Method and apparatus for analyzing alarms coming from a communications network

• US 20030088669A1
• Filed: 11/01/2002
• Published: 05/08/2003
• Est. Priority Date: 11/08/2001
• Status: Active Grant

• Current Assignee: WSOU Investments, LLC (WSOU Holdings, LLC)
• Original Assignee: Alcatel SA (Nokia Corporation)
• Inventors: Toure, Famory, Marilly, Emmanuel
• Granted Patent: US 7,103,801 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: H04L 41/16 using machine learning or artificial intelligence

Abstract

The invention enables alarms (MA) coming from a network (4), such as a communications network, to be analyzed by a step of transforming the detected alarms (MA) into a signal (V(t)) which expresses variation in time of a numerical value representative of all of the detected alarms. This transformation can be implemented using a learning system, in particular a neural network (6). The signal can then be analyzed using various techniques, in particular time/frequency analysis techniques, in order to make a diagnosis.

Claims

1. / A method of analyzing alarms (MA) coming from a network (4), the method being characterized in that it includes a step of transforming the detected alarms (MA) into a signal (V(t)) which expresses the variation in time of a digital value representative of all of the detected alarms.

2. / A method according to claim 1, characterized in that the transformation step is performed by means of a learning system (6).

3. / A method according to claim 1, characterized in that the transformation step is performed by means of a neural network (6).

4. / A method according to claim 2, characterized in that each alarm is made up of data fields (16-28), the inputs (8) of the learning system (6) are subdivided into groups (8a-8d), with each group being specifically dedicated to a respective field that is taken into account.

5. / A method according to claim 4, characterized in that a particular input (8) of a group of inputs (8a-8d) is selected as a function of the content of the field to which said group is dedicated, with each input of a group corresponding to a specific content of a field.

6. / A method according to claim 5, characterized in that at least one group of inputs (8a-8d) is provided that is dedicated to a respective one of the following alarm fields MA:

a first level of alarm localization, e.g. a type of network equipment (20);

a second level of alarm localization, e.g. a piece of equipment (24);

a reason for an alarm (26); and

an indication of priority (22).

7. / A method according to claim 4, characterized in that a detected alarm (MA) is applied to the input of the learning system (6) by selectively setting one input (8) in each group (8a-8d) of inputs into a determined logic state.

8. / A method according to claims 3 and 4, characterized in that each input (8) corresponds to an input neurone of a neural network, the input neurones being subdivided into groups (8a-8d) of inputs.

9. / A method according to claim 2, characterized in that a prior stage of training the learning system (6) is performed using steps of correlating the input (6a) and the output (6c) of the system on the basis of examples of alarm messages (MA).

10. / A method according to claim 2, characterized in that said signal (V(t)) is obtained in the form of the value of a binary word expressed by the output neurones (10) of the learning system (6).

11. / A method according to claim 1, characterized in that it further comprises a step of processing (14) the signal (V(t)) delivered by the transformation step, on the basis of which usable information is extracted.

12. / A method according to claim 11, characterized in that said processing comprises time/frequency analysis of said signal (V(t)) in order to make a diagnosis.

13. / A method according to claim 12, characterized in that said time/frequency analysis is performed by a technique of the Cohen class.

14. / A method according to claim 13, characterized in that the Wigner Ville distribution is used to produce a bilinear distribution of the energy of said signal (V(t)).

15. / A method according to claim 12, characterized in that peaks in the distribution obtained by the time/frequency analysis are localized, and respective vectors are extracted therefrom which correspond to identified alarm signatures.

16. / A method according to claim 15, characterized in that said distribution is interpreted in terms of times and frequencies on a linear scale.

17. / A method according to claim 11, characterized in that said processing comprises time/scale analysis of said treatments (V(t)) in order to make a diagnosis.

18. / A method according to claim 17, characterized in that wavelet transforms are produced to change said signal (V(t)) into coefficients for use in making the diagnosis, e.g. by means of a scalogram.

19. / A method according to claim 11, characterized in that said processing comprises analysis of the frequency only of said signal (V(t)), e.g. Fourier analysis, in order to make a diagnosis on the basis of the resulting spectrum.

20. / A method according to claim 19, characterized in that peaks in the spectrum are localized, and respective vectors are extracted therefrom corresponding to identified alarm signatures.

23. / Apparatus for analyzing alarms coming from a network (4), the apparatus being characterized in that it comprises transformation means for transforming detected alarms (MA) into a signal (V(t)) which expresses variation in time of a numerical value representative of all of the detected alarms, the apparatus being suitable for implementing the method according to claim 1.

21. / Apparatus for analyzing alarms coming from a network (4), the apparatus being characterized in that it comprises transformation means for transforming detected alarms (MA) into a signal (V(t)) which expresses variation in time of a numerical value representative of all of the detected alarms.

22. / Apparatus according to claim 21, characterized in that it further comprises means (14) for processing the signal (V(t)) delivered by the transformation means, with usable information being extracted therefrom.