那么，由于Y所对应的故障原因为：基带信号异常，下行链路异常，上行链路异常，而此时若是根据最大隶属度原则，可知当前系统的故障原因为基带信号异常。
可选地，为了提高基站故障检测的灵活性，所述基站故障检测装置还包括：显示模块，设置为在接收到故障检测矩阵的更新指令时，通过所述显示终端显示所述故障检测矩阵对应的参数，以供用户对所述显示终端显示的所述参数进行修改；更新模块，设置为在接收到参数的更改完成指令时，根据更改后的所述参数更新所述故障检测矩阵。
在本实施例中，所述故障信息库中的规则表，可以预置也可以在线反馈更新，即用户可以在获取一定权限后，对规则表进行在线修订，即在接收到故障检测矩阵的更新指令时，所述显示模块通过所述显示终端显示所述故障检测矩阵对应的参数，以供用户对所述显示终端显示的所述参数进行修改，
所述参数包括专家经验权重ω1，经验数据权重ω2，专家经验故障隶属度Sij，经验数据故障隶属度vij等参数，在接收到参数的更改完成指令时，所述更新模块根据更改后的所述参数更新所述故障检测矩阵。通过设定相应权限，以满足不同用户的需求，且增加系统的安全程度。
可选地，为了提高基站故障检测的灵活性，在本实施例中，参照图8，所述基站故障检测装置还包括：
获取模块40，设置为根据基站故障原因获取所述基站故障原因对应的处理措施；
发送模块50，设置为将所述处理措施发送给预设的显示终端，以供所述显示终端显示所述处理措施。
在本实施例中，得到故障原因向量后，所述获取模块40根据基站故障原因获取所述基站故障原因对应的处理措施，并且所述发送模块50将所述处理措施发送给预设的显示终端，以供所述显示终端显示所述处理措施，可选地，还可通过人机接口发起故障信息查询请求，在收到通知后。逐次检索故障信息库中的结论表、规则表、故障表象表，将相应的过程数据打印出来，供用户进行确认。
本发明实施例还提供一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机可执行指令，所述计算机可执行指令被执行时实现基站故障检测方法。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其它变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其它要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。
上述本发明实施例序号仅仅为了描述，不代表实施例的优劣。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对相关技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请每个实施例所述的方法。
以上仅为本申请的可选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其它相关的技术领域，均同理包括在本申请的专利保护范围内。
可选地，上述实施例的全部或部分步骤也可以使用一个或多个集成电路来实现。相应地，上述实施例中的各模块/单元可以采用硬件的形式实现，例如通过集成电路来实现其相应功能，也可以采用软件功能模块的形式实现，例如通过处理器执行存储于存储器中的程序/指令来实现其相应功能。本申请不限制于任何特定形式的硬件和软件的结合。
上述技术方案提高了基站故障检测的智能性。
Claims (10)
- 一种基站故障检测方法，所述基站故障检测方法包括以下步骤：获取基站的故障信息，将所述故障信息转化为故障征兆向量；根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量；将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。
- 如权利要求1所述的基站故障检测方法，其中，所述获取基站的故障信息，将所述故障信息转化为故障征兆向量的步骤包括：获取基站的故障信息，确定所述故障信息对应的故障种类；根据所述故障种类获取所述故障种类对应的故障信息表，将所述故障信息与所述故障信息表进行一一比对；将所述故障信息表中与所述故障信息匹配的预存故障信息的对应值设为第一预设值，并将所述故障信息表中与所述故障信息不匹配的预存故障信息的对应值设为第二预设值；根据所述第一预设值和所述第二预设值组合成故障征兆向量。
- 如权利要求1所述的基站故障检测方法，其中，所述根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量的步骤包括：通过预设的故障种类与故障检测矩阵的映射关系，获取确定的故障种类对应的故障检测矩阵；对所述故障征兆向量与获取的所述故障检测矩阵进行矩阵运算，得到故障原因向量。
- 如权利要求1所述的基站故障检测方法，其中，所述方法还包括：所述将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因的步骤之后，根据基站故障原因获取所述基站故障原因对应的处理措施，并将所述处理措施发送给预设的显示终端，以供所述显示 终端显示所述处理措施。
- 如权利要求1-4任一项所述的基站故障检测方法，其中，所述故障检测矩阵由专家经验故障隶属度和经验数据故障隶属度加权平均得到。
- 一种基站故障检测装置，所述基站故障检测装置包括：第一处理模块，设置为获取基站的故障信息，将所述故障信息转化为故障征兆向量；第二处理模块，设置为根据所述故障信息的故障种类获取所述故障种类对应的故障检测矩阵，将所述故障征兆向量与所述故障检测矩阵进行矩阵运算，得到故障原因向量；第三处理模块，设置为将所述故障原因向量中满足预设条件的故障原因分量对应的故障原因作为基站故障原因。
- 如权利要求6所述的基站故障检测装置，其中，所述第一处理模块包括：处理单元，设置为获取基站的故障信息，确定所述故障信息对应的故障种类；所述处理单元，设置为根据所述故障种类获取所述故障种类对应的故障信息表，将所述故障信息与所述故障信息表进行一一比对；转化单元，设置为将所述故障信息表中与所述故障信息匹配的预存故障信息的对应值设为第一预设值，并将所述故障信息表中与所述故障信息不匹配的预存故障信息的对应值设为第二预设值；组合单元，设置为根据所述第一预设值和所述第二预设值组合成故障征兆向量。
- 如权利要求6所述的基站故障检测装置，其中，所述第二处理模块包括：获取单元，设置为通过预设的故障种类与故障检测矩阵的映射关系，获取确定的故障种类对应的故障检测矩阵；计算单元，设置为对所述故障征兆向量与获取的所述故障检测矩阵进行 矩阵运算，得到故障原因向量。
- 如权利要求6所述的基站故障检测装置，所述基站故障检测装置还包括：获取模块，设置为根据基站故障原因获取所述基站故障原因对应的处理措施；发送模块，设置为将所述处理措施发送给预设的显示终端，以供所述显示终端显示所述处理措施。
- 如权利要求6-9任一项所述的基站故障检测装置，其中，所述故障检测矩阵由专家经验故障隶属度和经验数据故障隶属度加权平均得到。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510644708.5 | 2015-10-08 | ||
| CN201510644708.5A CN106572494A (zh) | 2015-10-08 | 2015-10-08 | 基站故障检测方法及装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2016198006A1 true WO2016198006A1 (zh) | 2016-12-15 |
Family
ID=57503012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2016/086695 WO2016198006A1 (zh) | 2015-10-08 | 2016-06-22 | 基站故障检测方法及装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106572494A (zh) |
| WO (1) | WO2016198006A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112379200A (zh) * | 2020-11-10 | 2021-02-19 | 深圳市康必达控制技术有限公司 | 故障定位及解决方法 |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108401257B (zh) * | 2018-03-06 | 2021-01-22 | 北京全路通信信号研究设计院集团有限公司 | 一种铁路基站一体化监测系统 |
| CN109274533B (zh) * | 2018-09-28 | 2022-02-25 | 中国电子科技集团公司第十五研究所 | 一种基于规则引擎的Web服务故障的定位装置和方法 |
| CN109905278A (zh) * | 2019-02-28 | 2019-06-18 | 深圳力维智联技术有限公司 | 基于大数据的基站故障检测方法、装置和存储介质 |
| CN110267288B (zh) * | 2019-05-22 | 2022-04-15 | 中国联合网络通信集团有限公司 | 移动网络投诉定位方法及装置 |
| CN111818566B (zh) * | 2020-06-30 | 2022-09-27 | 武汉虹信科技发展有限责任公司 | 一种射频信号故障监测装置 |
| CN114691398A (zh) * | 2020-12-30 | 2022-07-01 | 中兴通讯股份有限公司 | 一种基于规则的故障分析方法、装置、设备和存储介质 |
| CN117640346A (zh) * | 2024-01-25 | 2024-03-01 | 中兴系统技术有限公司 | 一种通信设备故障诊断方法、存储介质及计算机设备 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7392143B2 (en) * | 2006-03-14 | 2008-06-24 | The University Of Texas System Board Of Regents | Monitoring and fault diagnosis of single-and multi-converter power systems |
| CN101446827A (zh) * | 2008-11-06 | 2009-06-03 | 西安交通大学 | 一种流程工业系统的过程故障分析装置及方法 |
| CN102818948A (zh) * | 2012-07-16 | 2012-12-12 | 北京航空航天大学 | 基于模糊故障诊断和相关性模型诊断的dr合成诊断方法 |
| CN104504607A (zh) * | 2014-09-04 | 2015-04-08 | 国家电网公司 | 一种基于模糊聚类算法的光伏电站故障诊断方法 |
-
2015
- 2015-10-08 CN CN201510644708.5A patent/CN106572494A/zh active Pending
-
2016
- 2016-06-22 WO PCT/CN2016/086695 patent/WO2016198006A1/zh active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7392143B2 (en) * | 2006-03-14 | 2008-06-24 | The University Of Texas System Board Of Regents | Monitoring and fault diagnosis of single-and multi-converter power systems |
| CN101446827A (zh) * | 2008-11-06 | 2009-06-03 | 西安交通大学 | 一种流程工业系统的过程故障分析装置及方法 |
| CN102818948A (zh) * | 2012-07-16 | 2012-12-12 | 北京航空航天大学 | 基于模糊故障诊断和相关性模型诊断的dr合成诊断方法 |
| CN104504607A (zh) * | 2014-09-04 | 2015-04-08 | 国家电网公司 | 一种基于模糊聚类算法的光伏电站故障诊断方法 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112379200A (zh) * | 2020-11-10 | 2021-02-19 | 深圳市康必达控制技术有限公司 | 故障定位及解决方法 |
| CN112379200B (zh) * | 2020-11-10 | 2024-02-13 | 深圳市康必达控制技术有限公司 | 故障定位及解决方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106572494A (zh) | 2017-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2016198006A1 (zh) | 基站故障检测方法及装置 | |
| CN107566358B (zh) | 一种风险预警提示方法、装置、介质及设备 | |
| CN108173670B (zh) | 检测网络的方法和装置 | |
| US8751414B2 (en) | Identifying abnormalities in resource usage | |
| US8751417B2 (en) | Trouble pattern creating program and trouble pattern creating apparatus | |
| US20090125916A1 (en) | Event-processing operators | |
| CN108257659A (zh) | 故障处理方法、故障处理装置及电子终端 | |
| JP7478226B2 (ja) | 充電ステーション監視方法および装置 | |
| CN108241114B (zh) | 混合电路芯片的在线故障检测方法、装置及故障检测电路 | |
| CN110222925B (zh) | 绩效量化考核方法、装置及计算机可读存储介质 | |
| EP3648433B1 (en) | System and method of training behavior labeling model | |
| CN113313280B (zh) | 云平台的巡检方法、电子设备及非易失性存储介质 | |
| US20160063078A1 (en) | Automatic identification and tracking of log entry schemas changes | |
| CN113568812A (zh) | 一种智能机器人的状态检测方法和装置 | |
| JP6148505B2 (ja) | 在室確率推定装置およびその方法、ならびにプログラム | |
| CN116467468A (zh) | 基于知识图谱技术的电力管理系统异常信息处置方法 | |
| US10805186B2 (en) | Mobile communication network failure monitoring system and method | |
| CN115081969A (zh) | 异常数据确定方法及相关装置 | |
| CN111400114A (zh) | 基于深度递归网络大数据计算机系统故障检测方法及系统 | |
| CN108804574B (zh) | 告警提示方法、装置、计算机可读存储介质及电子设备 | |
| US20210089960A1 (en) | Training a machine learning model using a batch based active learning approach | |
| CN110582091B (zh) | 定位无线质量问题的方法和装置 | |
| US20200372982A1 (en) | Imputing an outcome attribute to a pers record missing an outcome attribute using a structured situation string or unstructured case note text associated with the record | |
| US20220376989A1 (en) | Management of predictive models of a communication network | |
| CN115118631A (zh) | 链路异常处理方法、装置、电子设备及存储介质 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16806904 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 16806904 Country of ref document: EP Kind code of ref document: A1 |