Skip to main content
Defence Industry and Space
Call for tendersClosed

IPS – Ionosphere Prediction Service


Publication date
Deadline date


The objective of the Ionosphere Prediction Service (IPS) project was to design, develop and operate a web-based prototype platform to monitor and predict the behaviour of the ionosphere and its potential effects on the performance of global navigation satellite system (GNSS)-based applications.

Contract Number: Call for Tenders No 434/PP/GRO/RCH/15/8381

Project Segment: Horizon 2020, Galileo Mission and Services evolution

Duration: 24 months

Budget: € 669 000

Project Partners: Telespazio (IT); Telespazio Vega (Germany); Nottingham Scientific Ltd (NSL, UK), The University of Nottingham (UK), the University of Rome Tor Vergata (Italy) and the Istituto Nazionale di Geofisica e Vulcanologia (INGV, Italy)

Project Coordinator:
Filippo Rodriguez, Telespazio Italy
Filippo [dot] Rodriguezattelespazio [dot] com (Filippo[dot]Rodriguez[at]telespazio[dot]com)

Project Manager:
Eric Guyader, European Commission
Eric [dot] Guyaderatec [dot] europa [dot] eu (Eric[dot]Guyader[at]ec[dot]europa[dot]eu)


Ionosphere effects are a well-known source of disturbances on GNSS. In some cases, the impact is so great that it is important to anticipate upcoming disturbances and put in place mitigation measures. This is particularly relevant to GNSS-based applications sensitive to a sudden degradation of performance, such as those related to aviation, high accuracy (construction works, civil engineering), and synchronised networks (energy grids, banks, telecom).

Ionosphere now-casting and forecasting can help mitigate the impact of significant space weather-related geophysical events on the final performance of specific GNSS-based applications.

IPS is an initiative of the European Commission in the framework of the Galileo Programme.

It is a prototype of a service for the monitoring and prediction of Ionosphere effects on final GNSS users. The web-based IPS platform monitors and forecasts solar and ionospheric activity and its well-known effect on GNSS signals and on the final performance of user applications. The predictions are delivered for both ionosphere-related parameters and GNSS performance indicators, over two geometric scales (European region and global) and three time scales (now-cast, 30 minutes and 24 hours ahead).

Project objectives

  • Design, develop, deliver and operate a service prototype platform
  • Monitor ionospheric behaviour
  • Predict potential effects on the performance of GNSS-based applications
  • Translate the prediction and forecasting of the ionosphere into results and user-devoted metrics

The project is structured along two main activity streams: the design, development, operation and maintenance of the web-based prototype; and scientific research activity at the base of the implemented algorithms.


The project delivered predictions for:

  • solar flares and other solar related events
  • ionosphere-related measurements such as total electron content (TEC)
  • high-accuracy related products
  • aviation-related GNSS products

The project developed and delivered a prototype for predicting ionosphere behaviour, since GNSS systems can be disturbed by such space weather-related events. In this sense, the project was conceived around the idea of offering operators of GNSS-based applications the capacity to anticipate and plan for these events.

Several similar web-based platforms exist today, but they mostly deliver monitoring products rather than forecasting products. This means that the user must  predict the impact themselves. Moreover, most of the services do not provide products dedicated to users and operators of infrastructure; they are more suited to scientific usage.

Another distinguishing characteristic of IPS is that it was conceived as a centralised and coherent set of different processing facilities. All processors are located within the same stand-alone platform and synchronised with each other, enabling a greater capacity for expansion and replication in other places. This is a novel approach as existing platforms use distributed architectures among several laboratories, which is less practical to handle, evolve and replicate.

The potential to tailor the service to the regional level is particularly innovative, as it may be a better fit for users more interested in monitoring a given area (airport, surveyed field, etc.) than the entire planet. In fact, the platform is built to deliver products at any scale, depending on the data collected. As a result, there is huge potential for deployment at local and regional level.

One other key element of the project was the elaboration of retro-validation products, i.e. products immediately showing how the forecast compared to the reality. This never-before-seen element was central to the prototype, and proved to be essential in assessing the accuracy of predictions. It ensures transparency with the user by providing an indication of the quality of predictions. This function is essential as it increases confidence in the products. 

Some EGNOS disruption events occurred during the project period, the cause of which was not immediately identified. IPS was used to retrieve ionosphere activity during the period, and could confirm the absence of any critical ionosphere activity that could have caused the disruption. After a few days, it was confirmed that the disruption was caused by an internal anomaly. IPS also has a role in real-time monitoring of the ionosphere state, and it can later be used in the elimination of hypothesis during investigations.

Many articles were submitted during the project’s execution, and several conferences on the topics were held. These included discussions on the prototype. More information on the project’s results are available in the final report (PDF, 738 KB).


This project allowed for the development of novel approaches in both a scientific field (ionosphere and sun-earth interaction) and in engineering (forecasting methods).

One of the project’s greatest challenges was delivering predictions in real-time. This required real-time access to observation networks to collect data, real-time processing of this data and real-time publishing of the results on the portal. Only a few minutes of latency could be tolerated to maintain the prediction’s validity. This required novel approaches (neural networks, high speed servers, real-time fluxes of input data, specific software techniques) to absorb huge quantities of data within a reduced time frame.

The subject of space weather and ionosphere phenomena is also a topic that is difficult to bring to the attention of GNSS users who do not have any prior specialised knowledge of the processes at play. It was therefore important to design products and functions that are easily understandable by users. More importantly, it was critical to raise the confidence level of users by providing accurate predictions. To ensure clarity and ease of use, a great deal of pedagogy and ergonomics was applied during the conception of the Graphical User Interface (GUI) and the system’s overall design.

The prototype was tested thoroughly and it will be further tested and improved within the European Commission, with the aim to eventually deploy it at the GNSS Service Centre in Madrid, for the benefit of all GNSS users.

Operational space weather initiatives from a variety of institutions (EU, academia, national research institutes, semi-public institutions, etc.) could also benefit from the service, e.g. through the collection and reuse of prediction products generated by IPS, comparisons with their own predictions, or by feeding their own algorithms with additional observation and prediction data to improve forecasts.

The original contributions of IPS are:

  • a prototype dedicated to GNSS users and not only scientific communities;
  • fully suited to operational users thanks to its high level of customisation and alarms configurability
  • most IPS products provide forecast analysis, instead of the now-cast analysis of most similar services
  • the service can provide real-time “quality of the service” analysis to estimate forecasting accuracy
  • the user has complete access to the repository of prediction products for further offline processing and time series analysis 

Disclaimer: The IPS service prototype cannot be liable for the availability or quality of its products, and the final user is the only one responsible for their use and interpretation.