24-09-2016, 01:20 PM
1456066585-61648.pdf (Size: 7 MB / Downloads: 6)
We present an overview of an air-to-ground laser communications demonstration performed at MIT Lincoln
Laboratory. Error-free communication at 2.5 Gb/s was demonstrated along a 25-km slant path between a 1-in
transmit aperture on an aircraft at 12 kft altitude and ground terminal with 4 separate 1-cm receivers. Power
fluctuations from turbulence-induced scintillation are mitigated in the spatial domain by use of the multiple
ground receivers and in the time domain by the use of forward error correction and interleaving. The optical
terminals are monitored by multiple high-rate sensors which allow us to quantify total system performance
INTRODUCTION
An important future application of lasercom is support of tactical intelligence, surveillance, and reconnaissance
(ISR) missions, delivering near real-time sensor data to the operational theater. Representative objective requirements
include the ability to transport data at rates up to tens of Gb/s over a range of roughly 50 km. The
air terminal should have low size, weight, and power (SWaP), and low integration impact on the aircraft. Use
of standard terrestrial network protocols can simplify the implementation, particularly at the client interface. A
key advantage of lasercom is its ability to provide high bandwidth without adding to RF congestion in theater.
An example of such an application, currently under development at Lincoln Laboratory, is the Multiple
Aperture Sparse imager Video System (MASIVS), an imaging system with 880 megapixels generating in excess
of 20 Gb/s of raw data that can be near-losslessly compressed to about 2 Gb/s. The work described in this paper
is directed towards providing an air-to-ground optical link that will allow the MASIVS system to relay data to
users on the ground in real time.
Optical links based on single-mode fibers in the 1550-nm band are commonly used in commercial telecom and
provide a readily available base of COTS components, including efficient pre-amplified receivers and components
that support a wide range of modulation formats. In free space communications the fiber connectivity between
transmit and receive nodes present in terrestrial telecom is replaced with a power delivery subsystem the sole
purpose of which is to deliver power from the transmit fiber to the receive fiber as efficiently as possible.
Communications performance depends on both the mean power in fiber and the fluctuations of power in fiber.
Design approaches must maximize average power and minimize its variance in the presence of atmospheric
turbulence and platform jitter.
Figure 1 diagrams the effects of the channel on communications performance. Power in fiber is a product of
both power delivered to the receiver aperture and the coupling efficiency from aperture to fiber. Transmitter
platform jitter and turbulence in the far field of the receiver both lead to fluctuations in power delivered to
the aperture, the well-known scintillation problem. In addition, turbulence near to the receiver causes phase
distortion in the aperture that reduces the efficiency of power coupling into fiber. Traditionally, phase distortion
can be corrected with adaptive optics; the lowest order distortion, tilt, can be corrected with fast tracking
systems. Moreover, as long as the receiver aperture is small compared to the phase distortion scale, r0, tilt
tracking is sufficient to ensure high coupling efficiency, avoiding need for higher-order phase corrections.
The design used in this work exploits 1-cm ground receiver apertures which use tilt-only tracking and which
have 8-m diffraction limited imaging resolution at a 50-km slant range to the aircraft. This provides the flexibility
to use multiple apertures at the aircraft, if desired, to avoid structural blockage or to allow seamless handover
between multiple aircraft apertures with limited field of regard. The small apertures ensures that high-order phase
aberrations are negligible; scintillation is the dominant atmospheric turbulence effect, and must be mitigated.
PERFORMANCE OF THE POWER DELIVERY SYSTEM
For initial signal acquisition, the pointing burden is placed on the aircraft since it has all the information required
to point to the ground terminal. The ground terminal requires knowledge of the aircraft position, which can
be obtained through a RF downlink if available. For this work, the aircraft was acquired optically by flying
through predetermined waypoints. After initial acquisition, any reacquisition can be achieved by propagating
the aircraft motion. In an actual ISR mission with the aircraft in a pre-known orbit, a wide acquisition camera
on the ground would suffice to acquire the downlink beacon.
Sample frames from tracking camera on the aircraft and the four tracking camera on the ground are shown in
Figure 9. The tilt-only tracking system is sufficient to deliver well-contained distributions in the focal plane. The
beam spot is smaller for the ground terminal, with nearly diffraction limited performance. The aircraft power
distribution is reasonably well contained, but is spread a bit in elevation. A known handover issue between the
fast steering mirror and the gimbal contributes to this distortion on the aircraft The potential impact of airframe
boundary layer turbulence is still under investigation, although it appears to be modest. Not shown in Figure 9
is the track jitter for the air and ground terminals, which we measure to be less than 0.3 λ/D.
Figure 10(a) overlays probability densities of power-in-aperture and power-in-fiber. With good tracking and
low higher-order phase distortion, the densities should closely match. A stiction-induced tracking disturbance in
the wide area beam director caused occasional mirror jumps. Contours of the residual centroid log probability are
shown in Figure 10b. The low probability jumps along the direction of the two gimbal axes are easily identified.
The result is deep fades in the fiber power distribution that do not appear in the aperture distribution, resulting
in the rising tail seen in Figure 10a. We can remove these jumps by ignoring points where the plane orientation
(as measured by the IMU) is changing quickly. The result is well matched distributions of fiber and aperture
power, indicating good tracking performance and the absence of higher-order phase disturbances.
The distributions in Figure 11 illustrate the single and combined 4-aperture distributions for two cases: 50-
km range afternoon and 25-km range evening conditions. Overlaid on the charts are wave propagation model
simulations for the respective cases, scaled with a multiplier for best fit relative to the well known Hufnagel-Valley