Abstract:
In 2009, NASA will launch an optical communications payload to Mars capable of supporting data rates of 1 to 30 Megabits per second. A key element of the Mars Laser Communications Demonstration (MLCD) project is the development of a distributed optical receiver array known as LDES: Link Development and Evaluation System. This paper will describe the LDES system, discuss the key technical challenges in its implementation, and show performance of hardware developed for the receiver system.

Introduction
NASA is presently developing the Mars Laser Communications Demonstration Project (MLCD), which will provide an optical link from Mars to Earth at rates from 1 to 30 Megabits per second. In order to realize such data rates, the system has several key components: a flight terminal that can produce and modulate optical power and precisely point it at an earth receive terminal, a large aperture receiver to collect and decode the signal, and a beacon reference system to assist in flight terminal pointing. This paper describes one realization of a receiver for the MLCD project that demonstrates a distributed aperture approach to achieving the large collection areas required to support megabit per second data rates.
Breaking the total required aperture into a number of pieces has several attractive advantages over a single monolithic receiver. First, the system is less costly to manufacture for large collection areas, since the cost of large optical quality mirrors grows faster than the collection area. Second, a distributed receiver architecture is scalable: collection area can easily be added to improve system performance. Third, distributed receivers are flexible: the user can choose to use pieces of the array to support multiple links, or to gang the array together to maximize sensitivity. Finally, receiver arrays are robust. Individual elements can be serviced without impacting the entire system.

Figure 1: Number of bits per detected photon on the LDES testbed

Table 1: Key System Parameters of the LDES Receiver

Table 2: Background Mitigation Techniques

In order to realize a useful receiver array, there are three principal technical challenges that must be overcome to attain the same performance that is achievable with a single large collector. In particular, since each receive telescope will require at least one detector, very low noise detectors are required to maintain equivalent performance to a monolithic approach. Also, a method must be devised to efficiently combine the signals from the independent telescopes such that the total signal received is nearly equal to the sum of the individual parts. Finally, for a communication mission with Mars, it is important that the receive telescopes are capable of pointing close to the sun and can operate with high levels of background flux.
The MLCD Link Development and Evaluation System (LDES) is addressing these technical challenges to realize a functional array receive system for the Mars to Earth Link. The remainder of this paper will focus on how LDES is solving these three technical challenges, with performance results from prototype hardware.

LDES System Overview
LDES will demonstrate the distributed aperture approach with four 0.8m telescopes that will be arrayed together to collect equivalent light to a single 1.6m telescope. The key parameters of the system are summarized in Table 1.
The Mars flight terminal will transmit high order pulse position modulation (64-PPM is the baseline) coupled with near capacity achieving turbo codes. To realize the full potential of this transmit waveform, the receiver must be capable of timing the arrival of individual signal photons to a small fraction of the transmit pulse width. We start by describing the detector technology we are developing for this project: Geiger mode APD arrays which are sensitive to single photons. Geiger mode APDs generate a nearly digital pulse for each detected photon arrival. The devices that Lincoln Laboratory is developing have high photon detection efficiency (PDE approaches 50%), with low dark count rates (DCR of about 40kcnts/sec). Geiger mode APDs must be refreshed after each detection event, which lead to a dead time for the detector. We minimize the effect of this dead time by employing an array of these detectors, and designing an integrated readout IC that allows for each detector to operate asynchronously.
The photon timing of the APD is clocked from a receiver oscillator that must be precisely synchronized to the transmitted waveform. This sync process is complicated by the large range of Doppler and rates associated with the mission, as well as the fact that there are very few photons arriving at a given telescope to measure timing with. The LDES sync approach uses a combination of very low phase noise oscillators that allow long measurements intervals coupled with a low overhead pilot tone embedded into the downlink waveform. With bandwidths of less than 2 Hz, we have demonstrated the ability to lock each telescope independently to the Mars waveform with timing jitter of less than 1/10 of the slot interval. Once synced to the downlink signal, the task of combining the telescopes can be carried out with near zero loss.
Operating LDES during daylight conditions drives the system design from a system performance point of view. To optimize communications performance, we use temporal, spatial, spectral, and polarization filtering to mitigate the amount of background photons that would potentially confuse the decoding system. Our approach for each of these mitigation techniques is shown in Table 2.
Lincoln has developed technology to meet each of the three challenges outlined above. Additionally, we have combined a prototype APD detector with the synchronization and timing system and the demodulation and decoder that will be employed for the operational system. We have successfully shown the ability to communicate reliably with photon efficiencies of ~1.5 incident photons per source bit. Actual measurements are shown in Figure 1.
* This work was sponsored by NASA under contract F19628-00-C-0002. Opinions, interpretations, conclusions and recommendations are those of the authors, and are not necessarily endorsed by the United States Government.