A bidirectional low Earth orbit-to-ground optical link experiment
Free-space optical communications using existing space-borne assets is a cost-effective way to show the maturity of the technology for future missions. To this end, NASA's Jet Propulsion Laboratory (JPL) undertook a joint experiment with the Japanese Aerospace Exploration Agency to perform a bidirectional optical link between the Laser Utilizing Communication Experiment (LUCE) instrument run by the National Institute of Information and Communications Technology on the Optical Inter-orbit Communications Experiment Test Satellite (OICETS)1 and the JPL Optical Communications Telescope Laboratory (OCTL) 1m ground station (see Figure 1). Originally tasked for intersatellite communications with the geostationary earth orbit (GEO) Artemis spacecraft that were successfully completed in 2006, the low-Earth-orbit (LEO) OICETS also supported space-to-ground links by inverting the spacecraft and pointing LUCE towards the ground using onboard gyros, providing navigation for extended periods of time. There were several objectives for OTOOLE (OCTL-To-OICETS Optical Link Experiment). First, the experiment aimed to demonstrate acquisition and tracking of a LEO satellite from the OCTL and perform an optical communications link with 2Mb/s uplink and 50Mb/s downlink. Second, it sought to validate operational issues to support an optical communications link, such as link models and aperture-averaging effects. And finally, the experiment hoped to characterize link performance for a variety of atmospheric and background conditions.
The campaign began in January 2009 and quickly completed four successful link demonstrations at each attempt in May and June 2009.2 Experiment windows occurred between 3 and 4am. They required that the ground station be in the dark and were determined by the sun angles incident on the LUCE and spacecraft maintenance (due to updating pointing verification requiring spacecraft inversion).
The ground system consisted of a receiver module at the focus of the boresighted 20cm acquisition telescope of the 1m-diameter primary mirror at OCTL. Twenty percent of the received downlink was split off to a CCD camera, and the remainder was equally split between a high-bandwidth avalanche photodiode module for data detection and multimode fiber coupled to a low-bandwidth photodiode to measure power fluctuations in the downlink (see Figure 2).
The multibeam approach, first shown by JPL in a 1995 experiment with the National Space Development Agency of Japan Engineering Test Satellite (ETS)-VI, was used to mitigate uplink beam scintillation.3 Four 1W diode lasers at 801nm served as the beacon source for coarse tracking. Three 10–25mW 819nm single-mode communication lasers (directly modulated at 2.048Mbps with a binary-pulse-position-modulation pseudo-random bitstream and temperature tuned to match the receiver filter on the spacecraft) served both for communications and fine tracking. A Raman notch filter coupled the beacon and communication sources through the coudé focus of the main 1m telescope via an off-axis parabolic mirror. Beam divergences exiting the telescope were on the order of 1–2mrad for both beams. Data acquired included uplink beam diagnostics, downlink beam power, a video of the imaged downlink, and downlink bit error rate. The downlink retrieved at the receiver was split to a clock-and-data-recovery chip and to a peripheral component interconnect-based digitizer with global position system time stamps.
Typical passes lasted for a maximum of 6min with the sun-illuminated satellite initially acquired from blind pointing using recently updated consolidated predict files. The sequence of events was then as follows. The ground beacon and communication beams illuminated the spacecraft as it rose above the OCTL tree line. Once the downlink was acquired (see Figure 3), the beacon beam was turned off, leaving the communications beam to maintain ground-station tracking by the fine tracking sensor. The actual link was established for up to approximately 140s, and it could, at times, be maintained for the entire track within the constraints of the LUCE pointing issues, i.e., sun-probe-Earth angle. Designed primarily to track with much slower angular velocities of the LEO-GEO links, dropouts occurred on most passes due to issues with the fast tracking required for LEO-to-ground tracking. Figure 3 shows a representative plot of the received power for the pass of 11 June. Bit-error rates nominally were less than 10−5 except during dropouts. Link models were used to compare predicted with measured received data at 500ms intervals (see Figure 4). The data was bounded by two atmospheric aerosol models for good agreement of the prevailing conditions. The scintillation index was also calculated and varied between 0.3 and 0.1 over elevation angles of 20 and 50°, respectively, for the 11 June pass.
In summary, the experiment provided an excellent opportunity to show a LEO-to-ground bidirectional optical communications link in a short time frame with existing assets. It allowed the collection of valuable experimental data to validate link models under the joint international collaboration. We and others are currently planning future demonstrations from OCTL to other satellites to support various NASA projects.
This work was performed by JPL, California Institute of Technology, under contract with NASA and supported by the Japanese Aerospace Exploration Agency and the National Institute of Information and Communications Technology.
After receiving his PhD in physics from the University of New Mexico, Malcolm Wright was with the Air Force Research Laboratory developing high-power semiconductor lasers. Since 1998 he has been with the JPL, California Institute of Technology, developing laser systems for NASA's free-space optical communications technology program and future flight projects.
