Pen-Shu Yeh is a senior engineer at NASA's Goddard Space Flight Center, Greenbelt, Md. and a grant technical officer. She's also the person that helped create unique solutions for data compression for future missions with a team from the University of Idaho (U-Idaho). Those solutions are now going to fly on two of NASA's upcoming missions.
> Larger image
Compression chips mounted on board at Univ. of Idaho ready for radiation testing. Credit: Univ. of Idaho
Under Yeh's guidance, engineers at the University of Idaho's Center for Advanced Microelectronics Biomolecular Research (CAMBR) located in Post Falls, Idaho, have developed three unique advanced special purpose processors. These processors, along with a previously developed processor, will be used in the upcoming Landsat Data Continuity Mission (LDCM) and the Magnetospheric Multiscale Mission (MMS) missions that will study the Earth and Sun-Earth connection.
Development of the Algorithms
Yeh lead the creation of a tunable data compression algorithm that will increase science data return for NASA missions. "She demonstrated that the algorithm is superior to existing technology for onboard applications," said Gary Maki, Principal Investigator on the CAMBR U-Idaho team.
The data compression algorithm, adopted by the international Consultative Committee on Space Data Systems (CCSDS) in 2005, allows missions to precisely control compression factor, therefore the data rate from instruments. "The algorithm is over 50 times more complicated than the previous CCSDS compression standard," says Yeh. "However, this algorithm is much more versatile, but it does pose a great challenge for implementation."
NASA Goddard engineer Dave Fisher testing the channel coder implemented in a communication subsystem in a lab. Credit: NASA
In the 90's, Yeh also assisted the development of the first CCSDS lossless data compression standard.
Yeh was also the grants officer for the design of an algorithm, called Low Density Parity Check (LDPC) coder. The LDPC code was created by Prof. Shu Lin of University of California at Davis and based on a recently rediscovered technique for protecting data fidelity in a communication channel. This was part of a larger effort called the Bandwidth Efficient Channel Coding Project under the aegis of NASA's Space Communications and Navigation program. Wai Fong of Goddard is the Principal Investigator for the Bandwidth Efficient Channel Coding project and the LDPC effort; and his advocacy at CCSDS led the LDPC code to a soon to be standard. This new algorithm is shown to perform better than previous CCSDS standards.
The team of specialized processor chip designers at U-Idaho were able to implement these difficult algorithms and create high speed versions suitable for various space missions. They tested the processor chips in October. Aeroflex, a company in Colorado Springs is now testing for flight qualification.
"We at NASA are extremely fortunate to have the U-Idaho team developing these chips from given algorithms;" says Yeh. "The team's capability allows high-end specialized space processors to be developed at a budget level considered 'shoe-string' by industry standards. I hope other NASA centers and maybe other government agencies can benefit from this capability and the developed technology."
What Do the Chips Do?
CAMBR/Univ. of Idaho engineers Chad Orbe, Paul Winterrowd, Ron Nelson doing radiation testing of the compression chips at Texas A&M University. Credit: Univ. of Idaho
Generally speaking, two of the chips - the Discrete Wavelet Transformer and Bit Plane Encoder – compress science data to decrease the data volume to be sent back to Earth. "If we didn't produce these chips, instruments on the MMS satellites would have to greatly reduce science data return or use an inferior technique with less performance," Yeh said.
Chad Orbe, research engineer, spent six years building the BPE chip. "Some teams felt that it was impossible to build a single high speed BPE chip. So the impossible takes a little bit longer," said Sterling Whitaker, research professor at U-Idaho.
The third chip – the Low Density Parity Check (LDPC) encoder – adds redundancy information to data before it is transmitted to Earth. This allows for error corrections caused by signal degradation. Instead of just adding binary information of 1's and 0's, it adds a probability factor to each bit with just how strong of a 1 it is or how strong of a 0 it is. This chip is rated 7/8, meaning 7 out of 8 bits are data and 1 out of 8 is for error correction.
This allows for efficient use of available bandwidth. With this coder, scientists are approaching the theoretical limit for the best you can do called the Shannon limit. The bit/error rate is 1 in 10^14, comparable to the error rate seen in hard drives. "We're getting as reliable as talking to our hard drives on our desk, except here they're millions of miles away in space," said Whitaker, one of the LDPC designers.
A fourth chip, the previously developed lossless data compression chip called Universal Source Encoder for Space (USES) and the newly developed LDPC channel coder will be implemented in the LDCM mission.
The Challenges Overcome
The biggest challenge to insert custom-designed chips into space missions is the potential chip failure caused by radiation environment. The U-Idaho team is able to utilize a sleek technology, the radiation-hardness-by-design (RHBD) method, to overcome the failure mechanism. With this technique, the radiation tested LDPC coder shows a worst-case of failure once every 1700 years. The DWT and BPE chips are being tested and expect to fair even better. Shown in the attached picture are CAMBR engineers Paul Winterrowd, Chad Orbe, and Ron Nelson testing the BPE and DWT at the radiation test facility at Texas A&M.
The Missions the Chips Will Fly Inside
LDCM is the future of Landsat satellites. It will continue to obtain valuable data and imagery to be used in agriculture, education, business, science, and government. The Landsat Program provides repetitive acquisition of medium resolution near-infrared and visible data of the Earth's surface on a global basis.
The MMS will have four spacecraft acting in concert to measure the three-dimensional structure and motion of magnetic and electric fields around the Earth. MMS will determine the small-scale basic plasma processes which transport, accelerate and energize plasmas in thin boundary and current layers – and which control the structure and dynamics of the Earth's magnetosphere. MMS will for the first time measure the 3-D structure and dynamics of the key magnetospheric boundary regions, from the subsolar magnetopause to the distant tail. Launch is planned for 2014. Karen Halterman, MMS Project Manager at Goddard said, "MMS plans to use the data compression chips to increase the amount of science data that can be sent to the ground on two instruments: the Fast Plasma Investigation and the Hot Plasma Composition Analyzer."
"Currently, this new technology is not available anywhere else in the world," Maki said.
> Landsat Data Continuity Mission site
> Magnetospheric Multiscale Mission site