Atmospheric turbulence can significantly degrade the bit error rate performance of a free-space laser communication link. We describe an approach for calculating the average bit error rate of a direct detection binary optical communication link in clear-air atmospheric turbulence, and discuss strategies for optimizing system performance.

Free-space atmospheric direct detection optical communication at 980 nm with multiple quantum well retro-reflectors is attractive because of the availability of high power, good beam quality 980 nm laser diodes as transmitters and very low noise silicon avalanche photodiodes for use in the receivers. The paper will report the results of measurements of intensity fading statistics for broadband, low coherence volume 980 nm laser sources. In addition, results of measurement of both intensity spatial correlation lengths and optical field spatial correlation lengths will be described. The former determines receiver performance and the latter the optimal positioning of arrays of MQW retro-reflectors to make best use of the spatial diversity of the atmospheric channel.

We evaluate a simple model for predicting and understanding the structural behavior of Cn2 for a specific location, date, time, and given environmental parameters. This model is compared with Cn2 data taken at the Chesapeake Bay Detachment of the Naval Research Laboratory in Chesapeake Beach, Maryland. This simplified model predicts and explains the fluctuation in Cn2 reasonably well, and also shows that Cn2 is a strong function of solar irradiation.

Fading due to scintillation has been measured at path lengths of up to approximately 13.3 Km. Theoretical fading and aperture averaging models based on the gamma-gamma distribution were used to generate predictions of the fade parameters. Data collected at the optical receiver was compared to the theoretical values to determine if the fade theory remains valid over greater distances then has previously been investigated in a tropical environment.

The European Space Agency (ESA) has launched the geostationary data-relay satellite ARTEMIS with one of its payloads being a laser communication terminal (LCT). The LCT is used within the semiconductor-laser intersatellite link experiment (SILEX) to receive Earth observation data transmitted from a similar LCT onboard the SPOT-4 satellite. To support SILEX, ESA has also reached an agreement with the Instituto de Astrofisica de Canarias (JAC) to build the Optical Ground Station (OGS), in the Teide Observatory ofthe IAC. ARTEMIS and the OGS are an ideal and unique test-bed to study and characterise laser beam propagation through atmospheric turbulence. Theoretical models of laser beam propagation through atmospheric turbulence have been reviewed and developed, to predict the performance of the optical links from the propagation and communication point of view. Special effort has been invested in modelling the uplink effects. Optical link experiments have been planned in detail, to gather the necessary data required to be statistically representative, to compare the results with theoretical predictions and to validate and adjust the theoretical models. This comparison will pave the way towards the implementation of deep-space laser communication links. The first results ofthese experiments, presenting the theoretical models, analysing separately downlink and uplink measurements, and comparing the data with the theoretical predictions, are presented.

The e ffects o f a tmospheric turbulence o n the g round—to-space p ropagation p ath a nd P oisson sh ot n oise o n a n a ctive laser-based imaging system for high-resolution imaging of Geosynchronous (GEO) satellites are investigated using a wave-optics simulation code. The phase and scintillation statistics in tilt corrected and uncorrected beams are examined at the top ofthe atmosphere and at the satellite. The effects of intensity and phase variations in the illuminating beams on the fringe visibility and spatial frequency of the interference pattern formed by the illuminating beams at the satellite are investigated. The Fourier phase variance caused by Poisson shot noise and turbulence on the uplink path is evaluated. We found that tilt correction reduces the scintillation in the laser beam at the satellite. In the Fourier telescopy system the scintillation variance at the edge of the beam is reduced by a factor of up to 3 for a tilt corrected beam. Long-range propagation in free space reduces scintillation in the illuminating beam. The scintillation variance in the Fourier telescopy system on the optical axis at the satellite is reduced by 26% to 36 %, as compared to that at the top of the atmosphere. The latter is due to diffraction of the laser beam in free space and enhancement of the spatial coherence of the beam described by the Van Cittert-Zernike theorem. The intensity spatial correlation scale in the scintillation pattern exceeds the satellite dimensions. This leads to a so-called residual turbulent scintillation effect, when the scintillation in the illuminating beam modulates the total reflected energy flux. As a result, an arbitrarily large receiver on the ground cannot average the received signal variations. This degrades the Fourier telescopy system performance. Also intensity and phase variations in the illuminating beams degrade the interference pattern formed at the satellite. The turbulence effect on the fringe visibility is stronger at high spatial frequencies. Intensity variations in the illuminating beams degrade the fringe visibility the most. Poisson shot noise and scintillation on the uplink path strongly impacts the Fourier phase of the object. In the turbulent atmosphere the Fourier phase variance increases by a factor of 1 .5-3, as compared to that in free space. The increase of the phase variance is caused by a non-linear interaction between the two statistically independent noise sources. For the nominal signal level and number of averaged pulses the Fourier phase variance is less than 0.1, or ()L I20)2 . This suggests that the Fourier telescopy method is feasible.

The effect of a (spatially) partially coherent quasi-monochromatic lowest order Gaussian-beam on the direct detection system is studied in weak and strong atmospheric turbulence. The analytic expression for the scintillation index has been developed and analysed as a function of the spatial correlation distance (lc) and the correlation time (ts), associated with the source, as well as the strength of turbulence and the integration time (td) of the detector. The probability of fade is also studied as a function of relative detector speed td/ts. The limiting cases of slow (td>>ts) and fast (td<