Session
Technical Session X: Communications
Abstract
The Galileo Near-Infrared Mapping Spectrometer (NIMS) is a sophisticated multispectral instrument which was developed to study both geological and the spatial compositional aspects of atmospheres at Jupiter. Since its original development, the communication capability of the Galileo spacecraft has been severely reduced as the result of a failure of ground controllers to open the main antenna. The data rate which will be available for all instruments at the first Jupiter encounter in 1995 has been reduced by several orders of magnitude (even after recent numerous improvements to data link efficiency. ) Because of this severe limitation, real-time return of science data is now secondary. The primary means of returning Galileo science data will now be to store encounter data using the on-board tape recorder and then gradually play it back over an extended period of time. But even this time period is restricted to approximately one month since the tape recorder must be ready for additional encounters as the spacecraft orbits Jupiter. The NIMS instrument designers built-in extraordinary flexibility to adjust parameters that affect spectral resolution, spatial resolution and field of view. Each of these affects the amount of data generated. As a result of the severe communication constraints, this flexibility has become of great importance in allowing the NIMS scientists to maximize their science return value. The NIMS instrument will also incorporate adaptive lossless data compression (by factors of 2:1 to 3:1) to further improve its return capabilities. This paper describes how this is accomplished such that the flexibility to arbitrarily alter instrument parameters is not jeopardized. The paper will also discuss a solution to the problem of NIMS allocation of a fixed number of bits over a full tape load in the face of changing data characteristics and associated greatly diverse data compression factors. The approach described is a global compression rate allocation strategy, designed to achieve nearly-optimal bit usage over a complete tape load. Data history and a priori planning are the major inputs to the strategy's control structure. This mechanism then autonomously makes adjustments in specific parameters which affect data generation to ensure compliance with the overall bit constraint. By "looking ahead" the strategy results in more accurate decision making which minimizes unnecessary reductions in data quality and/or ensures that unused bits will not be wasted.
Lossless Compression and Rate Control for the Galileo NIMS Instrument
The Galileo Near-Infrared Mapping Spectrometer (NIMS) is a sophisticated multispectral instrument which was developed to study both geological and the spatial compositional aspects of atmospheres at Jupiter. Since its original development, the communication capability of the Galileo spacecraft has been severely reduced as the result of a failure of ground controllers to open the main antenna. The data rate which will be available for all instruments at the first Jupiter encounter in 1995 has been reduced by several orders of magnitude (even after recent numerous improvements to data link efficiency. ) Because of this severe limitation, real-time return of science data is now secondary. The primary means of returning Galileo science data will now be to store encounter data using the on-board tape recorder and then gradually play it back over an extended period of time. But even this time period is restricted to approximately one month since the tape recorder must be ready for additional encounters as the spacecraft orbits Jupiter. The NIMS instrument designers built-in extraordinary flexibility to adjust parameters that affect spectral resolution, spatial resolution and field of view. Each of these affects the amount of data generated. As a result of the severe communication constraints, this flexibility has become of great importance in allowing the NIMS scientists to maximize their science return value. The NIMS instrument will also incorporate adaptive lossless data compression (by factors of 2:1 to 3:1) to further improve its return capabilities. This paper describes how this is accomplished such that the flexibility to arbitrarily alter instrument parameters is not jeopardized. The paper will also discuss a solution to the problem of NIMS allocation of a fixed number of bits over a full tape load in the face of changing data characteristics and associated greatly diverse data compression factors. The approach described is a global compression rate allocation strategy, designed to achieve nearly-optimal bit usage over a complete tape load. Data history and a priori planning are the major inputs to the strategy's control structure. This mechanism then autonomously makes adjustments in specific parameters which affect data generation to ensure compliance with the overall bit constraint. By "looking ahead" the strategy results in more accurate decision making which minimizes unnecessary reductions in data quality and/or ensures that unused bits will not be wasted.
