A basic concept of Expanse is the concept of data-chains or data-processing-chains (somewhat similar to service chains). Expanse focuses on end-to-end chains on which data is acquired, processed and used across the different domains. We would like to highlight two such chains as follows:
Chain 1: Optimization of the 5G system behaviour
- use of the satellite network within the 5G system to provide additional monitoring information for a better management of the overall system;
- use of insight from geo-observation data to optimize the usage of the 5G system – geo-observation data can be used by the 5G system management in order to be able to optimize its behaviour.
Chain 2: optimization of the applications and of the applications’ behaviour by combining status information available in the convergent 5G satellite-terrestrial environment with geo-observation data.
The envisaged high-level architecture of the combined big data / EO data and telco management system is as follows:
In our approach, we postulate the convergence of the different administrative and technical domains at the data level, including:
- Big Data Repositories Domain
- Telco Domain
- Terrestrial 5G System
- Converged SatCom-Terrestrial 5G System
- Edge-Central System
- Application and services domain
Expanse considers data from multiple sources, including earth observation, network management and devices and sensor data as inputs for a data processing layer which in turn enables the increase of the network adaptation capabilities as well as of the application level data enhancement.
A high-level overview of the role of Expanse
The data processing layer includes the functionality related to the:
- Data discovery and access – providing the mechanisms for how the different sources of data can be found and how the data can be accessed;
- Efficient and secure data exchange – especially concentrating on the reduction of network resources consumption as well as on the secure end-to-end connectivity;
- Data pre-processing – addressing the need of coherence in data retrieved from the different data sources;
- Data Insight generation – addressing the opportunities brought by different statistics and machine learning mechanisms to generate additional knowledge from data.
In Expanse the different aspects are addressed only in an initial form while implementing and validating a limited set of end-to-end use cases. Our approach helps to identify additional aspects warranting further study, but at the same time already provide initial results concerning real-life use cases.