System and method for fault prediction in home network

• US 8,255,351 B2
• Filed: 04/27/2009
• Issued: 08/28/2012
• Est. Priority Date: 09/04/2008
• Status: Expired due to Fees

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: Lee, Chang Eun, Park, Jun Hee, Moon, Kyeong Deok
• Primary Examiner(s): Fernandez Rivas, Omar
• Application Number:
  US12/430,166
• Publication Number:
  US 20100057649A1
• Time in Patent Office: 1,219 Days
• Field of Search: None
• US Class Current: 706/47
• CPC Class Codes: H04L 41/0631 using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/16 using machine learning or artificial intelligence

Abstract

A system for fault prediction in a home network includes: a context generator for generating context information based on status data collected in real time about components of the home network; a specification interpreter for generating knowledge rules for fault detection by using specifications of the components of the home network; a context analyzer for analyzing if the context information meet the knowledge rules to classify the context information into normal situation contexts and abnormal situation contexts; a context pattern learner for generating new knowledge rules based on the abnormal situation contexts and fault rules corresponding to the abnormal situation contexts; a knowledge rule database for storing and managing the knowledge rules and the new knowledge rules; and a fault predictor for analyzing a correlation between the knowledge rules or the new knowledge rules and the generated context information, thereby predicting faults to be generated.

Claims

1. A system for fault prediction in a home network, the system comprising:

a processor and a memory, the memory haying stored thereon;

a context generator for generating context information based on status data collected in real time about components of the home network;

a specification interpreter for generating knowledge rules for fault detection by using specifications of the components of the home network;

a context analyzer for analyzing if the generated context information meet the generated knowledge rules to classify the context information into normal situation contexts and abnormal situation contexts;

a context pattern learner for generating new knowledge rules based on the abnormal situation contexts and fault rules corresponding to the abnormal situation contexts;

a knowledge rule database for storing and managing the knowledge rules and the new knowledge rules; and

a fault predictor for analyzing a correlation between the knowledge rules or the new knowledge rules and the generated context information, thereby predicting faults to be generated.

2. The system of claim 1, wherein the context generator comprises:

a status data collector for collecting the status data in real time; and

a context mapper for establishing a correlation between the status data, and extracting features of the status data by digitizing and normalizing the status data to generate the context information.

3. The system of claim 2, wherein the status data is digitized and normalized into the context information based on 4W1H (when, where, who, what, and how).

4. The system of claim 1, wherein the specification interpreter comprises:

a resource interpreter for interpreting the specifications and giving the specifications semantics;

a relation establisher for establishing a relation between the semantics; and

a diagnosis rule generator for generating the knowledge rules based on functions that affect the establishment of the relation between the specifications of the components.

5. The system of claim 1, wherein the context analyzer comprises:

a context interpreter for interpreting the context information;

a condition checker for analyzing if the interpretation results of the context information meet the knowledge rules; and

a context classifier for classifying the context information into the normal situation contexts or the abnormal situation contexts depending on the analysis results from the condition checker, and providing the context pattern learner with the abnormal situation contexts and the fault rules corresponding to the abnormal situation contexts.

6. The system of claim 1, wherein the context pattern learner learns a correlation between the abnormal situation contexts and the fault rules by using a neural network, and creates new context patterns corresponding to new fault causes based on the correlation to generate the new knowledge rules.

7. The system of claim 6, wherein the context pattern learner generates a correlation probability table corresponding to the new knowledge rules.

8. The system of claim 7, wherein the fault predictor analyzes the correlation between the correlation probability table corresponding to the new knowledge rules and the context information collected in real time.

9. The system of claim 1, further comprising a context repository for storing and managing the context information generated by the context generator as well as the correlation and the features corresponding to the context information.

10. A method for fault prediction in a home network, comprising:

generating context information based on status data collected in real time about components of the home network;

generating knowledge rules for fault detection by using specifications of the components of the home network;

analyzing if the generated context information meet the generated knowledge rules to classify the context information into normal situation contexts and abnormal situation contexts;

generating new knowledge rules based on the abnormal situation contexts and fault rules corresponding to the abnormal situation contexts; and

analyzing a correlation between the new knowledge rules and the generated context information thereby predicting faults to be generated.

11. The method of claim 10, wherein said generating context information comprises:

collecting the status data in real time; and

digitizing and normalizing the status data to establish a correlation between the data, and extracting features of the data to generate the context information.

12. The method of claim 11, wherein the status data is digitized and normalized into the context information based on 4W1H (when, where, who, what, and how).

13. The method of claim 12, wherein the context information includes the correlation and the features, and is stored and managed in a context repository.

14. The method of claim 10, wherein said generating knowledge rules comprises:

interpreting the specifications and giving the specifications semantics;

establishing a relation between the semantics; and

generating the knowledge rules based on functions that affect the establishment of the relation between the specifications of the components.

15. The method of claim 14, wherein the knowledge rules are stored and managed in a knowledge rule database.

16. The method of claim 10, wherein said analyzing if the generated context information meet the generated knowledge rules comprises:

interpreting the context information; and

analyzing if the interpretation results of the context information meet the knowledge rules.

17. The method of claim 16, wherein said generating new knowledge rules comprises:

classifying the context information into normal situation contexts or abnormal situation contexts depending on the analysis results; and

providing the abnormal situation context and the fault rules to generate the new knowledge rules.

18. The method of claim 10, wherein the new knowledge rules are generated by learning the correlation between the abnormal situation context and the fault rules by using a neural network structure, and creating new context patterns corresponding to new fault causes based on the correlation.

19. The method of claim 18, wherein the new knowledge rules are organized as a corresponding correlation probability table.