Life cycle management

Patent metadata

Field	Value
Publication number	US-12229165-B2
Application number	US-201917624095-A
Country	US
Kind code	B2
Filing date	Jul 12, 2019
Priority date	Jul 12, 2019
Publication date	Feb 18, 2025
Grant date	Feb 18, 2025

Abstract

A method is provided for identifying operating conditions of a system. Input data relating to operation of the system is applied to a multi-class model for classification, where the multi-class model is configured for classifying the data into one of a plurality of predefined classes, and each class corresponds to a respective operating condition of the system. A confidence level of the classification by the multi-class model is determined. If the confidence level is below a threshold confidence level, the input data is applied to a plurality of binary models, where each binary model is configured for determining whether the data is or is not in a respective one of the predefined classes. If the plurality of binary models determine that the data is not in any of the respective predefined classes, the data can be taken into consideration when updating the multi-class model.

First claim

The invention claimed is:

1. A method for identifying operating conditions of a system comprising a telecommunications system, the method comprising:
applying input data, relating to operation of the telecommunications system, to a multi-class model to classify the input data into one of a plurality of predefined classes each corresponding to a faulty or non-faulty operating condition of the telecommunications system;
determining a confidence level of the classification by the multi-class model;
if the confidence level is below a threshold confidence level, applying the input data to a plurality of binary models, each binary model being configured for determining whether the input data is or is not in a respective one of the predefined classes; and
if the plurality of binary models determine that the input data is not in any of said respective predefined classes, taking said input data into consideration when updating the multi-class model.

2. The method according to claim 1 , wherein the telecommunications system comprises at least a part of a telecommunications network.

3. The method according to claim 1 , wherein taking said data into consideration when updating the multi-class model comprises:
performing a clustering operation on a plurality of input data that have been determined by the plurality of binary models not to be in any of said respective predefined classes; and
if a cluster is found, adding an additional predefined class into the multi-class model; and
generating a new binary model configured for determining whether the data is or is not in the additional predefined class.

4. The method according to claim 1 , wherein applying the input data to a plurality of binary models comprises applying the input data to a respective binary model corresponding to each of said plurality of predefined classes.

5. The method according to claim 1 , comprising:
receiving input data from one example of the telecommunications system, and updating the multi-class model and the binary models used in connection with said one example of the telecommunications system, or
receiving input data from a plurality of examples of the telecommunications system, and updating the multi-class model and the binary models that are used in connection with said plurality of examples of the telecommunications system.

6. The method according to claim 1 , comprising:
determining a measure of a proportion of the input data that is determined not to be in any of said respective predefined classes; and
if said measure is below a classification threshold, retraining the multi-class model and the binary models.

7. The method according to claim 1 , comprising:
performing further training of the multi-class model and the binary models, using data for which said confidence level is above the threshold confidence level.

8. The method according to claim 1 , comprising:
receiving raw data; and
normalising the raw data to form said input data.

9. The method according to claim 1 , comprising:
if the plurality of binary models determine that the data is in a plurality of said respective predefined classes, determining that a corresponding plurality of said respective operating condition of the telecommunications system have arisen.

10. A computer program comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out a method according to claim 1 .

11. A computer program product comprising non transitory computer readable media having stored thereon a computer program as claimed in claim 10 .

12. Apparatus for identifying operating conditions of a system comprising a telecommunications system, the apparatus comprising a processor and a memory, the memory containing instructions executable by the processor such that the apparatus is operable to:
apply input data, relating to operation of the telecommunications system, to a multi-class model to classify the input data into one of a plurality of predefined classes each corresponding to a faulty or non-faulty operating condition of the telecommunications system;
determine a confidence level of the classification by the multi-class model;
if the confidence level is below a threshold confidence level, apply the input data to a plurality of binary models, each binary model being configured for determining whether the input data is or is not in a respective one of the predefined classes; and
if the plurality of binary models determine that the input data is not in any of said respective predefined classes, take said input data into consideration when updating the multi-class model.

13. Apparatus as claimed in claim 12 , wherein the operation to take said data into consideration when updating the multi-class model comprises:
performing a clustering operation on a plurality of input data that have been determined by the plurality of binary models not to be in any of said respective predefined classes; and
if a cluster is found, adding an additional predefined class into the multi-class model; and
generating a new binary model configured for determining whether the data is or is not in the additional predefined class.

14. Apparatus as claimed in claim 12 , wherein the telecommunications system comprises at least a part of a telecommunications network.

15. Apparatus as claimed in claim 12 , wherein applying the input data to a plurality of binary models comprises applying the input data to a respective binary model corresponding to each of said plurality of predefined classes.

16. Apparatus as claimed in claim 12 , further operable to:
receive input data from one example of the telecommunications system, and updating the multi-class model and the binary models used in connection with said one example of the telecommunications system, or
receive input data from a plurality of examples of the telecommunications system, and updating the multi-class model and the binary models that are used in connection with said plurality of examples of the telecommunications system.

17. Apparatus as claimed in claim 12 , further operable to:
determine a measure of a proportion of the input data that is determined not to be in any of said respective predefined classes; and
if said measure is below a classification threshold, retrain the multi-class model and the binary models.

18. Apparatus as claimed in claim 12 , further operable to:
perform further training of the multi-class model and the binary models, using data for which said confidence level is above the threshold confidence level.

19. Apparatus as claimed in claim 12 , further operable to:
receive raw data; and
normalise the raw data to form said input data.

20. Apparatus as claimed in claim 12 , further operable to:
if the plurality of binary models determine that the data is in a plurality of said respective predefined classes, determine that a corresponding plurality of said respective operating condition of the telecommunications system have arisen.

21. The method according to claim 1 , wherein the faulty or non-faulty operating condition of the telecommunications system comprises at least one of cell load, downlink utilization, uplink utilization, processor load, physical uplink control channel performance, random access channel accessibility, signalling load, mobility, coverage, and interference.

Assignees

• Ericsson Telefon Ab L M

Inventors

• Frank Philipp
• Eng Chin Lam
• Martin Cuerdo Raul
• Ho Mitchell
• Ng Chee Wai

Classifications

• H04L41/16

  using machine learning or artificial intelligence · CPC title

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• H04L41/082

  the condition being updates or upgrades of network functionality · CPC title

• H04L41/0895

  Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements · CPC title

• H04L41/0823

  characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability (for optimising operational conditions of wireless networks H04W24/02) · CPC title