Arrangement and Method for Network Management

• US 20100049676A1
• Filed: 12/07/2006
• Published: 02/25/2010
• Est. Priority Date: 12/07/2006
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Danev, Boris, Devitt, Ann
• Granted Patent: US 8,407,162 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: H04L 41/0233 Object-oriented techniques, for representation of network management data, e.g. common object request broker architecture [CORBA]
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/149 for prediction of maintenance
• CPC Class Codes: H04L 41/16 using machine learning or artificial intelligence
• CPC Class Codes: H04L 41/5009 Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]

Abstract

The present invention relates to an arrangement for network management and adapted to be provided in or associated with a network node to be managed. It comprises, or is in communication with, modeling means adapted to, using substantially non-formal descriptions, model network domain and behavior using formal ontologies comprising inference capabilities by means of an inference engine, thus providing a formal ontology model describing domain and behavior. It also comprises annotating means adapted to add semantic information to the formal domain and behavior ontology model, generating means adapted to, using said formal ontology model and said inference engine, elaborate an algorithm adapted to generate and update a probabilistic causal network graph structure representing the domain and its behavior.

Claims

1. :

An arrangement for network management and adapted to be provided in, associated with or in communication with, a network node to be managed, wherein the arrangement comprises, or is in communication with, modelling means adapted to, using non-formal descriptions, model network domain and behaviour using formal ontologies comprising inference capabilities by means of an inference engine, thus providing a formal ontology model describing domain and behaviour and annotating means adapted to add semantic information to the formal domain and behaviour ontology model, generating means adapted to, using said formal ontology model and said inference engine, elaborate an algorithm adapted to generate and update a probabilistic causal network graph structure representing the domain and its behaviour.

2. :

An arrangement according to claim 1, wherein said modelling means comprise first modelling means with computing means adapted to, from an input substantially non-formal description of a network management service, create a first formal model of the network management service comprising formal ontologies, and second modelling means adapted to, from an input written description of a probabilistic network model, create a formal probabilistic network description comprising formal ontologies to provide a second formal model.

3. :

An arrangement according to claim 1, wherein the annotating means are adapted to receive information about network characteristics to be observed comprising said semantic information, and to annotate the first formal model using the formal ontologies of the second formal model hence providing an annotated formal model describing said characteristics.

4. :

An arrangement according to claim 2, comprising ontology holding means adapted to hold said first formal model, said second formal model and said annotated formal model.

5. :

An arrangement according to claim 4, wherein the inference engine is adapted to implement inference or derivation rules included in the formal ontologies, that the generating means are adapted to communicate with or comprise said inference engine, said generating means further being adapted to implement the ruling of the inference engine on the annotated formal model to build or generate a formal probabilistic graph model.

6. :

An arrangement according to claim 5, wherein the first and second formal models are adapted to be updated in agreement with the annotated formal model identifying characteristics to be observed and in that the generating means are adapted to, in communication with the inference engine, update the formal probabilistic graph model.

7. :

An arrangement according to claim 6, wherein the probabilistic graph model comprises Bayesian Network graph structures.

8. :

An arrangement according to claim 7, wherein the probabilistic network model comprises a Bayesian Network.

9. :

An arrangement according to claim 6, comprising storing means for holding the probabilistic graph structures.

10. :

An arrangement according to claim 2, wherein the first formal model comprises a formal domain and behaviour model.

11. :

An arrangement according to claim 2, wherein the second formal model comprises a probabilistic graph domain model.

12. An arrangement according to claim 7, wherein the generating means are adapted to automatically generate a Bayesian Network (BN) structure, said structure comprising BN nodes, BN causal relationships and transition probabilities.

13. :

An arrangement according to claim 2, wherein the first and second modelling means are adapted to use a modelling language supporting formal ontologies, based on Description Logics (DL), and/or Frame Logics (FL), or based on DL and FL using WSML to provide the formal ontologies.

14. :

An arrangement according to claim 13, wherein the annotating means are adapted to perform the annotation using semantics provided by the formal ontologies of the implemented ontology modelling language comprising one or more of sub-classes, sub-properties and inference rules, at least some of the annotations being generic.

15. :

An arrangement according to claim 1, wherein the arrangement is adapted to be at least partly arranged in a network node or connected to a network node comprising a radio access network node.

16. :

An arrangement according to claim 9, wherein at least the storing means holding the probabilistic network model or graph structures is provided in the network node.

17. :

An arrangement according to claim 16, wherein the generating means, and/or the inference engine and/or the ontology database is/are provided in the network node.

18. :

An arrangement according to claim 16, wherein the generating means and/or the inference engine and/or the ontology database is/are provided externally of a network node to be managed, in a supernode or in one or more information distribution nodes.

19. :

An arrangement according to claim 9, wherein the generating means, the inference engine, the ontology database, the storing means holding the probabilistic graph structure are provided externally of a network node to be managed in a supernode or one or more information distribution nodes.

20. :

An arrangement according to any one of the preceding claims claim 1, wherein the arrangement is adapted to receive current network data to be observed and use the probabilistic graph structures comprising prediction capabilities to detect or forecast node failure, node malfunctioning and to provide information assisting in self-management decisions.

21. :

A method for network management implemented in or in association with a network node, comprising the steps of;

modelling network domain and behaviour models using formal ontologies comprising inference capabilities to provide a formal ontology model domain and behaviour description;

adding, in an annotating step, semantic information to the formal domain and behaviour ontology model; and

elaborating an algorithm adapted to generate and update a probabilistic causal graph structure using said formal ontology model and said inference capabilities.

22. :

A method according to claim 21, wherein the modelling step comprises;

receiving a substantially non-formal description of a network management service;

creating therefrom a first formal model of the network management service comprising formal ontologies;

receiving a written description of a probabilistic network model; and

creating therefrom a second formal model of the probabilistic network model comprising formal ontologies.

23. :

A method according to claim 22, wherein the annotating step comprises;

achieving information about network characteristics to be observed comprising said semantic information; and

annotating the first formal model using the formal ontologies of the second formal model to provide an annotated formal model describing said characteristics.

24. :

A method according to claim 23, comprising the step of;

storing said first formal model, said second formal model and said annotated formal model in ontology holding means.

25. :

A method according to claim 24, wherein the algorithm elaboration step comprises;

applying inference or derivation rules of the inference engine on the annotated formal model to elaborate an algorithm for building or generating the formal probabilistic graph structure model.

26. :

A method according to claim 21, comprising the step of;

updating, automatically, the formal probabilistic causal graph structure model depending on the observed characteristics.

27. :

A method according to claim 21, comprising the step of;

storing the probabilistic graph structures in a database, in the network node or in an external information distribution node.

28. A method according to claim 21, wherein the modelling step comprises:

using a modelling language supporting formal ontologies based on Description Logic, Frame Logic or Description Logic and Frame Logic to provide the formal ontologies.

29. :

A method according to claim 28, wherein the annotating step comprises;

using semantics provided by the formal ontologies of the implemented ontology modelling language, at least some of the annotations being generic.

30. :

A method according to claim 21, wherein the method used to manage a network node.

31. :

A method according to claim 21, comprising the steps of;

receiving network events on a network interface of the network node, in the database holding the probabilistic graph structures; and

using the probabilistic causal graph structures to predict behaviour, and to make self-management decisions based on the updated graph structures.