Greg Anderson Principal Engineer gma@apl.washington.edu Phone 206-543-4648 |

Research Interests

Sonar System Performance Modeling, Simulation, Optimization, Inversion, Statistics, Computational Intelligence

Biosketch

Gregory Anderson designs and develops tools for ocean sensor acoustic performance prediction, deployment optimization, and environmental parameter estimation (inversion), primarily for U.S. Navy anti-submarine warfare (ASW) tactical decision aids. His areas of expertise are system design, digital modeling and simulation, statistics, optimization, and parallel computing. Mr. Anderson joined the Laboratory's professional staff in 1990.

Education

B.S. Agriculture, University of Idaho, 1974

B.S. Applied Mathematics, University of Idaho, 1975

M.S. Electrical Engineering, University of Idaho, 1980

Publications |
2000-present and while at APL-UW |

3-D filter methods for sensor optimization Krout, D.W., J. Hsieh, M. Antonelli, M. Hazen, and G.M. Anderson, "3-D filter methods for sensor optimization," U.S. Navy J. Underwater Acoust., 61, 137-148, 2011. |
15 Jan 2011 |

An at-sea, autonomous, closed-loop concept study for detecting and tracking submerged objects Stevenson, J.M., et al., including J. Luby, R.T. Miyamoto, M. Grund, G. Anderson, and M. Hazen, "An at-sea, autonomous, closed-loop concept study for detecting and tracking submerged objects," U.S. Navy J. Underwater Acoust., 59, 671-690, 2009. |
1 Jun 2009 |

Distributed environmental inversion for multi-static sonar tracking Pitton, J., A. Ganse, G. Anderson, and D.W. Krout, "Distributed environmental inversion for multi-static sonar tracking," Proc., 9th International Conference on Information Fusion, 10-13 July, Florence, Italy, 6 pp., doi:10.1109/ICIF.2006.301710 (IEEE, 2006). |
More Info |
10 Jul 2006 |
|||||||

This paper presents an approach for adapting a tracking algorithm to the acoustic propagation environment. This adaptation is performed by incorporating the expected target signal-to-noise ratio (SNR) into the data association step through the measured contact amplitude. In this work, expected SNR is provided via acoustic modeling; estimates of bottom loss and scattering strength, required by the acoustic model, are obtained via inversion of the acoustic model based on measured multi-static sonar reverberation data. This paper shows that the use of distributed sensors provides improved estimates of the environmental parameters, and hence better estimates of the expected SNR. |

Sonar Environmental Parameter Estimation System (SEPES) for the Environmentally Adaptive AN/SQQ-89 (EA-89) Anderson, G.M., K.-Y. Moravan, and W.L.J. Fox, "Sonar Environmental Parameter Estimation System (SEPES) for the Environmentally Adaptive AN/SQQ-89 (EA-89)," APL-UW TR 0407, December 2004. |
30 Dec 2004 |

Orthogonal transformation of output principal components for improved tolerance to error Mann, T.P., C. Eggen, W. Fox, D. Krout, G. Anderson, M.A. El Sharkawi, and R.J. Marks II, "Orthogonal transformation of output principal components for improved tolerance to error," Proc., International Joint Conference on Neural Networks, 20-24 July 2003, 1290-1294, doi:10.1109/IJCNN.2003.1223881 (IEEE, 2003). |
More Info |
20 Jul 2003 |
|||||||

Preprocessing of data to be learned by a neural network is typically done to improve neural network performance. Output processing is especially important since it directly affects the influence of error in the hidden layers on the error of the neural network output. Principal component analysis is a commonly used preprocessing method that can improve the network performance by reducing the output dimensionality and reducing the number of parameters in a neural network model. Transforming the principal components of the outputs with an orthonormal matrix prior to scaling can further improve network performance. |

Sonar Environmental Parameter Estimation System (SEPES) Anderson, G.M., R.T. Miyamoto, M.L. Boyd, and J.I. Olsonbaker, "Sonar Environmental Parameter Estimation System (SEPES)," APL-UW TR 0101, April 2002. |
30 Apr 2002 |

Neural network training for varying output node dimension Jung, J.-B., M.A. El-Sharkawi, G.M. Anderson, R.T. Miyamoto, R.J. Marks II, W.L.J. Fox, and C.J. Eggen, "Neural network training for varying output node dimension," In Proc., International Joint Conference on Neural Networks, 15-19 July, Washington, D.C., 1733-1738, doi:10.1109/IJCNN.2001.938423 (IEEE, 2001). |
More Info |
15 Jul 2001 |
|||||||

Considers the problem of neural network supervised learning when the number of output nodes can vary for differing training data. The paper proposes irregular weight updates and learning rate adjustment to compensate for this variation. In order to compensate for possible over training, an a posteriori probability that shows how often the weights associated with each output neuron are updated is obtained from the training data set and is used to evenly distribute the opportunity for weight update to each output neuron. The weight space becomes smoother and the generalization performance is significantly improved. |

Team optimization of cooperating systems: Application to maximal area coverage Jung, J.-B., M.A. El-Sharkawi, G.M. Anderson, R.T. Miyamoto, R.J. Marks II, W.L.J. Fox, and C.J. Eggen, "Team optimization of cooperating systems: Application to maximal area coverage," In Proc., International Joint Conference on Neural Networks, 15-19 July, Washington, D.C., 2212-2217, doi:10.1109/IJCNN.2001.938510 (IEEE, 2001). |
More Info |
15 Jul 2001 |
|||||||

The composite effort of the system team, rather, is significantly more important than a single player's individual performance. We consider the case wherein each player's performance is tuned to result in maximal team performance for the specific case of maximal area coverage (MAC). The approach is first illustrated through solution of MAC by a fixed number of deformable shapes. An application to sonar is then presented. Here, sonar control parameters determine a range-depth area of coverage. The coverage is also affected by known but uncontrollable environmental parameters. The problem is to determine K sets of sonar ping parameters that result in MAC. The forward problem of determining coverage given control and environmental parameters is computationally intensive. To facilitate real time cooperative optimization among a number of such systems, the sonar input-output is captured in a feedforward layered perceptron neural network. |