Free-space optical communication is often desired between two nodes having different capabilities. Using a modulating retroreflector (MRR) shifts most of the power, weight, and pointing requirements to one end of the link, allowing the other end to be extremely small, low-power, and requiring only rough pointing (to within +/-15 degrees). For a 5 Mbls link at 2 km, our entire MRR package including drive electronics weighs only 8.5 g and requires 60 mW. This same device has also been arrayed to further relax the pointing requirements. In the simplest MRR link, an unmodulated (CW) laser interrogates a MRR comprised of an absorptive modulator and a retroreflector. If the interrogation beam is within the retroreflector's field of view (FOV), the beam will return to the interrogator with data impressed on it. In this discussion, we present the range of MRRs developed since 1998 at the Naval Research Laboratory (NRL).NRL's MRRs include both corner cube retroreflector and cat's eye retroreflector types, both individually and in arrays. Each variation has unique characteristics which may be beneficial or disqualifying in different situations. Size, weight, FOV, power consumption, cost, speed x range product, modulator type, ruggedness, time-tomarket, and requirement for angle division multiplexing are all determining factors in MRR design and retroreflector choice. We compare strengths, weaknesses, and give current link performance data for several different systems. Links demonstrated include shore-toshore, boat-to-shore, and UAV-to-ground over a wide range ofdistances and data rates.

We present an optical turbulence model that has evolved from the PAMELA model. After a preliminary report in SPIE 2003 it became apparent that more data was needed to refine this adaptive model. This led us to take twelve months of over-land data (~100 meters pathlength) at the Chesapeake Bay Detachment of the Naval Research Lab. We present data throughout the year with varying environments with comparison with the model prediction. Our recent modification includes segmenting the windspeed to 3 sections, morning, afternoon, and night for better fitting. This is an attempt to incorporate variable wind speed into the model which is known to contribute significantly to the turbulence in the atmosphere. In addition, we present preliminary results from the over-the-bay data (10 km pathlength).

No abstract

No abstract

No abstract