Life is for sharing.

EAN – A Look beyond the Heartbeats

The aviation industry faced its biggest crisis to date in 2020 due to the Coronavirus pandemic and some challenges will undoubtedly persist in 2021 as airlines navigate towards a recovery. Nevertheless, we turned this difficult situation on its head and used the time to further enhance our award-winning European Aviation Network (EAN) inflight broadband solution. Over the coming weeks, our new EAN Tech Update Series will explore some of the key improvements and features that we implemented during this period to evolve our EAN service offering and make it more compelling for current and future airline customers.

EAN has been successfully developed by Inmarsat and Deutsche Telekom as the world’s first and only dedicated aviation connectivity solution to combine a satellite link with a complementary ground network (CGC) of around 300 antenna sites, delivering resilient inflight broadband to airlines over European airspace. 

Enabling Optimal Service Quality

An important prerequisite for delivering outstanding service quality to our customers is service awareness as it ensures that performance can be measured against pre-defined Key Performance Indicators (KPIs). After all, you cannot successfully operate what you cannot reliably monitor. We have therefore made further improvements to complement our EAN network monitoring capabilities.

EAN’s operational concept already included a wide range of monitoring features that are commonly used by both the mobile satellite industry (for our S-Band satellite link) and the legacy retail market mobile radio business (for the CGC link). For example, small pings are sent and received at regular intervals using a communication protocol that has been customized for inflight broadband applications. These so-called ‘aircraft heartbeats’ ascertain that the connectivity is ‘alive’ and serve as the main input to estimate EAN’s network availability. 

We have now enriched those heartbeats for our CGC link by working with our partner FlightAware, which provides an Automatic Dependent Surveillance Broadcast (ADS-B) data feed via its Firehose product. ADS-B is a well-established surveillance method for Air Traffic Control (ATC) monitoring with many advantages over traditional radar in terms of cost and update rate. It works by periodically broadcasting information about an aircraft's GPS location, altitude, ground speed and other data to ground stations and other aircraft, enabling it to be tracked by relevant third parties. The addition of Firehose’s feed to the data collected by our own monitoring tools means that we can now combine EAN’s instantaneous service measurements with information on the location and speed of an aircraft, bringing together two independent flight data collection sources for better aircraft tracking and service monitoring in real-time. 

Powerful Back-End Systems and Automation

However, collecting the flight data is only the beginning of our aircraft monitoring story. We have also created powerful back-end systems to store and process the vast amount of data collected. As a result, we can not only set-up alarms for specific event triggers, issue comprehensive reports and instantly initiate the right troubleshooting procedures, but also do historic benchmarking of all key performance indicators (KPIs). These KPI measurements can then be clustered as needed, in whatever pertinent time window we choose, to fine-tune our monitoring capabilities.

The key to efficient monitoring of the data collected by our enhanced monitoring platforms is of course automation as it would be impossible to sift through the so-called ‘data lake’ manually. Modern technology such as advanced analytics and big data analysis structures the indexing of the collected data for easier reporting. Furthermore, intelligent algorithms have been developed that process the data lake for any suspicious behavior. The system can identify outlier events that deviate from expected behavior by comparing real-time performance to benchmark KPIs, a process known as anomaly detection. As mentioned above, alerting the monitoring crew quickly and pointing specifically to potential problems enables them to act precisely where needed and rectify any issues with the least impact to EAN’s inflight service experience. And we won’t stop there. With a loopback, we will ‘train’ the anomaly algorithms to improve and become even more sophisticated over time. This eventually will also open the path to cross-correlate records with other source as needed, such as weather data for example.

Intuitive Data Visualization for Easy Monitoring

The final element needed for optimizing EAN’s monitoring and operation features is intuitive data visualization. In addition to the many existing dashboards reporting on service performance and overall EAN network status, we have created EAN flight maps. These are designed as layered context that can be customized to support more insightful monitoring, troubleshooting or simply internal service demonstration. The elements we want to focus on – such as the specific airline, aircraft, region and even down to individual cell tower performance KPIs – can be easily selected by pre-defined filters. Deep dive information on the current aircraft service can simply be zoomed with a single click.

The beauty of the end-to-end monitoring system we now have in place for EAN is that the more customers we serve and the more real-time data we collect, the better our optimization algorithms will work. This allows us to raise the bar for a best-in-class passenger inflight broadband experience.

We of course hope to see the pandemic receding and the airline industry picking up this year. Until then, stay safe and look out for the second EAN Tech Update, where we will focus on how geo-redundancy has been leveraged to benefit our network reliability and hence our customers.