TWI752577B - Obstacle management system and method thereof - Google Patents
Obstacle management system and method thereof Download PDFInfo
- Publication number
- TWI752577B TWI752577B TW109126194A TW109126194A TWI752577B TW I752577 B TWI752577 B TW I752577B TW 109126194 A TW109126194 A TW 109126194A TW 109126194 A TW109126194 A TW 109126194A TW I752577 B TWI752577 B TW I752577B
- Authority
- TW
- Taiwan
- Prior art keywords
- obstacle
- circuit
- concentration
- abnormal
- equipment
- Prior art date
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
- Telephonic Communication Services (AREA)
Abstract
Description
本發明係關於網路管理技術，尤其是關於用於網路服務架構的障礙管理系統及其方法。 The present invention relates to network management technology, and more particularly, to a barrier management system and method for network service architecture.
在現今通訊產業中，為了滿足多樣化的網路服務，使得電信網路服務架構越趨複雜，而其中之障礙管理亦越趨困難。例如，在端對端寬頻網路服務架構中，一寬頻網路路由經過的網路設備與元件可能非常繁複，因而當此寬頻網路路由上之某個終端用戶通報異常時，由於障礙點難以判斷，服務營運商往往只能派遣技術人員沿著寬頻網路路由逐段查修，以尋找可能的障礙點來進行故障排除，因而需耗費大量時間與人力。 In today's communication industry, in order to satisfy diversified network services, the telecommunication network service architecture is becoming more and more complex, and the obstacle management is more and more difficult. For example, in an end-to-end broadband network service architecture, the network equipment and components that a broadband network route passes through may be very complicated, so when an end user on the broadband network route reports an abnormality, it is difficult to Judging from the judgment, service operators often can only dispatch technicians to repair section by section along the route of the broadband network to find possible obstacles for troubleshooting, which requires a lot of time and manpower.
因此，亟需一種障礙管理系統及其方法，以解決上述對於網路服務架構中障礙管理的人工查修與派工以及障礙點難以判斷之問題，並增加判斷的準確率與效率。 Therefore, there is an urgent need for an obstacle management system and a method thereof, which can solve the above-mentioned problems of manual inspection, repair and dispatch of obstacle management and difficulty in judging obstacle points in the network service architecture, and increase the accuracy and efficiency of judgment.
本發明係提供一種障礙管理系統，包括：事件管理模組，係用於維護電路查測資料；分析模組，係用於根據該電路查測資料產生設備濃度異常歷史資料；以及障礙預測模組，係用於根據該設備濃度異常歷史資料及障礙報表產生模糊關聯矩陣，其中，該電路查測資料係查測事件之紀錄，該設備濃度異常歷史資料係各類該查測事件中軌跡濃度異常之網路設備之紀錄，且該障礙報表係經申報之障礙電路之紀錄。 The present invention provides an obstacle management system, comprising: an event management module for maintaining circuit inspection data; an analysis module for generating abnormal history data of equipment concentration according to the circuit inspection data; and an obstacle prediction module , which is used to generate a fuzzy correlation matrix based on the abnormal historical data of the equipment concentration and the obstacle report. Among them, the circuit inspection data is the record of the inspection event, and the historical data of the abnormal concentration of the device is the abnormal track concentration in the various inspection events. The record of the network equipment, and the obstacle report is the record of the declared obstacle circuit.
在前述之障礙管理系統中，復包括用於儲存網路服務路由資訊之路由資料庫。 In the aforementioned obstacle management system, a routing database for storing the routing information of the network service is further included.
在前述之障礙管理系統中，該分析模組產生該設備濃度異常歷史資料之方式係包括：查找該路由資料庫之該網路服務路由資訊，以獲得各類該查測事件對應之查測軌跡，其中，該查測軌跡係包含各類該查測事件對應之路由及該路由上的節點；計算該路由上各該節點對應的軌跡數值加總及軌跡濃度；選擇各該節點中軌跡濃度超出軌跡臨界值且具有最高軌跡數值加總者對應之網路設備為該軌跡濃度異常之網路設備；以及紀錄該軌跡濃度異常之網路設備，以形成該設備濃度異常歷史資料，其中，該軌跡數值加總係對應各類該查測事件的查測次數，且該軌跡濃度係各該軌跡數值加總與各該節點之設備收容量的比值。 In the aforementioned obstacle management system, the method of generating the abnormal historical data of the equipment concentration by the analysis module includes: searching the network service routing information in the routing database to obtain the inspection trajectories corresponding to various inspection events , wherein the inspection track includes various routes corresponding to the inspection event and the nodes on the route; calculate the sum of the track values and the track concentration corresponding to each node on the route; select the track concentration in each node that exceeds the The network device corresponding to the trajectory threshold value and the one with the highest total trajectory value is the network device with the abnormal concentration of the trajectory; and the network device that records the abnormal concentration of the trajectory, so as to form the historical data of abnormal concentration of the equipment, wherein the trajectory The total value is the number of inspections corresponding to each type of the inspection event, and the track concentration is the ratio of the total value of each track to the equipment capacity of each node.
在前述之障礙管理系統中，該障礙預測模組產生該模糊關聯矩陣之方式係包括：查找該路由資料庫之該網路服務路由資訊，以獲得該障礙報表之各該障礙電路對應的電路資訊；比對各該障礙電路及該設備濃度異常歷史資料，以獲得該軌跡濃度異常之網路設備中存在於該電路資訊中者；依據各該障礙電路及該設備濃度異常歷史資料的比對結果，統計各該障礙電路所屬障礙類 型中各網路設備種類的障礙機率；以及依據各該障礙類型及對應之各該障礙機率產製該模糊關聯矩陣。 In the aforementioned obstacle management system, the method of generating the fuzzy association matrix by the obstacle prediction module includes: searching the network service routing information in the routing database to obtain circuit information corresponding to each obstacle circuit in the obstacle report ; Compare the abnormal concentration history data of each of the obstacle circuits and the equipment to obtain the network equipment with abnormal concentration of the trace that exists in the circuit information; according to the comparison results of the historical data of the abnormal concentration of each of the obstacle circuits and the equipment , and count the obstacle classes that each obstacle circuit belongs to The obstacle probability of each network device type in the model; and the fuzzy correlation matrix is produced according to each obstacle type and the corresponding obstacle probability.
在前述之障礙管理系統中，該障礙預測模組復用於診斷新申報之障礙電路的障礙類型，其診斷方式包括：接收該新申報之障礙電路；查找該路由資料庫之該網路服務路由資訊，以提取該新申報之障礙電路之電路資訊；比對該新申報之障礙電路及該設備濃度異常歷史資料，以獲得該軌跡濃度異常之網路設備中存在於該電路資訊中者；依據該新申報之障礙電路及該設備濃度異常歷史資料的比對結果生成該電路資訊中對應網路設備種類之故障特徵值集合，其中，該故障特徵集合係該新申報之障礙電路中各該網路設備種類的障礙機率之記錄；以及將該故障特徵值集合與該模糊關聯矩陣執行合成運算，以判斷該新申報之障礙電路之障礙類型。 In the aforementioned obstacle management system, the obstacle prediction module is reused for diagnosing the obstacle type of the newly reported obstacle circuit, and the diagnosis method includes: receiving the newly reported obstacle circuit; searching the network service route in the routing database information to extract the circuit information of the newly reported obstacle circuit; compare the newly reported obstacle circuit and the abnormal historical data of the equipment concentration to obtain the circuit information of the network equipment with the abnormal concentration of the trace; basis The comparison result of the newly declared obstacle circuit and the abnormal historical data of the equipment concentration generates a set of fault characteristic values corresponding to the type of network equipment in the circuit information, wherein the fault characteristic set is each network in the newly declared obstacle circuit. record the probability of failure of the road equipment type; and perform a composite operation on the set of fault feature values and the fuzzy correlation matrix to determine the type of failure of the newly reported obstacle circuit.
本發明復提供一種障礙管理方法，包括：取用電路查測資料；分析該電路查測資料，以產生設備濃度異常歷史資料；以及取用障礙報表，以比較該障礙報表及該設備濃度異常歷史資料，俾產生模糊關聯矩陣，其中，該電路查測資料係查測事件之紀錄，該設備濃度異常歷史資料係各類該查測事件中軌跡濃度異常之網路設備之紀錄，該障礙報表係經申報之障礙電路之紀錄。 The present invention further provides an obstacle management method, comprising: obtaining circuit inspection data; analyzing the circuit inspection data to generate equipment concentration abnormal history data; and obtaining obstacle report to compare the obstacle report and the equipment concentration abnormal history data to generate a fuzzy correlation matrix, in which the circuit inspection data is the record of the inspection event, the historical data of the abnormal concentration of the equipment is the record of various network devices with abnormal track concentration in the inspection event, and the obstacle report is Records of reported obstacle circuits.
在前述之障礙管理方法中，該分析該電路查測資料以產生設備濃度異常歷史資料之步驟係包括以下子步驟：查找路由資料庫之網路服務路由資訊，以獲得各類該查測事件對應之查測軌跡，其中，該查測軌跡係包含各類該查測事件對應之路由及該路由上的節點；計算該路由上各該節點對應的軌跡數值加總及軌跡濃度；選擇各該節點中軌跡濃度超出軌跡臨界值且具有最高軌跡數值加總者對應之網路設備為該軌跡濃度異常之網路設備；以及紀錄該軌跡濃 度異常之網路設備，以形成該設備濃度異常歷史資料，其中，該軌跡數值加總係對應各類該查測事件的查測次數，且該軌跡濃度係各該軌跡數值加總與各該節點之設備收容量的比值。 In the aforementioned obstacle management method, the step of analyzing the circuit inspection data to generate the abnormal history data of the equipment concentration includes the following sub-steps: searching the network service routing information in the routing database to obtain the corresponding inspection events of various types The inspection trajectory, wherein the inspection trajectory includes various routes corresponding to the inspection event and the nodes on the route; calculate the sum of the trajectory values and the trajectory concentration corresponding to each node on the route; select each node The network device corresponding to the trace concentration exceeding the threshold value of the trace and the sum of the highest trace value is the network device with abnormal concentration of the trace; and recording the concentration of the trace The network equipment with abnormal degrees is used to form the historical data of abnormal concentration of the equipment, wherein the sum of the trace values is the number of inspections corresponding to various types of inspection events, and the trace concentration is the sum of the values of the traces and the sum of the values of the traces. The ratio of the equipment capacity of the node.
在前述之障礙管理方法中，該比較該障礙報表及該設備濃度異常歷史資料以產生模糊關聯矩陣之步驟係包括以下子步驟：查找路由資料庫之網路服務路由資訊，以獲得該障礙報表之各該障礙電路對應的電路資訊；比對各該障礙電路及該設備濃度異常歷史資料，以獲得該軌跡濃度異常之網路設備中存在於該電路資訊中者；依據各該障礙電路及該設備濃度異常歷史資料的比對結果，統計各該障礙電路所屬障礙類型中各網路設備種類的障礙機率；以及依據各該障礙類型及對應之各該障礙機率產製該模糊關聯矩陣。 In the aforementioned obstacle management method, the step of comparing the obstacle report and the abnormal historical data of the equipment concentration to generate the fuzzy correlation matrix includes the following sub-steps: searching the network service routing information in the routing database to obtain the obstacle report. The circuit information corresponding to each of the obstacle circuits; compare the history data of the abnormal concentration of each of the obstacle circuits and the equipment to obtain the information of the network equipment with the abnormal concentration of the trace that exists in the circuit information; according to each of the obstacle circuits and the equipment The comparison result of the historical data of abnormal concentration, statistics the obstacle probability of each network device type in the obstacle type of the obstacle circuit; and produce the fuzzy correlation matrix according to each obstacle type and the corresponding obstacle probability.
在前述之障礙管理方法中，復包括：接收新申報之障礙電路；查找路由資料庫之網路服務路由資訊，以提取該新申報之障礙電路之電路資訊；比對該新申報之障礙電路及該設備濃度異常歷史資料，以獲得該軌跡濃度異常之網路設備中存在於該電路資訊中者；依據該新申報之障礙電路及該設備濃度異常歷史資料的比對結果生成該電路資訊中對應網路設備種類之故障特徵值集合，其中，該故障特徵集合係該新申報之障礙電路中各該網路設備種類的障礙機率之紀錄；以及將該故障特徵值集合與該模糊關聯矩陣執行合成運算，以判斷該新申報之障礙電路之障礙類型。 In the aforesaid obstacle management method, it further includes: receiving the newly reported obstacle circuit; searching the network service routing information of the routing database to extract the circuit information of the newly reported obstacle circuit; comparing the newly reported obstacle circuit and The historical data of abnormal concentration of the equipment is obtained to obtain the network equipment with abnormal concentration of the trace that exists in the circuit information; according to the comparison result of the newly declared obstacle circuit and the historical data of abnormal concentration of the equipment, the corresponding circuit information is generated. A set of fault characteristic values of network equipment types, wherein the fault characteristic set is a record of the failure probability of each network equipment type in the newly reported faulty circuit; and the set of fault characteristic values is synthesized with the fuzzy correlation matrix Calculation is performed to determine the obstacle type of the newly declared obstacle circuit.
在前述之障礙管理方法中，復包括每隔預定時段更新該電路查測資料、該濃度異常歷史資料及該障礙報表。 In the aforesaid obstacle management method, it further includes updating the circuit inspection data, the concentration abnormality history data and the obstacle report every predetermined period.
綜上所述，本發明之障礙管理系統及其方法係利用網路設備之軌跡濃度分析電路查測資料，以產生設備濃度異常歷史資料，再將設備濃度異常 歷史資料與歷史障礙電路比對，以產生關聯模糊矩陣，進而藉由關聯模糊矩陣判斷障礙電路最有可能的障礙類型，故工作人員不需沿著障礙電路的路由逐個查測障礙原因，因而大幅減少障礙排除時間及派工成本，並增加判斷的準確率與效率。 To sum up, the obstacle management system and method of the present invention use the trace concentration analysis circuit of the network device to check the data to generate the abnormal history data of the device concentration, and then analyze the abnormal device concentration. The historical data is compared with the historical obstacle circuit to generate a correlation fuzzy matrix, and then the most likely obstacle type of the obstacle circuit can be judged by the correlation fuzzy matrix, so the staff does not need to check the cause of the obstacle one by one along the route of the obstacle circuit. Reduce obstacle removal time and dispatch costs, and increase the accuracy and efficiency of judgment.
10:網路設備健檢裝置 10: Network equipment health check device
11:事件管理模組 11: Event Management Module
12:分析模組 12: Analysis module
13:障礙預測模組 13: Obstacle prediction module
20:障礙電路申報裝置 20: Obstruction circuit reporting device
30:電路查測客服裝置 30: Circuit test customer service device
41:設備濃度異常歷史資料庫 41: History database of equipment concentration abnormality
42:路由資料庫 42: Routing database
S100~S600:步驟 S100~S600: Steps
圖1及圖2係本發明之障礙管理系統及其方法的實施概念示意圖； FIG. 1 and FIG. 2 are schematic diagrams of the implementation concept of the obstacle management system and the method thereof of the present invention;
圖3係本發明之障礙管理系統的架構示意圖； 3 is a schematic diagram of the structure of the obstacle management system of the present invention;
圖4及圖5係本發明之障礙管理系統及其方法的實施階段示意圖； 4 and 5 are schematic diagrams of the implementation stages of the obstacle management system and the method thereof of the present invention;
圖6係本發明之障礙管理系統及其方法的實施階段示意圖； 6 is a schematic diagram of the implementation stage of the obstacle management system and method thereof of the present invention;
圖7係本發明之障礙管理系統及其方法的實施階段示意圖； 7 is a schematic diagram of the implementation stage of the obstacle management system and method thereof of the present invention;
圖8係本發明之障礙管理系統及其方法的實施階段示意圖；以及 FIG. 8 is a schematic diagram of the implementation stage of the obstacle management system and method thereof of the present invention; and
圖9係本發明之障礙管理方法的步驟流程圖。 FIG. 9 is a flow chart of the steps of the obstacle management method of the present invention.
以下藉由特定的具體實施例說明本發明之實施方式，熟悉此技藝之人士可由本說明書所揭示之內容輕易地瞭解本發明之其他優點及功效。須知，本說明書所附圖式所繪示之結構、比例、大小等，均僅用以配合說明書所揭示之內容，以供熟悉此技藝之人士之瞭解與閱讀，並非用以限定本發明可實施之限定條件，故不具技術上之實質意義，任何結構之修飾、比例關係之改變或大小之調整，在不影響本發明所能產生之功效及所能達成之目的下，均應仍落在本發明所揭示之技術內容得能涵蓋之範圍內。 The following specific embodiments are used to illustrate the implementation of the present invention, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. It should be noted that the structures, proportions, sizes, etc. shown in the drawings in this specification are only used to cooperate with the contents disclosed in the specification for the understanding and reading of those who are familiar with the art, and are not intended to limit the implementation of the present invention. Therefore, it has no technical significance. Any modification of the structure, change of the proportional relationship or adjustment of the size should still fall within the scope of the present invention without affecting the effect and the purpose that the present invention can achieve. The technical content disclosed by the invention can be covered within the scope.
參考圖1至圖2，本發明之障礙管理系統及其方法主要是根據網路服務架構(例如，寬頻網路服務架構)中各網路設備發生障礙之軌跡濃度達成最有效率的障礙管理。此處所述軌跡濃度係指網路服務架構於某時段內被反應發生障礙之路由上各網路設備對應遺留的軌跡數值，類似於蟻群覓食路徑上遺留之費洛蒙，其應用方式係如以下所詳述者。 1 to 2 , the obstacle management system and method of the present invention mainly achieve the most efficient obstacle management according to the track concentration of obstacles occurring in each network device in a network service architecture (eg, broadband network service architecture). The trace concentration mentioned here refers to the trace value left by each network device on the route where the network service architecture is reacted to the obstacle within a certain period of time, similar to the pheromones left on the foraging path of the ant colony. are as detailed below.
在圖1圖示之典型的寬頻網路架構圖中，假設在某時段內用戶A反應其個人連網設備(例如，個人電腦、手機、或類似之連網設備)所連接之路由上發生某類型之障礙，則可先藉此路由判斷(從用戶端往局端方向)可能有障礙的網路設備依次為設備1、設備2及設備3並被分別累積相應的軌跡濃度。此時，若用戶B在相同時段內也反應發生同樣類型之障礙，則其連網之路由上之各網路設備(例如，設備4、設備2及設備3)亦將被累積相應的軌跡濃度，並可由用戶A與用戶B路由重複處判斷設備2及設備3為軌跡濃度累積較高的網路設備。
In the typical broadband network architecture diagram shown in Figure 1, it is assumed that during a certain period of time, user A responds that a certain occurrence occurs on the route to which his personal networking device (eg, personal computer, mobile phone, or similar networking device) is connected. Type of obstacle, you can first use this route to determine (from the client to the central office) that the network devices that may have obstacles are
圖2係圖示前述圖1之態樣中各網路設備的軌跡濃度隨時間的變化態樣。詳細而言，圖2中橫軸上之時間t1係代表前述用戶A反映其路由上有障礙的時點(例如，以秒為單位)，而時間t2係代表前述用戶B反映其路由上有障礙的時點。由此可看到，用戶A及用戶B反映有障礙時都會造成其路由上各網路設備的軌跡濃度增加及/或累積。進一步地，還可預設一門檻值以辨別有最可能有障礙之網路設備，如圖2中當用戶B在時間t2反映其路由上有障礙時使得設備2的軌跡濃度累積超過門檻值，則可依此門檻值推估設備2最有可能為出現障礙的網路設備。
FIG. 2 is a graph showing the variation of the trace density of each network device with time in the aspect of FIG. 1 . In detail, the time t1 on the horizontal axis in FIG. 2 represents the time point (eg, in seconds) when the aforementioned user A reports that his route has obstacles, and the time t2 represents the aforementioned user B reports that his route has obstacles. time. From this, it can be seen that when user A and user B report an obstacle, the trace density of each network device on their route will increase and/or accumulate. Further, a threshold value can also be preset to identify the network equipment that is most likely to have obstacles. As shown in Figure 2, when user B reflects that there are obstacles on its route at time t2, the trajectory concentration of
額外地，相同於蟻群覓食路徑上費洛蒙隨時間揮發之方式，圖2所示之軌跡濃度亦會隨時間而呈指數消減(揮發)，從而指示某網路設備之障礙狀態 已修復或排除。舉例來說，圖2係各網路設備的軌跡濃度將於最後增加(或累積)的時間T(例如，以秒為單位)內以指數消減方式歸零，以指示此些網路設備無後續反應出現障礙的情形。 In addition, similar to the way pheromones volatilize over time on the foraging path of the ant colony, the concentration of the trace shown in Figure 2 will also decrease exponentially (volatilize) over time, thereby indicating the obstacle status of a certain network device. Fixed or excluded. For example, Figure 2 shows that the trace concentrations of each network device will exponentially decrease to zero within the last increased (or accumulated) time T (eg, in seconds) to indicate that these network devices have no follow-up Response to a situation where there is an obstacle.
因此，為體現上述以軌跡濃度為排除障礙之依據的障礙管理，本發明係提出如圖3所示之障礙管理系統及對應之障礙管理方法，其實施方式及態樣係如以下所詳述者。 Therefore, in order to embody the obstacle management based on the trajectory concentration as the basis for eliminating obstacles, the present invention proposes the obstacle management system and the corresponding obstacle management method as shown in FIG. .
參閱圖3，本發明之障礙管理系統至少包括相互通訊連接之網路設備健檢裝置10、障礙電路申報裝置20及電路查測客服裝置30。
Referring to FIG. 3 , the obstacle management system of the present invention at least includes a network equipment
在本文之障礙管理系統中，網路設備健檢裝置10係包括事件管理模組11、分析模組12及障礙預測模組13，並與設備濃度異常歷史資料庫41及路由資料庫42介接，以藉由各模組及資料庫之協同運作以執行本發明之障礙管理方法。
In the obstacle management system of this paper, the network equipment
在本文之障礙管理系統中，障礙電路申報裝置20係提供有操作介面，以協助記錄用戶申報之障礙電路以及執行障礙電路之查修追蹤、原因判定、任務轉派、查詢及統計報表等作業。另外，障礙電路申報裝置20還用於跟據用戶所申報之障礙電路維護障礙報表(例如，藉由障礙電路申報裝置20內建之資料庫儲存所申報障礙電路的歷史資料)，以作為網路設備健檢裝置10執行本發明之障礙管理方法的參考資料之一。
In the obstacle management system of this paper, the obstacle
電路查測客服裝置30係用於提供查測電路品質的客戶服務。舉例來說，當客戶發現電路有問題時，可先透過電路查測客服系統30查測此電路的品質，並在經確認有故障的情況下再至障礙電路申報裝置20進行障礙電路的申報。
進一步地，電路查測客服裝置30還用於維護客戶之查測事件所產生的電路查測資料，以作為網路設備健檢裝置10執行本發明之障礙管理方法的參考資料之一。
The circuit testing
以下將以圖4至圖8說明上述各裝置及模組執行障礙管理的實施方式。 The following will describe the implementation of failure management of the above devices and modules with reference to FIG. 4 to FIG. 8 .
首先，本發明之障礙管理係起始於接收電路查測客服裝置30之電路查測資料時。此時，網路設備健檢裝置10的事件管理模組11會接收並正規化查測事件以及藉其過濾機制排除任何非相關的查測事件，例如，在用戶端計畫性改接(例如，換門號或替換網路服務方案)而產生的查測事件或經確認為非故障情形之查測事件。接著，經事件管理模組11過濾之電路查測資料可交由分析模組12進行設備濃度異常歷史資料的收集。
First, the failure management of the present invention begins when circuit inspection data from the circuit inspection
在進一步實施例中，事件管理模組11復經配置以固定周期或應需求更新從電路查測客服裝置30所接收之電路查測資料，以確保電路查測資料在後續分析過程中的時間代表性。
In a further embodiment, the
圖4及圖5係分別圖示分析模組12收集設備濃度異常歷史資料的不同實施態樣。
FIG. 4 and FIG. 5 respectively illustrate different implementations in which the
圖4係圖示當分析電路查測資料(來自事件管理模組11)時，發現僅一筆「ARM01」類型之查測事件的實施態樣。此時，分析模組12可根據此查測事件查找路由資料庫42(其內建有完整之網路服務架構之路由資訊)中對應的電路資料(即，圖4中粗體線所標示的路由)及端對端查測軌跡(即，如上述標示之路由對應之網路設備及節點編號)。接著，分析模組12係分析此電路資料及端對端查測軌跡以產生設備濃度異常資料，如以下表一所示。
FIG. 4 is a diagram illustrating an implementation of finding only one "ARM01" type of test event when analyzing the circuit test data (from the event management module 11). At this time, the
在表一中，分析模組12首先由上到下依次紀錄整理出此筆查測事件所對應電路資料中自客戶端至局端(即，以路由L01、L09及L11至L12之順序)經過的各節點所屬網路設備種類、網路設備名稱、及節點名稱與編號(亦即，進出埠名稱及其編號)，如左側開始第一至三欄所示。接著，分析模組12萃取各節點對應之設備、機框、卡片及埠的軌跡數值加總(如第四欄所示)，並計算各網路設備的最大設備收容量及軌跡濃度(如第五、六欄所示)，進而與預設之軌跡臨界值(如第七欄所示)比較而推估軌跡濃度異常之網路設備。
In Table 1, the
在上述表一之態樣中，軌跡濃度係定義為各節點對應之軌跡數值加總與設備收容量的比值，故可以看出此筆查測事件內軌跡濃度異常(即，超出
軌跡臨界值「1/2」)者僅有編號為「P01」之節點(如灰色背景之欄位所示)，故推測網路設備「ONU01」為此筆查測事件中的潛在障礙點，因此，此項資訊將會被分析模組12在設備濃度異常歷史資料庫41中儲存為一筆設備濃度異常資料。
In the aspect of Table 1 above, the trajectory concentration is defined as the ratio of the sum of the trajectory values corresponding to each node to the equipment capacity, so it can be seen that the trajectory concentration in this detected event is abnormal (that is, exceeding the
The trajectory threshold "1/2") has only the node numbered "P01" (as shown in the column with gray background), so it is presumed that the network device "ONU01" is a potential obstacle point in this inspection event. Therefore, this information will be stored as a piece of equipment concentration abnormality data in the equipment concentration
圖5係圖示當分析電路查測資料(來自於事件管理模組11)時，發現有複數筆「ARM02」類型(例如，四筆)之查測事件的實施態樣。此時，分析模組12係根據多筆「ARM02」類型之查測事件查詢路由資料庫42中對應的電路資料(即，圖5中粗體線所標示的路由)及端對端查測軌跡(即，如上述標示之路由對應之網路設備及節點編號)。接著，分析模組12係分析此些電路資料及端對端查測軌跡以產生設備濃度異常資料，如以下表二所示。
FIG. 5 is a diagram illustrating an implementation of a plurality of “ARM02” type (eg, four) test events when analyzing the circuit test data (from the event management module 11 ). At this time, the
在表二中，類似地，分析模組12首先首先由上到下依次紀錄整理出此些查測事件所查測電路中自客戶端至局端(即，以路由L01、L02、L03及L04、L09至L11至L12之順序)經過的各節點所屬網路設備種類、網路設備名稱、及節點名稱與編號(亦即，進出埠名稱及其編號)，如左側開始第一至三欄所示。接著，分析模組12萃取各節點對應之設備、機框、卡片及埠的軌跡數值加總值(如第四欄所示)，並計算各網路設備的最大設備收容量及軌跡濃度(如第五、六欄所示)，進而與預設之軌跡臨界值(如第七欄所示)比較而推估軌跡濃度異常之網路設備。
In Table 2, similarly, the
此處可看出，軌跡數值加總係對應此些查測事件的查測次數，而軌跡濃度係定義為各節點對應之軌跡數值加總與設備收容量的比值，且此些查測事件中節點編號為「P01，P02，P03，P04與P17」處皆超出軌跡臨界值(即，「1/2」)，但考量同一網路設備的進出埠處累加而來的軌跡數值加總在「P17」處最高，故
推論節點「P17」對應之網路設備「光分歧器01」為此些查測事件中的潛在障礙點，因此，此項資訊(如灰色背景之欄位所示)將會被分析模組12在設備濃度異常歷史資料庫41中儲存為一筆設備濃度異常資料。
It can be seen here that the sum of the trajectory values corresponds to the number of inspections of these inspection events, and the trajectory concentration is defined as the ratio of the sum of the trajectory values corresponding to each node to the equipment capacity, and among these inspection events The nodes numbered "P01, P02, P03, P04 and P17" all exceed the trajectory threshold (ie, "1/2"), but the accumulated trajectory values from the ingress and egress ports of the same network device are summed up in " P17” is the highest, so
The network device "
接續地，分析模組12係將累積一段時間(例如，一小時)的設備濃度異常資料再萃取並簡化，以在設備濃度異常歷史資料庫41中整理為設備濃度異常歷史資料，如表三所示。
Next, the
在表三中，設備濃度異常歷史資料係例示2019年11月1日14點至15點間發生軌跡濃度異常(即，超出軌跡臨界值)的每一個網路設備，其包括：網路設備種類、網路設備名稱、發生異常的節點名稱及編號以及濃度異常發生時間等。並且，此設備濃度異常歷史資料係提供障礙預測模組13作為後續障礙預測的參考資料。須知，上述表一、表二及表三中各節點名稱及節點編號依據產生時對應電路的態樣而定可能相同或不相同，在本文中並不特別限定。
In Table 3, the historical data of abnormal device concentration is an example of each network device with abnormal trajectory concentration (that is, exceeding the trajectory threshold) between 14:00 and 15:00 on November 1, 2019, including: network device types , the name of the network device, the name and number of the node where the abnormality occurred, and the time when the concentration abnormality occurred. In addition, the abnormal history data of the equipment concentration provides the
須知，分析模組12亦可以固定周期或應需求隨時更新(例如，重新收集設備濃度異常歷史資料)儲存於設備濃度異常歷史資料庫41中的設備濃度異常歷史資料，以確保設備濃度異常歷史資料在後續分析過程中的時效代表性。
It should be noted that the
以下將介紹障礙預測模組13進行障礙預測之實施態樣。
The following will introduce the implementation aspect of the
在執行障礙預測前，障礙預測模組13首先將取用並分析障礙電路申報裝置20所維護之障礙報表(障礙電路申報裝置20內建之資料庫所儲存之申報障礙電路的歷史資料)以產生關聯模糊矩陣，作為障礙預測的診斷標準。舉一實施例來說，假設發現障礙報表中記錄之由光纖到大樓(FTTB)所申報的障礙電路中，有164筆障礙電路是「L3交換器(L3SW)」類型，則障礙預測模組13係進一步分析此164筆障礙電路，以統計L3SW類型的障礙電路中各網路設備種類發生軌跡濃度異常的障礙機率。
Before executing the failure prediction, the
在統計障礙機率時，障礙預測模組13係根據各筆障礙電路(例如，前述實施例之164筆L3SW類型之障礙電路)取用路由資料庫42中對應的電路資料。圖6係圖示前述164筆L3SW類型之障礙電路中一者的電路資料，可以發現此筆障礙電路的路由經過的網路設備包含一顆光纖網路終端(OLT-101)、兩顆十億位元乙太網路交換器(GESW-101及GESW-102)、一顆高效能邊緣路由器(HPER-
101)、兩顆多重服務邊緣路由器(MSER-101及MSER-102)及一顆寬頻遠端接入系統(BRAS-101)。進一步地，根據此筆障礙電路的電路資訊及其申報時間，障礙預測模組13將查找設備濃度異常歷史資料庫41儲存之設備濃度異常歷史資料(例如，表三內容)中是否有上述各網路設備的濃度異常紀錄。
When calculating the obstacle probability, the
此時，假設上述障礙電路的申報時間為2019年11月1日的15點，則障礙預測模組13係查詢設備濃度異常歷史資料(例如，表三)中此申報時間之前的一預定時段(例如，一小時)各網路設備出現軌跡濃度異常的紀錄，並得出2019年11月1日的14點至15點間分別記錄有GESW-101(P14)、GESW-102(P15)及HPER-101(P19)等網路設備(其係對應至上述障礙電路中包含的網路設備)的軌跡濃度有超過軌跡臨界值的情形。藉此查詢結果，障礙預測模組13可為GESW及HPER之網路設備種類各紀錄一筆「1/1(命中/總筆數)」統計結果。另外，由於此筆障礙電路的其餘網路設備種類在此預定時段內並無出現軌跡濃度異常之紀錄，故對於其餘網路設備種類可記錄「0/1(命中/總筆數)」統計結果。
At this time, assuming that the reporting time of the above-mentioned obstacle circuit is 15:00 on November 1, 2019, the
接續，由於L3SW類型障礙電路有164筆，故障礙預測模組13係將164筆障礙電路全部分析，並將各網路設備種類「命中/總筆數」的統計結果累加，遂整理成如以下表四之統計結果，作為L3SW類型障礙電路中各網路設備種類發生軌跡濃度異常的評估指標。
Continuing, since there are 164 L3SW type obstacle circuits, the
進一步說明，以下表四第一欄係表示在分析完L3SW類型障礙電路後，統計各網路設備種類在這164筆障礙電路中發生軌跡濃度異常的比例，即各網路設備種類的障礙機率。接著，可將障礙機率從分數轉換為小數形式，如第二欄所示。最後，可參考表五之分級標準將各障礙機率模糊化，如第三欄的模糊化結果所示。 For further explanation, the first column of Table 4 below indicates that after analyzing the L3SW type obstacle circuit, the proportion of abnormal track concentration in these 164 obstacle circuits for each network device type is counted, that is, the obstacle probability of each network device type. Next, the handicap probability can be converted from fractional to decimal, as shown in the second column. Finally, the probability of each obstacle can be fuzzified by referring to the grading criteria in Table 5, as shown in the fuzzification results in the third column.
上述障礙機率的模糊化用途在於使得後續障礙預測的運算中數值得以統一化。如上述表四中，GESW的障礙機率「0.402」可依表五表示為中等機率(M)；而HPER的障礙機率「0.299」可依表五表示為中低機率(ML)；而其餘網路設備種類之障礙機率皆在「0.01」及「0.2」之間，故可表示為低機率(L)。 The purpose of fuzzification of the above obstacle probability is to unify the numerical value in the calculation of the subsequent obstacle prediction. As shown in Table 4 above, the obstacle probability of GESW "0.402" can be expressed as a medium probability (M) according to Table 5; and the obstacle probability of HPER "0.299" can be expressed as a medium-low probability (ML) according to Table 5; The obstacle probability of the equipment type is between "0.01" and "0.2", so it can be expressed as a low probability (L).
對於障礙電路申報裝置20所維護之障礙報表中其他類型之障礙電路(例如，光纖收發器(FOT)、超高速數位用戶回路(VDSL)、光纖網路終端(OLT)、L2交換器(L2SW)、高效能邊緣路由器(HPER)、寬頻遠端接入系統(BRAS)等)，障礙預測模組13係套用上述表四之方法以統計各類型障礙電路中各網路設備種類發生軌跡濃度異常的障礙機率及其模糊化結果，遂整合為如圖7所示之關聯模糊矩陣。
For other types of barrier circuits in the barrier report maintained by the barrier circuit reporting device 20 (eg, fiber optic transceiver (FOT), very high speed digital subscriber loop (VDSL), optical network terminal (OLT), L2 switch (L2SW) , High-Performance Edge Router (HPER), Broadband Remote Access System (BRAS), etc.), the
完成關聯模糊矩陣之後，障礙預測模組13即可針對障礙電路進行診斷以執行障礙預測(例如，以障礙電路申報裝置20之操作介面指定障礙預測模組13對任一障礙電路執行障礙預測)。
After completing the associated fuzzy matrix, the
舉例來說，假設障礙電路申報裝置20處新申報(例如，透過電路申報裝置20的操作介面)一障礙電路，且障礙預測模組13經查此障礙電路的路由(例如，比對路由資料庫42之電路資料)上經過的網路設備包括四顆GESW、一顆OLT、三顆MSER、四顆HPER及一顆BRAS，其中在過去一小時內發生軌跡濃度異常的網路設備(例如，比對設備濃度異常歷史資料庫41中的設備濃度異常歷史資料)有一顆GESW、一顆OLT、零顆MSER、三顆HPER及一顆BRAS。則障礙預測模組13可藉由關聯模糊矩陣(如圖7之關聯模糊矩陣)及上述資訊判斷此電路最可能的障礙類型。
For example, it is assumed that a barrier circuit is newly reported (eg, through the operation interface of the circuit reporting device 20 ) at the barrier
在障礙預測之運算過程中，障礙預測模組13首先將圖7之關聯模糊矩陣去模糊化，亦即將被分級的障礙機率賦予對應量化數值。在本實施例中，係將高機率(H)量化為0.9、中高機率(HM)量化為0.7、中機率(M)量化為0.5、中低機率(ML)量化為0.3、低機率(L)量化為0.1以及零機率量化為0，所得到去模糊化的關聯模糊矩陣的結果係如圖8的矩陣R所示。
In the operation process of obstacle prediction, the
接著，障礙預測模組13係針對此障礙電路替各網路設備種類提取故障特徵值集合。例如，由於此電路中四顆GESW中有一顆曾發生軌跡濃度異常，則GESW的故障特徵值可表示為1/4=0.25(即，GESW的障礙機率)，並依同理類推可得到故障特徵值集合X為GESW=0.25、OLT=1、MSER=0、HPER=0.75、NRAS=1，即X={0.25,1,0,0.75,1}。
Next, the
最後，障礙預測模組13係將故障特徵集合X與去模糊化的關聯模糊矩陣R進行最大最小值(Max-Min)合成運算(如圖8所示之運算過程)，以得到模糊故障集合Y={0.5,0.1,0.7,0.25,0.3,0.3,0.5}，即FOT=0.5、VDSL=0.1、OLT=0.7、L2SW=0.25、L3SW=0.3、HPER=0.3及BRAS=0.5。由此推論此障礙電路可能之障礙類型的機率由高至低依序為：OLT>BRAS=FOT>L3SW=HPER>L2SW>VDSL。故當工作人員欲快速排除此障礙電路時，可依據模糊故障集合Y優先針對OLT之障礙類型進行檢測，再依序檢測BRAS或FOT之障礙類型，依此類推。
Finally, the
須知，障礙預測模組13亦可以固定周期或應需求隨時更新關聯模糊矩陣及去模糊化的規則，以確保關聯模糊矩陣在執行障礙預測過程中的時效代表性。
It should be noted that the
圖9係揭示本發明之障礙管理方法的步驟流程圖。 FIG. 9 is a flow chart showing the steps of the obstacle management method of the present invention.
首先在步驟S100，事件管理模組11係收集並過濾來自電路查測客服裝置30的電路查測資料。
First in step S100 , the
接續於步驟S200，分析模組12係記錄電路查測資料之各筆查測事件中軌跡濃度超過軌跡臨界值且軌跡數值加總最高的網路設備，以記錄為設備濃度異常資料。
Continuing in step S200, the
接續於步驟S300，將設備濃度異常資料於設備濃度異常歷史資料庫41中儲存為設備濃度異常歷史資料。
Following step S300, the abnormal equipment concentration data is stored in the abnormal equipment concentration
接續於步驟S400，障礙預測模組13係將障礙電路申報裝置20儲存之障礙報表與設備濃度異常歷史資料進行比對，以計算各障礙類型關聯之網路設備種類的障礙機率。
Following step S400, the
接續於步驟S500，障礙預測模組13係產生模糊關聯矩陣，以表示步驟S400之計算結果。
Following step S500, the
最後於步驟S600，根據模糊關聯矩陣，障礙預測模組13可對障礙電路執行最大最小值合成運算，以預測障礙電路之障礙類型。
Finally, in step S600 , according to the fuzzy correlation matrix, the
綜上所述，本發明之障礙管理系統及其方法係利用網路設備之軌跡濃度分析電路查測資料，以產生設備濃度異常歷史資料，再將設備濃度異常歷史資料與歷史障礙電路比對，以產生關聯模糊矩陣，進而藉由關聯模糊矩陣判斷障礙電路最有可能的障礙類型，故工作人員不需沿著障礙電路的路由逐個查測障礙原因，因而大幅減少障礙排除時間及派工成本，並增加判斷的準確率與效率。 To sum up, the obstacle management system and method of the present invention utilizes the trace concentration analysis circuit of the network equipment to check the data to generate the abnormal historical data of the equipment concentration, and then compares the abnormal historical data of the equipment concentration with the historical obstacle circuit, In order to generate an associative fuzzy matrix, and then use the associative fuzzy matrix to determine the most likely obstacle type of the obstacle circuit, so the staff does not need to check the cause of the obstacle one by one along the route of the obstacle circuit, thus greatly reducing the obstacle removal time and labor cost. And increase the accuracy and efficiency of judgment.
10:網路設備健檢裝置 10: Network equipment health check device
11:事件管理模組 11: Event Management Module
12:分析模組 12: Analysis module
13:障礙預測模組 13: Obstacle prediction module
20:障礙電路申報裝置 20: Obstruction circuit reporting device
30:電路查測客服裝置 30: Circuit test customer service device
41:設備濃度異常歷史資料庫 41: History database of equipment concentration abnormality
42:路由資料庫 42: Routing database
Claims (12)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW109126194A TWI752577B (en) | 2020-08-03 | 2020-08-03 | Obstacle management system and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW109126194A TWI752577B (en) | 2020-08-03 | 2020-08-03 | Obstacle management system and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TWI752577B true TWI752577B (en) | 2022-01-11 |
| TW202207032A TW202207032A (en) | 2022-02-16 |
Family
ID=80809300
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW109126194A TWI752577B (en) | 2020-08-03 | 2020-08-03 | Obstacle management system and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI752577B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030037167A1 (en) * | 1998-12-23 | 2003-02-20 | Nokia Wireless Routers Inc. | Unified routing scheme for ad-hoc internetworking |
| TWI372972B (en) * | 2003-12-29 | 2012-09-21 | Ebay Inc | Apparatus and method to process issue data pertaining to system, and related machine readable storage medium |
| CN103402216A (en) * | 2007-04-13 | 2013-11-20 | Hart通信基金会 | Support for network management and device communication in wireless network |
| TW201515401A (en) * | 2013-10-07 | 2015-04-16 | Chunghwa Telecom Co Ltd | Group circuit breakdown checking system and method thereof |
| CN204689281U (en) * | 2015-05-12 | 2015-10-07 | 广州广日电梯工业有限公司 | For the lift management system that elevator faults is diagnosed in advance |
| CN106022529A (en) * | 2016-05-26 | 2016-10-12 | 中国电力科学研究院 | Abnormal data detection method for power distribution network based on high dimensional random matrix |
| WO2017161403A1 (en) * | 2016-03-23 | 2017-09-28 | Westpac Banking Corporation | A method of and system for anonymising data to facilitate processing of associated transaction data |
| TWI662809B (en) * | 2017-08-08 | 2019-06-11 | 中華電信股份有限公司 | Obstacle location system and maintenance method for image streaming service |
| TWI684139B (en) * | 2018-10-16 | 2020-02-01 | 中華電信股份有限公司 | System and method of learning-based prediction for anomalies within a base station |
-
2020
- 2020-08-03 TW TW109126194A patent/TWI752577B/en active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030037167A1 (en) * | 1998-12-23 | 2003-02-20 | Nokia Wireless Routers Inc. | Unified routing scheme for ad-hoc internetworking |
| TWI372972B (en) * | 2003-12-29 | 2012-09-21 | Ebay Inc | Apparatus and method to process issue data pertaining to system, and related machine readable storage medium |
| CN103402216A (en) * | 2007-04-13 | 2013-11-20 | Hart通信基金会 | Support for network management and device communication in wireless network |
| TW201515401A (en) * | 2013-10-07 | 2015-04-16 | Chunghwa Telecom Co Ltd | Group circuit breakdown checking system and method thereof |
| CN204689281U (en) * | 2015-05-12 | 2015-10-07 | 广州广日电梯工业有限公司 | For the lift management system that elevator faults is diagnosed in advance |
| WO2017161403A1 (en) * | 2016-03-23 | 2017-09-28 | Westpac Banking Corporation | A method of and system for anonymising data to facilitate processing of associated transaction data |
| CN106022529A (en) * | 2016-05-26 | 2016-10-12 | 中国电力科学研究院 | Abnormal data detection method for power distribution network based on high dimensional random matrix |
| TWI662809B (en) * | 2017-08-08 | 2019-06-11 | 中華電信股份有限公司 | Obstacle location system and maintenance method for image streaming service |
| TWI684139B (en) * | 2018-10-16 | 2020-02-01 | 中華電信股份有限公司 | System and method of learning-based prediction for anomalies within a base station |
Also Published As
| Publication number | Publication date |
|---|---|
| TW202207032A (en) | 2022-02-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020238810A1 (en) | Alarm analysis method and related device | |
| CN107196804B (en) | Alarm centralized monitoring system and method for terminal communication access network of power system | |
| CN112398680B (en) | Fault delimiting method and equipment | |
| KR101234326B1 (en) | Distributed traffic analysis | |
| Sasisekharan et al. | Data mining and forecasting in large-scale telecommunication networks | |
| KR100617310B1 (en) | Apparatus for detecting abnormality of traffic in network and method thereof | |
| US20020152185A1 (en) | Method of network modeling and predictive event-correlation in a communication system by the use of contextual fuzzy cognitive maps | |
| WO2007143943A1 (en) | Method, system and network device of centralized maintenance of multiple devices | |
| KR20050048019A (en) | Method for detecting abnormal traffic in network level using statistical analysis | |
| CN108632269A (en) | Detecting method of distributed denial of service attacking based on C4.5 decision Tree algorithms | |
| CN113542017A (en) | Network fault positioning method based on network topology and multiple indexes | |
| CN111953530A (en) | Network equipment monitoring management method and system | |
| TWI752577B (en) | Obstacle management system and method thereof | |
| CN112804250A (en) | LDoS attack detection and mitigation scheme based on integrated learning and peak-finding algorithm | |
| CN108521346B (en) | Method for positioning abnormal nodes of telecommunication bearer network based on terminal data | |
| US8792482B2 (en) | Method and apparatus for correlation of data sources in a Voice over Internet Protocol network | |
| CN112752172B (en) | Optical channel fault diagnosis method and system based on transfer learning | |
| CN117459365A (en) | Fault cause determining method, device, equipment and storage medium | |
| KR20110004101A (en) | Method and apparatus for analyzing abnormal traffic using hierarchical clustering | |
| US7406260B2 (en) | Method and system for network wide fault isolation in an optical network | |
| KR101469283B1 (en) | The enterprise network analysis system and its method | |
| CN117675230A (en) | Knowledge-graph-based oil well data integrity identification method | |
| JPH09307550A (en) | Network system monitoring device | |
| US8194568B1 (en) | Telecommunications network traffic monitoring | |
| CN118075201A (en) | Interactive flow multidimensional monitoring analysis system based on NetFlow big data acquisition technology |