A 25 Mbit/s direct-detection optical communication system that used Q=4 PPM signaling was constructed and its performance measured under laboratory conditions. The system used a single mode AlGaAs laser diode ( lambda =834 nm) and low-noise silicon avalanche photodiode (APD). A procedure is given to numerically compute system performance which uses the nearly exact Webb's approximation of the true Conradi distribution for the APD output that does not require excessive amounts of computer time (a few CPU minutes on VAX 8600 per system operating point). Comparison revealed that modeling the APD output as a Gaussian process under conditions of negligible background radiation and low (less than 10/sup -12/A) APD bulk leakage currents leads to substantial underestimates of optimal APD gain and overestimates of system bit error probability. Examples are given which illustrate the breakdown of the Gaussian approximation in assessing system performance. The measured performance of the system was found to be in excellent agreement with the performance predicted by the nearly exact computational procedure. This system achieved a bit error probability of 10/sup -6/ at a received signal energy corresponding to an average of 60 absorbed photons/bit and optimal APD gain of 700. >

Gaussian approximation versus nearly exact performance analysis of optical communication systems with PPM signaling and APD receivers

This correspondence compares OOK and low-order PPM signaling formats in terms of bit error probabilities versus required signal counts per bit. The results show that QPPM requires 3 dB less signal than OOK, while BPPM requires the same or slightly more than OOK for the same performance. Optimum APD gain values are from 200 to 400. When using QPPM, k_{eff} = 0.006 , and optimum gain, 60 signal counts/bit are required at 500 Mbits/s for a 10-6bit error probability.

Performance of OOK and Low-Order PPM Modulations in Optical Communications When Using APD-Based Receivers

Performance bounds of free space optical (FSO) systems with multiple laser transmitters and multiple avalanche photodetectors (APDs) are evaluated for terrestrial, line-of- sight communication. Specifically, analytical upper bounds on average uncoded error rate in turbulent multiple-input multiple- output (MIMO) channels with Q-ary pulse position modulation (PPM) are derived for arbitrary observable densities, and are used to quantify APD-based performance improvements over pin detector arrays in non-fading, lognormal, and negative exponential channels. Moreover, it is shown that despite the high complexity of APD array statistics, these performance gains can be accurately predicted using a jointly-Gaussian model for APD array outputs. Both maximum likelihood (ML) and sub-optimal combining rules are derived, and a simple equal gain combiner (EGC) that performs close to optimum is demonstrated. Finally, APD gain selection is discussed.

Performance Bounds for Free-Space Optical MIMO Systems with APD Receivers in Atmospheric Turbulence

Small satellites, particularly CubeSats, have become
popular platforms for a wide variety of scientific, commercial and
military remote sensing applications. Inexpensive commercial o the
shelf (COTS) hardware and relatively low launch costs make these
platforms candidates for deployment in large constellations that
can offer unprecedented temporal and geospatial sampling of the
entire planet. However, productivity for both individual and
constellations of CubeSats in low earth orbit (LEO) is limited by
the capabilities of the communications subsystem. Generally, these
constraints stem from limited available electrical power, low-gain
antennas and the general scarcity of available radio spectrum. In
this thesis, we assess the ability of free space optical
communication (lasercom) to address these limitations, identify key
technology developments that enable its application in small
satellites, and develop a functional prototype that demonstrates
predicted performance. We first establish design goals for a
lasercom payload archi- tecture that offers performance
improvements (joules-per-bit) over radio-frequency (RF) solutions,
yet is compatible with the severe size, weight and power (SWaP)
constraints common to CubeSats. The key design goal is direct
LEO-to-ground downlink capability with data rates exceeding 10
Mbps, an order of magnitude better than COTS radio solutions
available today, within typical CubeSat SWaP constraints on the
space terminal, and with similar COTS and low-complexity
constraints on the ground terminal. After defining the goals for
this architecture, we identify gaps in previous implementations
that limit their performance: the lack of compact, power-efficient
optical transmitters and the need for pointing capability on small
satellites to be as much as a factor of ten better than what is
commonly achieved today. One approach is to address these
shortcomings using low-cost COTS components that are compatible
with CubeSat budgets and development schedules. In design trade
studies we identify potential solutions for the transmitter and
pointing implementation gaps. Two distinct transmitter
architectures, one based on a high-power laser diode and another
using an optical amplifier, are considered. Analysis shows that
both configurations meet system requirements, however, the optical
amplifier offers better scalability to higher data rates. To
address platform pointing limitations, we dene a staged control
framework incorporating a COTS optical steering mechanism that is
used to manage pointing errors from the coarse stage (host
satellite body-pointing). A variety of ne steering solutions are
considered, and microelectromechanical systems (MEMS) tip-tilt
mirrors are selected due to their advantage in size, weight and
power. We experimentally validate the designs resulting from the
trade studies for these key subsystems. We construct a prototype
transmitter using a modified COTS fiber amplifier and a
directly-modulated seed laser capable of producing a 200mW average
power, pulse position modulated…

Optical communications for small satellites

http://ieeexplore.ieee.org/iel5/49/4481388/04481390.pdf

The output voltage of an optical receiver is statistically a mixture random variable, composed of the sum of a discrete count variable and a continuous Gaussian thermal noise variable. Based on some computer analyses, it is shown here that threshold crossing probabilities using the mixture density can be reliably approximated by integrations of an equivalent continuous Gaussian density. The conditions for this approximation and an accurate assessment of its error are derived.

G. Prati

Papers

On Gaussian Error Probabilities in Optical Receivers