Network analysis system

• US 8,918,345 B2
• Filed: 03/31/2010
• Issued: 12/23/2014
• Est. Priority Date: 03/31/2009
• Status: Active Grant

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Kings, Nicholas J, Davies, Nicholas J, Fortuna, Blaz, Grobelnik, Marko, Wittgreffe, John P
• Primary Examiner(s): Starks, Wilbert L
• Assistant Examiner(s): Seck, Ababacar
• Application Number:
  US13/262,069
• Publication Number:
  US 20120136816A1
• Time in Patent Office: 1,728 Days
• Field of Search: 706/12
• US Class Current: 706/12
• CPC Class Codes: H04L 41/0681 Configuration of triggering conditions
• CPC Class Codes: H04L 41/147 for predicting network behaviour

Abstract

The present invention provides a method of operating a network comprising the steps of: analyzing a first datastore comprising data representing historical network performance; creating or more indices within the first datastore; creating one or more probability networks in accordance with one or more of the created indices; determining from the one or more probability networks a conditional probability associated with an alarm event; and'"'"' if the conditional probability determined is less than a threshold value, disregarding the associated alarm event; or if the conditional probability determined is greater than a threshold value, using the associated alarm event in conjunction with other historical network data to predict future alarm events.

Claims

1. A method of operating a network comprising a plurality of network elements, the method comprising the steps of:

a) analysing, via a computer comprising one or more processors, a first datastore comprising data representing historical network performance;

b) creating within the first datastore;

i) a first index relating to the probability that a network element generates an alarm event, said probability being proportional to the number of alarm events issued by said network element;

ii) a second index relating to the probability that a first network element and a second network element both generate an alarm event within a predetermined time period, said probability being proportional to the number of alarm event pairs where an alarm event is issued by said first network element and an alarm event is issued by said second network element; and

iii) a third index relating to the probability that a second network element generates an alarm event given that a first network element previously generated an alarm event within a predetermined time period, said probability being proportional to the number of alarm events where said first network element issues a first alarm event and then said second network element issues a second alarm event;

c) creating one or more probability networks in accordance with the indices created in step b);

d) determining from the one or more probability networks a conditional probability associated with an alarm event; and

e) if the conditional probability determined in step d) is less than a threshold value, disregarding the associated alarm event;

or

f) if the conditional probability determined in step d) is greater than a threshold value, using the associated alarm event in conjunction with other historical network data to predict future alarm events.

2. The method according to claim 1 further comprising

g) identifying one or more root cause alarm event from a set of such events.

3. The method according to claim 1, wherein

the or each probability networks comprise a plurality of nodes, each of the nodes corresponding to a network element comprised within the network.

4. The method according to claim 3, wherein

each of the probability network nodes are weighted in accordance with the probability that the corresponding network element generates an alarm event.

5. The method according to claim 1, wherein

the or each probability networks comprise a plurality of undirected edges, each of the undirected edges connecting a respective first node and a respective second node.

6. The method according to claim 5, wherein

each of the plurality of undirected edges is weighted in accordance with the probability that the network element associated with the respective first node and the network element associated with the respective second node both generate an alarm event within a predetermined period of time.

7. The method according to claim 1, wherein

the or each probability networks comprise a plurality of directed edges, each of the directed edges connecting a respective first node and a respective second node.

8. The method according to claim 7, wherein

each of the plurality of directed edges is weighted in accordance with the probability that:

i) the network element associated with the respective first node generates a first alarm event;

ii) the network element associated with the respective second node generates a second alarm event;

iii) the second alarm event being generated after the first alarm event and within a predetermined period of time after the generation of the first alarm event.

9. A non-transitory computer-readable storage medium having stored therein a computer executable code configured to perform the method according to claim 1.

10. An apparatus comprising a central processing unit, volatile data storage and non volatile data storage, the apparatus being configured to perform the method according to claim 1.