Data driven decision making

Leveraging advanced atmospheric science


We enable direct to Earth laser communication thanks to cutting edge climate understanding

Laser communications are easily blocked by clouds and scattered by turbulence. Continuity of service can only be ensured through site diversity and precise understanding of the atmospheric communication channel. Unlike fiber optics, the atmosphere is an ever-changing propagation medium inducing continuous bandwidth variations of the theoretical link capacity.

Atmospherical characterization

Modeling, measuring and forecasting of all relevant parameters including cloud cover, turbulence and absorption.

Survey potential sites for Optical Ground Stations

Modeling and simulation to choose the best laser communication paths, between space and ground segments.



Miratlas offers its clients real-time and continuous monitoring of atmospheric turbulence. The data collected allows us to mitigate the impact of the atmosphere on laser communication, by anticipating and minimizing it.

We want to ensure that our clients can get critical information to optimize their infrastructure and improve their decision making.

SkyMonitor Network
SkyMonitor Network

The map above shows a subset of the instrument network that Miratlas has already deployed.



We offer prediction of atmospheric conditions as a service.

In particular, the measurement and prediction of atmospheric turbulence has an important scientific and economic value as it helps operators optimize their optical ground segment.

To predict these conditions, we use machine learning, which is central to the extrapolation and interpretation of data from our network of instruments. The quality of short-term forecasts of atmospheric disturbances depends on the quality of the data collected, on its quantity, and on its consistency.

The data quantity is increasing as our Sky Monitors are deployed: our network of instruments offers the diversity of observation sites necessary to study different climates. Data accumulated over time and continuity of measurements over several years make it possible to obtain the necessary seasonal variations.


Miratlas uses disruptive technologies to predict atmospheric absorption and scattering. We use machine learning to extrapolate and interpret data from our instrument network.

The quality of short-term atmospheric forecasts depends on the quality, quantity and consistency of the data collected.

Miratlas designs, manufacture and deploys its own instruments, ensuring the quality of the data. 

The quantity increases over time as our network is deployed, offering a worldwide diversity of observation sites.

The consistency is ensured by the continuity of measurements over several years, which makes it possible to characterize seasonal variations.