Logical inventory database integration into alarm correlations discovery environment. (16th November 2010)

• Record Type: Journal Article
• Authors: Jukic, O.; Kunstic, M.
• Is Part Of: International journal of intelligent information and database systems. Volume 4:Number 6(2010)
• Page Start: 643
• Page End: 655
• Publication Date: 2010-11-16
• Subjects: network alarms -- network problems -- alarm correlation -- alarm filtration -- Apriori algorithm -- correlation rules -- unsolicited messages -- notifications -- network malfunction -- logical inventory databases -- networks; Database management -- Computer programs -- Periodicals; Information retrieval -- Computer programs -- Periodicals; Information storage and retrieval systems -- Computer programs -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligent agents (Computer software) -- Periodicals; 006.33
• Languages: English
• ISSNs: 1751-5858

Abstract

Network elements from telecommunications network have capability to generate unsolicited messages known as notifications. In the case of network resource malfunctioning, notification, called alarm, is carrying details about malfunction. When general network problem occurs, it is represented as a sequence of alarms coming from one or more different network elements. Typically, alarms are processed by network operators, requiring response action within reasonable time interval. If network operator is overloaded with huge number of alarms, time needed for network problem recognition may increase. The worst case scenario – problem will be detected after occupation of call centre by customer calls. Hence, it is necessary to recognise critical network problems automatically, reducing and correlating incoming alarms. This paper describes architecture of alarm basic correlation discovery environment (ABCDE). It is aimed for correlation rules discovery for both types of correlations described in paper: low-level ('smart' filtrations) and high-level (recognition of typical alarm sequences) correlations. Potential usage of mathematical Apriori algorithm is presented, together with integration of logical inventory database, used for including network structure knowledge in correlation process. Finally, some experimental results are presented.