Title of Paper: Real-Time Data Processing Onboard Remote Sensor
Platforms
Principal Author: Mr. Scott Lindell
Abstract: With EO-1 Hyperion and MightySat in orbit and
Warfighter-1 scheduled for launch this summer NASA and the DoD are showing its
continued commitment to HSI. As these systems are fielded and made operational,
the large amounts of sensor data generated can result in expensive
communication and data handling systems. Lockheed Martin with its experience
with spacecraft design and expertise in developing special purpose real-time
onboard processors has teamed with Applied Signal & Image Technology (ASIT)
who has an extensive heritage in HSI to develop a real-time and intelligent
onboard processing (OBP) system for reducing the HSI sensor downlink
requirements. Our objective is to reduce the downlink requirement by a factor
> 100, while retaining the necessary spectral fidelity of the sensor data
that is needed to reach the many science, military, and intelligence goals of
these systems. Our initial spectral compression simulations leverage the
commercial-off-the-shelf (COTS) spectral exploitation algorithms for
segmentation, material identification and spectral compression that ASIT has
developed. ASIT will also support the modification and integration of this COTS
software into the OBP. Other commercially available COTS software for lossless
spatial compression will also be employed as part of the overall compression
processing sequence. Demonstrations presenting our initial capability will be
presented at the August meeting.
Next year and in the 3rd year elements of a high-performance
re-configurable OBP will be brass-boarded to implement proven preprocessing
steps that distill the HSI data stream in both spectral and spatial dimensions.
The system will intelligently reduce the volume of data that must be stored,
transmitted to the ground and processed while minimizing the loss of
information. Data will be reduced in a manner that maintains the necessary
spectral, spatial and radiometric properties of the important data and creates
scientifically useful products, which increases the utility and accessibility
of the data.