David Gibbons, Chief Technology Officer, Atheras Analytics SAS
Our world is being primed for hyper connectivity, where everyone and everything is connected and optimized to enable the world to run more efficiently through 5G and the network of networks.
To enable this, ubiquitous connectivity must be achieved with different access technologies interwoven, interoperable and working together in a seamless manner to provide the most suitable communications access for the use case in hand. This will ensure that seamless connectivity is made available to all, and a user could then move between each access technology without so much as noticing it.
David Gibbons
In order to make this happen, and alongside the changes that the industry is already encountering with multi-orbit networks, the transformation has begun. Satellite operators everywhere are either in the process of or looking at the virtualization of their network.
Satellite connectivity is going through a migration to the cloud as it embraces telco standards and a move to become integrated into the broader communications landscape.
Both traditional and multi-orbit satellite networks will be key to this new era. Satellite is enjoying a renaissance with the emergence of smallsat constellations in LEO consisting of small, highly capable satellites that move quickly across the sky and deliver high performance, low latency connectivity. These complement traditional GEO satellite networks.
AI has a significant role to play in these networks, enabling intrinsic functions that help the network to manage itself such as automation and orchestration. Its precision and efficiency, monitoring capabilities and its basic ability to respond faster than a human to any issues are important characteristics.
However, AI also has an important role to play on the ground, where the situation is extremely complex and difficult to manage, bringing together multiple orbits and capabilities.
For satellite networks, location of the ground stations is a critical consideration that every satellite operator needs to make when designing their networks. First, decisions must be made on where the ground station should be located, and secondly, the operation and management of them becomes the priority.
________________________________
AI for ground network design
________________________________
When a network is in the design phase, there are a multitude of considerations that must be taken into account. The location must provide longevity. It will, after all be there for a long time.
Excellent line of sight with the satellites is vital, especially in the case of LEO constellations that require good horizon availability in order to track satellites. Access is critical, along with a lack of noise pollution that could cause interference.
Another important aspect is climate and the potential for weather disruption. The constellations of today operate in high frequency bands to allow for better performance. Weather patterns are of relevance for higher bands such as Ka-band and Q/V-band as these bands are highly susceptible to attenuation caused by precipitation. To determine the influence of weather on candidate ground station locations, it’s necessary to take into account historical weather data.
Directly from the beginning, it is essential to consider how many gateways and diversity gateways are required and where each of these gateways is located for best availability in terms of the impact of weather and establishing the optimum link budget/fade margin and optimum gain for each gateway antenna.
Traditionally, determining the optimal number and location of gateways, and planning for link availability and weather resilience, involved significant manual analysis and conservative assumptions.
It’s also worth noting here that when assessing the impact of weather, conventional approaches to ground network design do not always capture the impact of recent climate change. To address this, it’s essential that recent years’ weather data is included to ensure that climate change impact is considered.
Using AI tools with advanced algorithms, operators can now analyze historical rainfall as well as more recent weather data to quickly determine link availability for potential gateway locations.
The same AI-based approach can also be used to identify correlations between different sites to help determine how likely they are to be affected by the same weather system.
Operators can use these types of AI algorithms to run simulations and model various groupings of gateway sites to find the combinations that offer the highest availability at the lowest cost.
However, choosing the correct number of gateways, and where to put them, is only part of the challenge. There’s a constant trade-off between service availability and the cost of infrastructure.
Operators must decide how many diversity gateways are required, determine optimal antenna gain, and calculate the right link budgets and fade margins. AI helps here, as well, enabling operators to reduce capex. By selecting the right number of gateways, with locations that complement each other and reduce the chance of simultaneous weather outages, operators can build in resilience and ensure the desired throughput is reached, all without adding unnecessary diversity gateways that increase complexity and cost.
_________________________________
Outage prediction—
forewarned is forearmed
_________________________________
Real-time monitoring and proactive management are integral to maintain high levels of performance. For a seamless service, operators need to switch traffic to an alternative gateway before an outage occurs.
To accomplish that, they need to be able to accurately predict an outage a number of hours before that happens in order to have the time to synchronize gateways and switch traffic. Additionally, network operators also need to know—with a level of certainty—that the diversity gateway the traffic will be switched to isn’t also impacted or about to be impacted by a weather event.
AI is already making this kind of network management possible. It can be used to process historical weather data together with real-time data collected from the network, to predict outages ahead of an outage occurring, as well as to predict weather at diversity gateways.
These AI generated predictive analytics enable network operators to proactively take corrective actions before performance is impacted, by manually or automatically rerouting traffic through an unaffected gateway. This proactive approach minimizes service interruptions to maintain quality of service, while also optimizing network efficiency.
___________________________________
Huge potential for efficiency
___________________________________
The satellite sector is only just scraping the surface of how AI can be used to enhance its operations. It undoubtedly has a bright future in terms of enabling the automation and orchestration in a world where software is taking over from legacy hardware.
AI can also be employed on the ground, where extremely complex networks are dealing with huge amounts of satellites and terminals. Moreover, the thirst for performance is only going increase as end users demand more and more data, pushing satellite operators into higher frequency bands to cope with demand.
Already, AI is proving to be an indispensable tool in overcoming the challenges associated with operating at high frequencies and ensuring that HTS networks remain resilient, efficient, and reliable—and we’re only just starting to see what’s truly possible.
Author David Gibbons is a veteran of the satellite and defence business. He was Global Space Sales Director at TERMA Group (formal Atos/Siemens) where he focused on forging new markets and business opportunities and was also the Head of the ESA Space Solutions Centre, Ireland, and has held senior positions at Magellan and Moog.
atherasanalytics.com