NASA SBIR 2003 Solicitation


PROPOSAL NUMBER:03-E3.02-8986 (For NASA Use Only - Chron: 034019)
SUBTOPIC TITLE:Automation and Planning
PROPOSAL TITLE:Reconfigurable Environmentally Aware Computing Technology for Earth Observing Systems (7284-060)

SMALL BUSINESS CONCERN (Firm Name, Mail Address, City/State/Zip, Phone)
Physical Sciences Inc
20 New England Business Ctr
Andover ,MA 01810 - 1077
(978) 689 - 0003

PRINCIPAL INVESTIGATOR/PROJECT MANAGER (Name, E-mail, Mail Address, City/State/Zip, Phone)
Robin L Coxe
20 New England Business Ctr
Andover ,MA  01810 -1077
(978) 689 - 0003
U.S. Citizen or Legal Resident: Yes

Over the past decade, many research groups have developed reconfigurable computing systems built from Field Programmable Gate Arrays (FPGAs) for on-board processing of remote sensing data. Physical Sciences Inc. (PSI) proposes to exploit the run-time reconfiguration capabilities of SRAM-based FPGAs to demonstrate an innovative extension to the current state-of-the-art -- an autonomous, adaptive digital signal processor. PSI?s Reconfigurable Environmentally Aware Computing Technology (REACT) will automatically customize applications executing on a reconfigurable FPGA processing system based on external inputs. In Phase I, we will validate the REACT concept by processing images from a digital video camera with a variable-precision 2D integer wavelet transform in a Xilinx Virtex-II FPGA. Wavelet transforms enable image data compression with dynamically tunable compression ratios. Quantization of the filter coefficients will be controlled by a master state machine with inputs from a GPS receiver and PSI?s SEU Alarm, a compact sensor that monitors the surrounding environment for radiation hazards. Furthermore, PSI will develop a Phase II system design capable of performing a full-scale demonstration using hyperspectral imagery.

PSI will build a high-performance, agile image compression engine on an FPGA processor. The compression scheme will be based on the emerging CCSDS standard, ensuring its applicability to a wide array of NASA missions. The compression ratio and data rate will be adjusted on the fly via plug-and-play inputs to a master controller unit, enabling future NASA Earth Sciences satellites to accommodate next-generation hyperspectral instruments and radar arrays-- with increasingly high data rates-- on platforms that make maximally efficient use of downlink bandwidth, size, weight, and power. The GOES-R Hyperspectral Environmental Suite (HES) stands out as a potential near-term beneficiary.

The results of the proposed SBIR effort will be equally applicable to military space missions and commercial communications satellites. Although PSI intends to develop a specific application during the course of this program, the REACT concept can also be extended to a myriad of terrestrial and underwater digital signal processing applications that would benefit from an adaptive processing component. Examples include adaptive filtering to autonomously detect countermeasures on torpedo sonar systems and dynamically tuned software radio transceivers for mobile communications.