In our series ’Progress in Physics’ we present research topics from Swiss physical institutes, but also engineering highlights from Swiss industries, performed by physicists. The following article describes an industrial space project, the manufacturing of Optical Terminals for fast data exchange within satellite networks, using laser beams. The terminals must generate optical links between fast orbiting satellites over large distances of up to 60’000 km with sub-arcsec pointing accuracy. These challenging link requirements need solutions at the limits of current technologies. The terminals are designed and manufactured in Switzerland, and the whole business is rapidly growing. We describe in the following, how to achieve the stringent technical performance of such masterpieces.


Optical terminals for data communication in space

Fig. 1: Network of satellites with Optical terminals for bi-directional data transfer. Low earth orbiting satellites interact with higher orbiting satellites and ground stations. The accurate mutual pointing of the laser beams from terminal to terminal is the main technical challenge.

Bernhard Braunecker, Leica-Geosystems AG, CH-9435 Heerbrugg
Edgar Fischer, Oerlikon Space AG, CH-8052 Zürich


System Layout

The increasing demand for high speed Internet data transfer over large distances will doubtless be dominated by optical technologies, using fiber and optical free space connections. One expects about 10% of the Internet traffic being transmitted in future via satellite networks. After the worldwide communication crisis in the early nineties a critical review of all technologies involved has started. New applications like the transfer of large 3D images in real time will need very powerful point-to-point and point-to-multi-point communication channels, focussing on satellite to satellite, but also on satellite to ground station communication links. Thereby laser beams transfer data information over distances of up to 60’000 km between satellites at different orbits (Fig. 1).

The free space data exchange by optical means requires at both link ends optical transceivers, so called ‘Optical Terminals’. The sender terminal emits the information towards the counter station, where this forward data stream beam is either intensity or phase modulated. In the reverse direction the terminal receives simultaneously the very weak backward data signals from the counter station.

The primary advantage of a free space laser link is its narrow beam width. If the laser wavefront is diffraction limited, the beam divergence angle δα is proportional to λ/D, where λ is the wavelength and D the diameter of the transmitter aperture. Taking e.g. λ = 1.064 μm (Nd-Yag Laser) and D = 135 mm results in δα ~1 arcsec. Such a small divergence causes a high ‘antenna gain’, since the emitted intensity can be kept within a diameter of only several 100 m at the counter terminal, 60’000 km away. This leads to a much better signal to noise ratio, compared to competing Radio Frequency (RF) links, and allows a faster data transmission at the same bit error rate. However, the narrow beam width causes also the main technical challenge for optical communication systems in space. The receiver must acquire the slender transmit beam and the optical connection must be maintained with sub-arcsec accuracy throughout communication.

Fig. 2: Optical Terminal with an all-mirror, afocal telescope of 10x magnification. In the exit pupil is a fast scanning mirror to fine adjust the laser beam to the counter station.

The requirements for a space terminal are low mass, compact packaging size and rigidity against space environmental loads (launch vibrations, separation shock, zero gravity operation, thermal gradients, and severe radiation impacts). We chose a layout (Fig. 2), which main part is an all-mirror, afocal telescope of 10x magnification, often called an ‚Optical Antenna’. The telescope should enlarge the diameter of the laser beam by a factor 10 to obtain the small beam divergence δα mentioned above. The received radiation is directed in opposite direction through the telescope to a point sensor for fast data demodulation.

The chosen arrangement of four folded mirrors M1...M4 is that of a Kepler telescope, where M1 and M2 act as ‘objective lens’ and M4 has the function of the ‘ocular lens’. The mirror M3 is only a plane folding mirror. We see in the exit pupil of the telescope a ‘fine pointing device’, which is a fast scanning mirror, to deflect the laser beam about ±10° in both directions. This leads, due to the 10x telescope magnification, to a beam sweep of ±1° on M1 at the space side. This controllable beam steering is needed to keep the laser beam direction stable, since satellite platform are rather instable and jittering.

To reduce the manufacturing complexity, the mirror surface functions are only conical aspheres: M1 and M4 are parabolic and M2 is hyperbolic. All mirrors are made of Zerodur, a glass ceramic of zero thermal expansion, known as preferred material for the large ground based telescopes in astronomy.

To achieve the specified high antenna gain, any central obscuration of the laser beam must be strictly avoided, not to broaden the beam spot at the counter terminal by diffraction at the obscuration. This forced us to design M1, M2 and M4 as ‘off-axis’ aspheres, i.e. mirror segments, cut out from rotational symmetric ‘mother’ mirrors.

We also note at the short side a 2D star-sensor, which permanently monitors the star pattern inside the field of view of the telescope, and compares it with pre-programmed star configurations. One obtains so the absolute angular orientation of the terminal in the orbit, which is an important information when switching the link from one satellite to another. In Fig. 3 we see an Optical Terminal, based on the described layout and optimized for mid range data links, a product of Oerlikon Space AG.

Fig. 3: Optical Terminal from Oerlikon Space AG. The large mirror left in front is a ‘coarse pointing unit’ to direct the laser beam to different counter stations without changing the orientation of the satellite.

System Improvement

The question was raised, how to further improve the wavefront flatness of the laser beam and thus the antenna gain, but without increasing the available space and keeping mirrors M1, M2 and M4 conical?
One elegant possibility would be to polish some ‘freeform’ surface deformations on M3, i.e. changing it from a plane to a structured mirror, called a reflective hologram. Since M3 is near to the entry pupil at M1, but also near to the intermediate image between M3 and M4, this surface manipulation should correct both, the spherical aberration and the field distortion, an important quantity for the star sensor.
Our design work shows that the wavefront quality can be improved by a factor 2. The polishing of the surface correction on M3 could be done by modern CNC technologies [1].

In conclusion, we showed that optical space terminals are rather complex devices to establish the data transmission over large distance between fast moving and mechanically jittering space platforms. To assure the high reliability under the severe space environmental conditions, a deep physical understanding of all parameters involved has to be part of the engineering work.



[1] “Advanced Optics using Aspherical Elements”, Editors: B. Braunecker, R. Hentschel, H.J. Tiziani, SPIE Press Book (2007), ISBN 978-0-8194-6749-2



[Released: January 2008]