Entities and methods for enabling control of a usage of collected data in multiple analytics phases in communication networks

Patent metadata

Field	Value
Publication number	US-12408065-B2
Application number	US-202418762366-A
Country	US
Kind code	B2
Filing date	Jul 2, 2024
Priority date	Jul 30, 2020
Publication date	Sep 2, 2025
Grant date	Sep 2, 2025

Abstract

A first entity for a communication network is provided. The first entity is configured to receive a feature mapping indication from a second entity of the communication network. The feature mapping indication comprises characteristics of a relationship between a set of data samples and properties of data in the data samples of the set of data samples for an analytics model at an analytics stage, wherein the feature mapping indication comprises a request for a feature mapping data structure. The feature mapping data structure includes a second set of data samples based on the relationship between the set of data samples and the properties of the data in the data samples of the set of data samples for use with the analytics model at the analytics stage for an analytics consumer.

First claim

What is claimed is:

1. A first entity for a communication network, the first entity comprising a processor and a memory having processor-executable instructions stored thereon, wherein the processor executes the instructions which cause the first entity to:
receive a feature mapping indication from a second entity of the communication network,
wherein the feature mapping indication comprises characteristics of a relationship between a set of data samples and properties of data in the data samples of the set of data samples for an analytics model at an analytics stage,
wherein the feature mapping indication comprises a request for a feature mapping data structure, which defines a second set of data samples based on the relationship between the set of data samples and the properties of the data in the data samples of the set of data samples for use with the analytics model at the analytics stage for an analytics consumer; and

generate the feature mapping data structure for use with the analytics model at the analytics stage based on at least one of the feature mapping indication received from the second entity and a feature provisioning policy, wherein the feature provisioning policy defines a set of at least one of properties and processes to be applied to the data in the data samples of the set of data samples;
wherein the feature provisioning policy comprises any one or more of the following information fields: an identification of an analytics consumer of a feature, a network slice identification, a network operator identification, a type of feature mapping, at least one of an allowed and restricted feature selection technique, at least one of an allowed and restricted feature type, at least one of an allowed and restricted area of interest, at least one of an allowed and restricted type of analytics models, an aggregation level per type of feature, and an anonymization rule; and
send the feature mapping data structure back to the second entity.

2. The first entity as claimed in claim 1 , wherein the processor further executes the instructions which cause the first entity to:
obtain one or more feature mapping data structures for use with the analytics model at the analytics stage based on one or more further feature mapping data structures from one or more other entities of the communication network.

3. The first entity as claimed in claim 1 , wherein the analytics stage comprises one or more of an inference stage, a training stage, a testing stage, a validation stage, an offline data collection, and an online data collection.

4. The first entity as claimed in claim 1 , wherein the feature mapping indication comprises a direct request for generating the feature mapping data structure, wherein the direct request comprises at least one of an information field defining a request for creating the feature mapping data structure and a flag indicating a requirement for immediate retrieval of the feature mapping data structure.

5. The first entity as claimed in claim 1 , wherein the feature mapping indication comprises a direct request for retrieval of the feature mapping data structure according to a set of criteria comprising one or more filters enabling a selection of the data samples to be associated with the feature mapping data structure.

6. The first entity as claimed in claim 1 , wherein the feature mapping indication comprises an indirect request for the feature mapping data structure related to a request for an analytics output from the analytics model, and the first entity is further configured to generate or obtain the feature mapping data structure for supporting inference using the analytics model for the analytics output.

7. The first entity as claimed in claim 1 , wherein the feature mapping indication comprises an indirect request for the feature mapping data structure related to a request for model training, and the first entity is further configured to generate or obtain the feature mapping data structure for supporting the model training.

8. The first entity as claimed in claim 1 , wherein the request for the feature mapping data structure comprises at least one of:
data representing a flag for indicating immediate retrieval of the feature mapping data structure, and
a data collection mode.

9. The first entity as claimed in claim 1 , wherein the processor further executes the instructions which cause the first entity to:
provide the feature mapping data structure to another entity of the communication network in response to the feature mapping indication.

10. The first entity as claimed in claim 1 , wherein the processor further executes the instructions which cause the first entity to:
access one or more data repositories, or is logically co-located with a data repository, and
wherein the data repository comprises the data samples of the set of data samples.

11. The first entity as claimed in claim 1 , wherein the processor further executes the instructions which cause the first entity to:
provide at least one of an identification for the feature mapping data structure obtained in accordance with the feature mapping indication, and an identification of a storage repository of the communication network, and
wherein the storage repository comprises the feature mapping data structure.

12. The first entity as claimed in claim 1 , wherein a direct request for creation of the feature mapping data structure comprises any one or more of the following information fields:
an analytics type identification, a data collection mode, a model type identification, a model version identification, the analytics stage, an analytics consumer identification, a type of data, a type of feature, an aggregation level per type of feature, a statistical property of a data sample, a statistical method to be applied, an area of interest, a target of analytics reporting, analytics filter information, an interval of time for sample selection, at least one of a minimum and maximum number of samples for sample selection, a network slice identification, a network operator identification, a deadline for generating and providing the feature mapping data structure, and a data collection mechanism to be used for the retrieval of at least one of raw data and pre-processed data.

13. The first entity as claimed in claim 1 , wherein the feature mapping data structure comprises any one or more of the following information fields: a feature mapping identification, an identification of the analytics consumer related to the feature mapping data structure, an analytics type identification, a type of feature mapping, a model type for the analytics identification, a model version for each model type, a model stage, a statistical property of data samples, a type of data, a feature type, a feature sample value, a reference for the feature sample identification, a reference for an entity storing the feature sample, a reference for an entity storing the data samples, and a timestamp of at least one of a created/updated feature and data samples.

14. A method applied to a communication network, the method comprising:
receiving, at a first entity, a feature mapping indication from a second entity of the communication network,
wherein the feature mapping indication defining characteristics of a relationship between a set of data samples and properties of data in the data samples of the set of data samples for an analytics model at an analytics stage,
wherein the feature mapping indication comprises a request for a feature mapping data structure, the feature mapping data structure defining a second set of data samples based on the relationship between the set of data s

Assignees

• Huawei Tech Co Ltd

Inventors

• Marquezan Clarissa
• Wei Qing
• Xin Yang
• Wu Xiaobo
• Chong Weiwei

Classifications

• H04L41/14

  Network analysis or design · CPC title

• H04W24/04

  Arrangements for maintaining operational condition · CPC title

• H04W24/02

  Arrangements for optimising operational condition · CPC title

• H04L41/16

  using machine learning or artificial intelligence · CPC title

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title