Communication channel effects pose significant challenges to networking in a mobile free space optical environment. Link outages due to clouds and bursty fades due to atmospheric turbulence are chief examples. However, networking provides a powerful means to mitigate channel effects. Results from ITT's architecture effort on the DARPA THOR program identify the key trades and demonstrate how innovations at all protocol layers can be leveraged to overcome free-space optical communications challenges and successfully achieve wireless optical networking. Architecture recommendations were validated through extensive network modeling and simulation that included optical link budgets, cloud scenes, atmospheric turbulence and airborne node trajectories. With adequate link redundancy, link outages due to cloud occlusions are mitigated through the use of optimized network layer re-routing. While advances in receiver sensitivity technology and high power lasers offer larger link margins to mitigate atmospheric turbulence-induced scintillation, the use of error correction coding at the physical layer and link layer retransmission can reduce the dependency on link margin for small disadvantaged nodes. Quality of service-based provisioning (differentiated services) is employed to provide expedited and assured service for high priority, low-latency data. Finally, an alternative to TCP is recommended to appropriately respond to packet loss resulting from intermittent, transient outages due to scintillation. Simulation results demonstrate that the switching and self-healing power of the network can be leveraged to mitigate the challenges of optical channels, including cloud occlusions at up to 80 % cloud cover.
[1]
Lennart Lindegren,et al.
Atmospheric Intensity Scintillation of Stars. III. Effects for Different Telescope Apertures
,
1998
.
[2]
Lennart Lindegren,et al.
ATMOSPHERIC INTENSITY SCINTILLATION OF STARS, I. STATISTICAL DISTRIBUTIONS AND TEMPORAL PROPERTIES
,
1997
.
[3]
Christopher C. Davis,et al.
Free-space optical wireless links with topology control
,
2002,
SPIE Optics + Photonics.
[4]
William C. Danchi,et al.
Atmospheric Fluctuations: Empirical Structure Functions and Projected Performance of Future Instruments
,
1992
.
[5]
Kwo-Sen Kuo,et al.
Clustering, randomness, and regularity in cloud fields: 5. The nature of regular cumulus cloud fields
,
1998
.
[6]
Judith A. Curry,et al.
Characteristics of small tropical cumulus clouds and their impact on the environment
,
1998
.
[7]
Robert F. Cahalan,et al.
Fractal Statistics of Cloud Fields
,
1989
.
[8]
Vernon G. Plank,et al.
The Size Distribution of Cumulus Clouds in Representative Florida Populations
,
1969
.
[9]
Douglas Comer,et al.
Internetworking with TCP/IP
,
1988
.
[10]
Ronald M. Welch,et al.
Clustering, randomness, and regularity in cloud fields: 4. Stratocumulus cloud fields
,
1992
.
[11]
R. A. Silverman,et al.
Wave Propagation in a Turbulent Medium
,
1961
.
[12]
Ronald M. Welch,et al.
Cumulus Cloud Field Morphology and Spatial Patterns Derived from High Spatial Resolution Landsat Imagery
,
1990
.