Core Network Data Pipeline for Multi-Source AI Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile communication networks face challenges in efficiently collecting and utilizing data from multiple sources and formats for artificial intelligence/machine-learning applications due to proprietary solutions and complex data collection procedures, limiting flexibility and scalability.
Innovation Solution
A core network data collection entity and machine-learning pipeline are introduced to streamline data collection and preprocessing, allowing data from various sources and formats to be processed and stored efficiently, enabling flexible analytics pipelines and lifecycle management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If proprietary data collection solutions and multiple data formats are used from various data sources, then data availability and coverage are improved, but system complexity and difficulty of integration increase
Solution Approach 1:
The patent introduces a core network data collection entity as an intermediary component that mediates between multiple data sources and the machine learning pipeline. This entity provides a unified interface for collecting data from diverse sources including radio access network, core network, and external sources, converting various data formats into a standardized structure that can be processed by the machine learning pipeline, thereby reducing system complexity while maintaining data availability
Solution Approach 2:
The core network data collection entity is designed with multi-functional capabilities to handle different data collection procedures, formats, and sources through a single unified interface. It can collect data from radio access network, core network functions, and external sources, and transform them into a universal format suitable for machine learning applications, eliminating the need for separate collection mechanisms for each data source
2Quantity of substance
If multiple data collection procedures and formats are implemented, then data comprehensiveness is improved, but ease of operation and data processing difficulty worsen
Solution Approach 1:
The core network data collection entity acts as a mediator that handles the complexity of multiple data collection procedures and formats internally, while presenting a simplified interface to the machine learning pipeline. It automatically transforms diverse data formats into a standardized structure, making data processing easier without compromising comprehensiveness
Solution Approach 2:
The system performs preliminary data collection, filtering, and formatting actions through the core network data collection entity before data reaches the machine learning pipeline. This pre-processing includes collecting data from multiple sources, validating formats, and transforming into standardized structures, thereby reducing the operational burden on subsequent processing stages
3Adaptability or versatility
If proprietary protocols and distributed data collection approaches are used, then adaptability to different suppliers' solutions is improved, but device complexity and integration requirements increase
Solution Approach 1:
The core network data collection entity is designed with universal capabilities to interface with proprietary protocols and solutions from different suppliers. It provides a standardized interface that can adapt to various data formats and collection procedures, enabling integration of multiple supplier solutions without increasing overall system complexity
Solution Approach 2:
The entity serves as an intermediary layer between proprietary supplier solutions and the core network, translating between different protocols and formats. This mediation capability allows the system to maintain adaptability to various supplier implementations while presenting a unified interface to the machine learning pipeline, reducing integration complexity
Data Source
Figure 1~2
Figure 3
AI summary
The invention relates to a method for realizing data collection and/or data analytics regarding artificial intelligence/machine-learning-usable data and/or statistics-usable data within a mobile communication network and from a plurality of data sources and/or using a plurality of data formats and/or collection procedures, wherein the mobile communication network comprises, or is associated with or assigned to a radio access network, and wherein the mobile communication network also comprises a core network, wherein the mobile communication network, especially the core network, comprises a core network data collection entity or functionality and a core network statistics and machine-learning pipeline entity or functionality, the core network data collection entity or functionality providing artificial intelligence/machine-learning and/or statistics usable data base content, especially to the core network statistics and machine-learning pipeline entity or functionality, wherein, in order to realize data collection and/or data analytics regarding artificial intelligence/machine-learning-usable data and/or statistics-usable data within the mobile communication network, the method comprises the following steps: -- in a first step, data collection is performed using a plurality of data sources and/or a plurality of data formats and using a plurality of data collection procedures, -- in a second step, the collected data are preprocessed and provided as artificial intelligence/machine-learning-usable data and/or statistics-usable data in a target store entity or functionality, -- in a third step, the core network statistics and machine-learning pipeline entity or functionality uses the collected and preprocessed data of the target store entity or functionality.