Lossless Compression Handbook

CHAPTER 16

The CCSDS Lossless Data

Compression Recommendation

for Space Applications 1

PEN-SHU YEH

16.1 INTRODUCTION

In the late 1980s, when the author started working at the Goddard Space Flight Center for the

National Aeronautics and Space Administration (NASA), several scientists were in the process

of formulating the next generation of Earth-viewing science instruments, such as the Moderate

Resolution Imaging Spectroradiometer. The instrument would have over 30 spectral bands and

would transmit an enormous amount of data through the communications channel. In an attempt

to solve the problem associated with buffering a large amount of data on board and transmitting

it to ground stations, the author was assigned the task of investigating lossless compression

algorithms for space implementation to compress science data in order to reduce the requirement

on bandwidth and storage.

Space-based algorithm implementations face extra constraints relative to ground-based imple-

mentations. The difference is mainly to the lack of computing resources in space and possible

bit errors incurred in the communication channel. A spacecraft is a physically compact environ-

ment where each subsystem (power, attitude control, instrument, data handling, telemetry, etc.)

is given only limited budget in terms of power, mass, and weight. Any computation other than

that which is absolutely necessary is normally not supported. Taking all these constraints into

consideration, a set of requirements was first formulated for selecting a lossless compression

algorithm:

a. The algorithm must adapt to changes in data statistics in order to maximize compression

performance.

b. The algorithm must be implemented in real time with small memory and little power usage.

1 Author's note: Part of the material from this chapter is taken from the Consultative Committee for Space Data

Systems (CCSDS) publications [ 1, 2] for which the author is a major contributor.

Copyright 2003, Elsevier Science (USA). 3 1 1

All rights reserved.