Study of Two ANN Digital Implementations of a Radar Detector Candidate to an On-Board Satellite Experiment

The Microelectronics and Photonics Testbed (MPTB) is a scientific satellite carrying twenty-four experiments on-board in a high radiation orbit since November 1997. The first objective of this paper is to summarize one year flight results, telemetred from one of its experiments, a digital “neural board” programmed to perform texture analysis by means of an Artificial Neural Network (ANN). One of the attractive features of MPTB neural board is its possibility of re-programmation from the earth. The second objective of this paper is to present two new ANN architectures, devoted to radar or sonar detection, intended to be telecommanded on the MPTB neural board. Their characteristics (performances and potential robustness with respect to parameter deviations due to the interaction with charged particles) are compared in order to predict their behavior under radiation.

[1]  Juan-Manuel Torres-Moreno,et al.  Apprentissage et généralisation par des réseaux de neurones : étude de nouveaux algorithmes constructifs. (Machine learning and generalization by Neural Networks: new constructive algorithms) , 1997 .

[2]  Scott E. Decatur,et al.  Application of neural networks to terrain classification , 1989, International 1989 Joint Conference on Neural Networks.

[3]  Juan-Manuel Torres-Moreno,et al.  Efficient Adaptive Learning for Classification Tasks with Binary Units , 1998, Neural Computation.

[4]  M. W. Roth.  Survey of neural network technology for automatic target recognition , 1990, IEEE Trans. Neural Networks.

[5]  Diego Andina,et al.  A comparison of criterion functions for a neural network applied to binary detection , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[6]  W. Miller,et al.  Office of Naval Research contributions to neural networks and signal processing in oceanic engineering , 1992 .

[7]  Raoul Velazco,et al.  Analysis and Improvement of Neural Network Robustness for On-Board Satellite Image Processing , 1997, ICANN.

[8]  Diego Andina,et al.  Quasi-optimum detection results using a neural network , 1996, Proceedings of International Conference on Neural Networks (ICNN'96).

[9]  W. Root.  An introduction to the theory of the detection of signals in noise , 1970 .

[10]  Raoul Velazco,et al.  Robustness against S.E.U. of an artificial neural network space application , 1995 .

[11]  Diego Andina,et al.  Optimization of a neural network applied to pulsed radar detection , 1996, 1996 8th European Signal Processing Conference (EUSIPCO 1996).

[12]  Bruno Raffin,et al.  Learning and Generalization with Minimerror, A Temperature-Dependent Learning Algorithm , 1995, Neural Computation.

[13]  Raoul Velazco,et al.  Artificial neural network robustness for on-board satellite image processing: results of upset simulations and ground tests , 1997 .

[14]  Jess Marcum,et al.  A statistical theory of target detection by pulsed radar , 1948, IRE Trans. Inf. Theory.