In this article, we present the first demonstration of an optical communications downlink from a low-earth orbiting free-flying CubeSat. Two 1.5U vehicles, AC7-B&C, built under NASA’s Optical Communications and Sensors Demonstration (OCSD) program were launched in November 2017 and subsequently placed into a 450-km, 51.6° inc. circular orbit. Pseudorandom data streams using on-off key (OOK) modulation were transmitted from AC-7B to a 40 cm aperture telescope located at sea level in El Segundo, CA. At 200 Mbps, without forward error correction (FEC), we achieved a 115-second link that was ~78% error free, with the remaining portion exhibiting an error rate below 1E-5. At the time of the engagement, the 1064-nm laser transmitter was operating at 2 W (half capacity) with a full width half maximum (FWHM) beam divergence of ~1 mrad, which was approximately double the anticipated pointing accuracy of the vehicle.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Small inexpensive satellite platforms offer opportunities for pathfinder experiments, space qualification of components and systems, and enhancement of larger assets. The Optical Communication and Sensor Demonstration (OCSD) was funded by NASA’s Small Spacecraft Technology Program (Space Technology Mission Directorate) to perform optical communication downlinks and proximity operation with a pair of CubeSats [1–3]. Each of the two vehicles, AeroCube-7B and -7C, host a 4-W, 1064-nm laser transmitters capable of data rates just over 600 Mbps. Limitations in the scope of the program precluded the development of ground station electronics needed to process data at rates beyond 200 Mbps. While 5.6 Gbps LEO satellite-to-ground links have been demonstrated from the NFIRE satellite to Tenerife  using precision gimbal pointing (see also OSIRIS , OPALS  and SOTA ), the present AeroCube7 architecture does not include an optical gimbal or any internal beam steering hardware. Eliminating the gimbal and moving parts reduces size, weight and power (SWaP), and cost, but requires setting the angular beam width to be on the order of a milliradian (0.06°) to accommodate a body steering approach. Increasing the beam divergence, however, reduces the optical intensity, which leads to lower data rates. For vehicles of this class, the low moment of inertia enables the vehicle to achieve sufficient pointing accuracy at slew rates that can support low complexity near-Gbps LEO-to-ground communication links. For this effort, The Aerospace Corporation has developed cm-scale star trackers that provide the required attitude references for our attitude control system. A key feature is that a ground-based uplink beacon is not necessary as sufficient attitude referencing and fine pointing control of the vehicles are obtained using an onboard pair of star trackers. Additionally, there is no downlink beacon provided to the ground station, which can track either in open-loop mode using the anticipated trajectory or in closed-loop mode using the incoming communication signal. While the OCSD experiments were not intended to demonstrate ultra-high bandwidth communications, the goal was to test the limits of technologies that are enablers for low-cost rapidly-deployable small satellite platforms for optical communications, remote sensing and other applications. Our strategy is to gain experience in manageable steps by setting reasonable goals and planning several low-cost missions to advance the state-of-the-art.
2. OCSD project status
The first OCSD vehicle, cataloged as “AeroCube-7A” (or AC-7A)  was launched on Oct. 8, 2015, as a pathfinder to test key spacecraft systems. The optical communications hardware incorporated a two-stage 10-W Yb fiber master oscillator power amplifier (MOPA) system. Initial on-orbit checkout proceeded as planned, until a software upload to the attitude determination and control system (ADCS) was interrupted by an unexpected spacecraft reboot. Subsequently, the ADCS was no longer functional, which disabled the laser subsystem and prevented any opportunity to perform optical downlinks. On Nov. 13, 2017, two additional CubeSats (AC-7B and -7C) were launched aboard Orbital ATK’s Cygnus resupply vehicle for NASA’s ISS following an unexpected year and a half delay due to issues with the original launch vehicles. The B and C spacecrafts, deployed Dec. 06, 2017, presently occupy an orbit with a nominal height of 450 km and an inclination of 51.6 deg. Both CubeSats host identical 4-W single-stage Yb fiber MOPA 1064-nm transmitters (see Section 4.1) with fixed output beam divergences set to 1 and 2.7 FWHM mrad for -7B and -7C, respectively. Silicon-based star trackers and mini reaction wheels constitute the critical hardware of the pointing subsystems. Communication downlink tests to our ground station in El Segundo, CA are presently being conducted with AC-7B, which by design, delivers ~7 times more irradiance than the -7C vehicle. To date, communication downlinks to our 40-cm telescope receiver have been conducted at 50, 100 and 200 Mbps with the laser transmitter power set to 2 W . At present, minimal communications testing has been performed with AC-7C.
A key result is that we have demonstrated a communication system that does notrequire an uplink or downlink beacon. The on-board star tracker based ADCS provides sufficient pointing to maintain the optical link while the vehicles are slewing. The ground station can either open-loop track on an anticipated trajectory or closed-loop track on the incoming communication signal. Spacecraft overpass trajectories are calculated from on-board GPS data that is downlinked to a supporting RF ground station network presently consisting of 5 remote sites. Operations are continuing in order to compile link and spacecraft pointing statistics. Mission lifetimes for these vehicles are anticipated to be ~3 years.
3. Spacecraft hardware
3.1. The spacecraft bus
As designed, each 1.5 U CubeSat weighs 2.3 kg. During lasercom engagements, which are expected to last up to ~3 minutes, the spacecraft consumes an additional 10-20 W power depending on the set point of the laser transmitter. Operating with this thermal load even over a short period in a 1.5 U CubeSat is a challenge. The laser system is mounted on an aluminum tray that is thermally connected to the spacecraft body allowing heat to be removed from the pump diode and optical fiber. Unlike many commercial CubeSats, the inhouse designed exoskeleton, or “body”, is machined from a single block of aluminum to provide structural and thermal stability. Figure 1 shows a detailed rendering of the AC-7B and -7C spacecraft. Figure 2 displays the actual vehicles undergoing final battery charging prior to integration with the launch vehicle.
Each spacecraft has two deployable wings to generate additional electric power. The wing surfaces and two of the rectangular side panels have three ~1-W, triple-junction solar cells, while each of the remaining two rectangular side panels have two 1-W triple-junction cells. Maximum solar input power is 9.5 W, and typical orbit-average power in 3-axis stabilized modes is 4.5 W. Two commercial-of-the-shelf (COTS) “high energy”  18650-size lithium ion batteries with a total capacity of 14 W-hr provide spacecraft power during eclipse, and two COTS “high current”  lithium ion batteries provide over 15 W to the laser transmitter during optical downlinks.
3.2. Fine attitude sensors and pointing
As a primary approach, precision pointing of about 0.024° is achieved using star trackers (see Table 1). A secondary approach that is available (for comparison), but has not been activated to date, implements an uplink ground laser beacon (5W at 1550 nm) and onboard quad cell for closed-loop pointing. Both methods are expected to provide better than 0.015° attitude knowledge, while heritage 3-axis reaction wheels provide a 0.015° control authority. This level of pointing accuracy is significantly better than the ~1° pointing accuracy provided by our previous attitude sensors .
The dual star trackers on each CubeSat use an ON Semiconductor MT9V022 monochrome WVGA imagers coupled to f/1.2 lenses. A Spartan-6 FPGA reads an image frame, processes the field, and outputs a collection of star locations. These data are further processed by a 16-bit microcontroller using a star catalog to output pointing directions as a set of quaternions. One of the star trackers points 20° off the + Z direction (typically zenith), while the other is canted by 70° to provide angular diversity in case stray light from the moon, sun, etc. interferes with one of the trackers. Star fields are imaged about once every 1.5 seconds, and quaternion outputs are combined with more rapid rate gyro data to provide continuous attitude information with much less than a 0.1° error.
During laser communications, a Sensonor STIM-210 3-axis rate gyro provides a bias stability of 0.5°/(hr)1/2 and can drift as much as 10°/hr due to thermal gradients. Frequent star tracker updates at a nominal interval of once every 1.5 seconds allows the on-board Kalman filter to estimate the gyro bias states and correct for them. The Sensonor STIM-210 gyro has a maximum power consumption of 1.5 W and needs a one-hour warm-up period . These devices are susceptible to helium exposure that can occur during launch vehicle preparations, so they are mounted inside hermetically-sealed containers. This issue was discovered on AeroCube-4 using the legacy STIM-202 units.
3.3. Attitude actuators
Attitude actuators include a triad of magnetic torque rods and a triad of reaction wheels. The torque rods have a magnetic moment of 0.2 A-m2 and provide torques for detumbling and reaction wheel unloading. The reaction wheels, shown in Fig. 3, have flight heritage on multiple spacecraft with 1 mN-m-s of total angular momentum. Slew rates in excess of 5° per second can be obtained using these wheels and pointing control to within 0.015° has been demonstrated on-orbit using them on a similar spacecraft. The vehicle needs to slew at a sufficient rate to keep the laser transmitter pointed at the ground station. Slew rates were typically < 1°/s for our links.
3.4. On-board ADCS data (attitude determination & control)
The AC-7B and -7C downlink engagements are expected to experience maximum slew rates of ~1°/s. Maintaining stars in track at the larger slew rates is challenging because the signal-to-noise is reduced as star images streak across multiple pixels during the integration period. In principle, only three stars are required to obtain an attitude solution. Typically, more star centroids are used to compare against the star catalog positions to derive residual errors. The root mean square (RMS) of these residual errors (example shown in Fig. 4) can provide a rough estimate of the star tracker solution uncertainty during the slewing. For the data shown in Fig. 4, the satellite was slewing faster than 0.5 degrees per second to support laser communications to the ground station.
The X and Y attitude control errors computed in Fig. 5 are comparisons of the commanded pointing trajectory to the measured attitude (based on a Kalman filtering of star tracker and rate gyro measurements) and reside approximately within a ± 0.01° window. Many of the transients observed in the attitude error result from new star tracker updates that are incorporated into the attitude determination solution. Given the star tracker solution uncertainty is estimated from the star matching residual errors, the magnitude of these transients is somewhat expected. The overall laser pointing error is dominated by the star tracker solution uncertainty. Higher accuracy star trackers with a smaller field of view have been developed for follow-on missions which should improve the body steered laser pointing accuracy. While the star tracker accuracy of each individual solution is unchanged as a function of slew rate, the number of solutions obtained is reduced when the vehicle is slewed because the signal-to-noise is reduced as fewer photons hit each individual pixel. Thus, overall system pointing performance is far better when the vehicle is inertially fixed due to the greater frequency of star tracker solutions.
4. The optical downlink
4.1. The laser transmitter
On-off key (OOK) data encoding is achieved by modulating the current of a low power DFB (master oscillator). For the initial tests reported here, we used a 223-1 PRBS pattern (without FEC) to evaluate the basic link performance. The encoded optical signal is amplified from ~10 mW to a maximum power of 4 W by a single stage polarization-maintaining ytterbium-doped fiber amplifier. Figure 6 shows a schematic of the transmitter, which is composed of COTS components and designed to operate over a temperature range of 10-50 °C. Due to power constraints, the only element that is actively cooled is the DFB in order to control the transmitter output wavelength. The Yb fiber is pumped at 915 nm rather than at 975 nm to accommodate the large expected temperature range on orbit. Pumping at 975 nm would have been more efficient and required less gain fiber. This arrangement, however, would have required the use of wavelength stabilized pumps which at the time of the build, were found to be significantly less efficient than unlocked devices at 915 nm and were unable to lock over the temperature range required. As designed, the AC7 transmit lasers are capable of delivering a max optical power of 4 W at 1064 nm with a wallplug efficiency of 20%. For the initial on-orbit lasercom tests (discussed here), both transmitters were initialized to operate at 2 W. Because of the uncertainty in the performance of the ADCS in this first-time demonstration, the divergences of the output beams for 7B and 7C were set conservatively at ~0.06 and 0.15° FWHM, respectively. The 7B divergence is roughly twice the allocated pointing accuracy of the ADCS (0.024°, as described in Sec. 3).
Figure 7 shows photos of the laser transmitter slice (during assembly) that occupies a volume of ~10x10x2.5 cm3. The components are mounted to an aluminum baseplate with fiber windings on both sides. Nusil and heat-conducting epoxy are used to dissipate the heat that is generated in the laser diode pumps and the gain fiber (not visible in the photos), respectively. A small lens, placed after the amplifier stage, is adjusted to yield the desired output divergence.
4.2. Optical ground station & receiver
The ground terminal consists of a 40-cm diameter, 3-m focal length Ritchey–Chrétien telescope (shown in Fig. 8) mounted on an in-house designed az-el gimbal that incorporates two Aerotech ALAR100 rotary stages. The gimbal is controlled by in-house developed flight software, which interfaces with Aerotech servo control software. The flight software also links two Xenics InGaAs camera arrays that can be used for active tracking. The principal InGaAs tracking camera shares a common path with a high-speed large area APD detector (800-µm dia. Si for the measurements reported in this article) through the main telescope. In the current configuration, a beam splitter is used to direct 10% of the incoming light to the tracking camera, which provides a 0.10° x 0.12° field of view (FOV). A secondary InGaAs camera is mounted on the side of the main telescope and provides a 1.5° x1.8° FOV.
The Si-APD detector converts the optical signal into an electrical waveform that is decoded and analyzed by an in-house built modem and software package. The modem performs clock and data recovery for the PRBS patterns transmitted by the spacecraft and logs the bit error count. For these initial downlink experiments, the FOV of the APD is sufficiently large to allow for the ground terminal to operate in an open-loop tracking mode with manual pointing offsets commanded as needed. Based on stellar line-of-sight calibrations, the post-cal open-loop pointing accuracy of the ground terminal was found to be about 40-µrad mean line-of-sight error over the gimbal field of regard. During downlink engagements, the initial total open-loop pointing error was noted to be about 3 to 4 times larger presumably due to uncertainties in the CubeSat ephemeris. Trajectory files for the ground telescope are generated from the vehicle GPS data that is telemetered to one of several RF ground stations prior to the optical engagements. Open-loop operation will continue in order to gain a better understanding of the ephemeris errors. The ground station is also capable of operating in a closed-loop mode, which will be implemented as needed for smaller FOV detectors to be used in upcoming experiments.
4.3. CubeSat space-to-ground pointing
Accurate pointing of the spacecraft is critical for successful optical downlinks and is dependent upon the accuracy of the star trackers and knowledge of the alignment between them and the laser. Initial tests to characterize this alignment involve successive spiral sweep patterns of the Cubesat about the projected location of the ground terminal. The alignment is estimated by reconciling two sets of data: 1) time stamped received signal strength recorded by the ground station’s tracking cameras, and 2) time-stamped satellite attitude determination data from the on-board star trackers and rate gyros. In this process, the signal strength is weighted by the link range, which is constantly changing during any given engagement. Further, repeat scans are needed to account for non-systematic atmospheric losses due to clouds. With successive tests, knowledge of the pointing is refined, and the angular extent of the spiral sweep pattern is reduced. For greater sensitivity, bit error rate (BER) measurements from the ground station modem are used for the alignment metric in place of the received signal strength. Examples of 1° and 0.2° diameter scans are shown in Figs. 9 and 10, respectively.
4.4. Optical downlink performance
Optical engagements were intentionally limited for our initial experiments to elevation angles between 30° and 70° and were thus typically 2-3 minutes in duration. After several downlinks with spiral scanning were concluded, the optimum pointing conditions for the vehicle were set. A number of preliminary BER measurements at 50 and 100 Mbps were collected during spiral operation (not shown here) to help with the alignment and to check out the data collection system. Once the alignment was optimized, the vehicle pointing was fixed and the data rate was increased to 200 Mbps. The data/BER collect at 200 Mbps (PRBS23) for our best pass to date is shown in Fig. 11, where the average link range was around 725 km. The bit errors were tabulated over 125 ms intervals for the full 115 s window of the engagement; ~77% of the time intervals were error free (set to 1E-9 for convenience) with ~92% having BERs ≤ 1E-6. BERs near 1E-7 signify that 1 or 2 errors occurred within the interval. Instances of higher BER counts are presumed to result from transient pointing errors. Further analysis is ongoing.
Under the NASA OCSD program we demonstrated the first optical communication downlink from a LEO orbiting body-pointed CubeSat implementing COTS components and subsystems. A nearly error-free link at 200 Mbps, from LEO orbit to sea level, was achieved over a two-minute engagement without the use of FEC. Sufficient vehicle pointing accuracy using on-board star trackers was realized, obviating the need for an uplink beacon and simplifying the design/operation of the ground station. Open-loop pointing of the ground station with as-needed manual offset commands was found to be adequate for the detector geometry implemented and the data rates achieved so far. Future testing will target higher data rates and will transmit captured image files from an on-board camera, rather than a PRBS test pattern. Alignment stability of the CubeSat vehicles will be monitored periodically.
All trademarks, service marks, and trade names are the property of their respective owners.
NASA’s Small Spacecraft Technology Program; Ames Research Center.
The authors wish to thank R. Dolphus, J. Wilson, P. Carian, B. Hardy, D. Hinkley and A. Berman of The Aerospace Corporation for their valued contributions to this effort and the NASA Ames POC, R. Hunter.
1. T. S. Rose, S. W. Janson, S. LaLumondiere, N. Werner, D. H. Hinkley, D. W. Rowen, R. A. Fields, and R. P. Welle, “LEO to ground optical communications from a small satellite platform,” Proc. SPIE 9354, 93540I (2015).
2. S. Janson, R. Welle, T. Rose, D. Rowen, B. Hardy, R. Dolphus, P. Doyle, A. Faler, D. Chien, A. Chin, G. Maul, C. Coffman, S. LaLumondiere, N. I. Werner, D. Hinkley, "The NASA Optical Communications and Sensor Demonstration Program: Initial Flight Results," in Proceedings of the 30th Annual AIAA/USU SmallSat Conference (AIAA, 2016), paper SSC16-III-03.
3. T. S. Rose, D. W. Rowen, S. LaLumondiere, N. I. Werner, R. Linares, A. Faler, J. Wicker, C. M. Coffman, G. A. Maul, D. H. Chien, A. Utter, R. P. Welle, and S. W. Janson, "Optical communications downlink from a 1.5U cubesat: OCSD program," Proc. SPIE10610, 109100T (2019).
4. R. Fields, D. Kozlowski, H. Yura, R. Wong, J. Wicker, C. Lunde, M. Gregory, B. Wandernoth, and F. Heine, “5.625 Gbps Bidirectional Laser Communications Measurements Between the NFIRE Satellite and an Optical Ground Station,” in International Conference on Space Optical Systems and Applications (IEEE, 2011), pp.44–53.
5. C. Schmidt, M. Brechtelsbauer, F. Rein, and C. Fuchs, “OSIRIS Payload for DLR’s BiROS Satellite,” International Conference on Space Optical Systems and Applications (IEEE, 2014).
6. O. Sindiy, M. Abrahamson, A. Biswas, M. W. Wright, J. H. Padams, and A. Konyha, “Lessons Learned from Optical Payload for Lasercomm Science (OPALS) Mission Operations,” in AIAA 2015–4657 (AIAA, 2015), https://doi.org/10.2514/6.2015-4657. [CrossRef]
7. A. Carrasco-Casado, H. Takenaka, D. Kolev, Y. Munemasa, H. Kunimori, K. Suzuki, T. Fuse, T. Kubo-oka, M. Akioka, Y. Koyama, and M. Toyoshima, “LEO-to-Ground Optical Communications using SOTA (Small Optical TrAnsponder) –Payload Verification Results and Experiments on Space Quantum Communications–,” Acta Astronaut. 139, 377–384 (2017). [CrossRef]
10. S. W. Janson and R. P. Welle, “The NASA Optical Communication and Sensor Demonstration Program: An Update,” in 28th Annual AIAA/USU Conference on Small Satellites (2014), paper SSC14-VI-1.
11. A. S. Sensonor, “Butterfly Gyro STIM210 Multi-Axis Gyro Module datasheet,” http://www.sensonor.com/media/90317/datasheet%20stim210%20ts1545.r10.pdf.