Optical communication has the potential to increase data rate without increasing the mass and power consumption of the onboard communication module. To achieve proper communication with deep space terminals, suitable ground based Optical Antenna must be developed providing large collecting area. Based on more than three decades of experience large optical telescopes development, AMOS has designed a 4m monolithic and a 6m segmented optical antenna for this purpose, in the frame of a GSTP contract with ESA/ESOC.

This study shows how to cope with several operational constrains specifically related to the optical communication antenna, and not to classical telescopes used for astronomy. The main one is the use of the antenna during daytime, and thus the direct exposure of the antenna to the sun. Another challenge is the ultra-narrow bandwidth spectral filtering needed to reject properly the signal background.
On the other side, the optical performance of the antenna is less demanding than astronomical telescopes. The main reason for this is that the antenna shall be able to perform properly under average environmental conditions (mainly seeing), while telescopes shall provide the best performance when the conditions are excellent. This leads to relaxed requirements on mirrors optical quality, on materials, on mirror segments positioning accuracies, on tracking performance, …

The optimal design in term of cost and performance has been selected based on analysis and comparison of the 4m and 6m configurations. This paper presents the optical, mechanical and control system design of the optical antenna. Special focus is given on the primary mirror unit and on the sun shielding. A potential implementation of adaptive optics is also introduced, as well as the associated performance improvement.

Target of the ESA TRP project “Optical Hybrid” [1] is a breadboard implementation of a high photon efficiency (HPE) link, which is compliant with the CCSDS standard recommended mainly for optical communication systems of scientific purposes (like Deep Space missions). The HPE specification includes the physical layer [2], but also the channel coding and modulation [3]. Therefore, the design and implementation of a breadboard with emulation capabilities of the downlink is of paramount importance for the project. In this context, the signal processing part was implemented on a software-defined-radio (SDR) platform.
In the first phase of the project, a detailed system simulation was setup. One key element of the HPE is the M-ary Pulse Position Modulation (PPM), which provides excellent power efficiency for optical links operated in deep space. In addition, a serially concatenated PPM (SCPPM) forward error correction (FEC) scheme [5] is employed in conjunction with iterative soft decoding, which is also necessary for efficient operation of a deep space link. All functional blocks described in the standard, from the CCSDS transfer frame to the encoded and mapped binary vector indicating the position of pulsed PPM slots, are implemented.
The focus of the simulation was on the implementation of a realistic model of the physical layer emulator. For the laser transmitter a wavelength of 1550nm is used. A bias-controlled Mach-Zehnder (MZ) modulator generates the optical pulses derived from the output of the Digital-to-Analog Converter (DAC) of the SDR platform. Pulse widths with 2ns or more might be generated. The extinction ratio (ER) is the figure of merit, which has been considered by simulation as well. The channel part of the emulator consists of fixed and variable attenuators to realize the given link budgets. The simulated channel model is a simple line-of-sight scenario. Moreover, the channel adds background noise according to the background flux defined in the test cases. A superconducting nanowire single-photon detector (SNSPD) of only two channels (for cost saving reasons) was bought from Single Quantum with a detection efficiency of 0.4 for a 300Hz dark count rate. The dead (blocking) time between detections is about 15ns. The blocking time is a key parameter of any SNSPD element, setting a theoretical limit on the maximum count rate of the detector. The impact of blocking can be mitigated by connecting several SNSPD elements (channels) to an SNSPD array. In fact, spreading the receive signal over many detectors reduces the photon flux per detector, thus reducing the per-detector blocking probability. Another mitigating aspect from the signal processing point of view is the spreading of PPM symbols (foreseen in the standard) to super symbols. The simulation takes into account the pulse shape of the detector, the number of channels, quantum efficiency, dead time, dark count rate and spreading factor. Therefore, the simulation has to consider several combinations of these parameters to verify trade-offs and degradations.

The SDR platform receives the detector pulses on two ADC channels and reduces information to a two-bit word per slot. This is the soft information to be gained by the two channels. In the module for optical synchronization we must determine the PPM symbol boundaries via guard slots. These slots are dark and an appendix to the symbol. The maximum likelihood (ML) approach, derived in [4], estimates the timing shift in a fraction of a slot and has been shown to approach the Cramér-Rao bound.

However, before passing the samples to the SCPPM decoder, synchronization of the super-symbol and code-block boundaries is necessary by using the code sync marker (CSM). Simulations results of the Symbol Error Rate (SER) and Coded Bit Error Rate (BER) down to 10-7 are presented, which could be achieved with a maximum of 32 iterations of the SCPPM module. Degradations to the ideal reference are measured. In the sequel, these simulation results will be the reference for the tests on the breadboard. Relevant and tested modules of the simulation environment are considered for the SDR platform. Therefore, the implementation is in C++ supported by the Intel AVX2 architecture, to optimize the throughput; this includes also the parallelization of the codec. The goal is to demonstrate a live stream processing for a slot width of 10ns. For higher throughputs (smaller slot widths), sample dumps are stored and post-processed.

[1] System Study of Optical Communications with a Hybridised Optical/RF Payload Data Transmitter, (ESA Contract No.4000115256/15/NL/FE)
[2] CCSDS Proposed Recommended Draft Standard, “High Photon Efficiency Optical Communications – Physical Layer”, White Book, September2016.
[3] CCSDS Proposed Recommended Draft Standard, “High Photon Efficiency Optical Communications – Coding & Synchronization”, White Book, September2016.
[4] R. Rogalin et al.,“Maximum Likelihood Synchronization for Pulse Position Modulation with Inter Symbol Guard Times”, Proc. IEEE GLOBECOM, Dec. 2016.
[5] Michael K. Cheng et al.,“Implementation of a Coded Modulation for Deep Space Optical Communications”, Jet Propulsion Laboratory, California Institute of Technology

Optical High-Speed Telemetry Downlinks for high data volumes from Earth Observation Satellites in Low Earth Orbit (OLEODL) is currently developing into a future standard in high-speed space communications. The German Aerospace Center’s (DLR) Institute for Communications and Navigation (IKN) has developed the miniaturized data downlink terminal OSIRIS (Optical Space InfraRed downlInk System) for space-to-ground links over LEO-distances, and has proven its performance by pre-operational downlinks of camera data from the University of Stuttgart’s Flying-Laptop (FLP) satellite. The optical transmitter OSIRIS consists of an optical communication module plus a fiber amplifier and another optional optical source, power supply control, and the optical beam emitter connected by fiber to the communication module. The signal is directed to a dedicated Optical Ground Station by body-pointing of the satellite during overflight, dynamically controlling its attitude by star sensors. The data rate of a testing sequence can be chosen between 39Mbps and 622Mbps, and operational live image data is transmitted at 80Mbps using standard CCSDS frame format. Live means that Earth observation images are transferred in near real time via the optical data downlink right after image acquisition.
The transmission follows the upcoming CCSDS-standard (Consultative Committee for Space Data Systems) for Optical On-Off Keying (O3K) (in preparation) in regard to the physical layer parameters, i.e. carrier wavelength, data rates, and modulation format.
Downlinks were performed to DLR’s transportable optical ground station, located at the IKN in Oberpfaffenhofen near Munich, as well as to ESA’s optical ground station (ESA-OGS) located on a mountain top at Tenerife, Canary Islands. We present here the OSIRIS-on-FLP structure and functionality, the measurement setup at the ESA-OGS, and results of downlink signal quality as well as bit error rate statistics and image transfer performance.
Steering of the optical signal spot towards the OGS during overflight is currently performed by body pointing of the satellite’s AOCS, assisted by closed loop control through star-sensor data. In future OSIRIS-implementations the satellite target pointing will no longer be necessary since a dedicated beam pointing opto-mechanic will perform autonomous beam steering while the satellite is not effected by the downlink operations. More OSIRIS-based terminals are currently being prepared for launch on Cubesats (OSIRIS4CUBE), as well as for the Bartolomeo-platform onboard the International Space Station.