Parametric reconstruction of internal building structures via canonical scattering mechanisms

In this paper, we describe a model-based, non-linear reconstruction method for mapping internal building structures using through-wall radar data. We based the model on canonical geometry constructs that are commonly used in construction practices. These constructs are then formulated as sets of simple scattering mechanisms, which can be estimated from the data. Our non-linear approach employs an iterative, conditional estimation method as a function of the intervening structures between the sensor and the object under consideration. Specific associations of scattering mechanisms are then used to re-create various building structures such as walls, doors, stairs, etc. We discuss some examples of estimating specific scattering mechanisms and a model-based reasoning approach for assembling them to reconstruct the interior structure of a building.

[1]  Stéphane Mallat,et al.  Matching pursuits with time-frequency dictionaries , 1993, IEEE Trans. Signal Process..

[2]  N.S. Subotic,et al.  Adaptive decomposition in electromagnetics , 1997, IEEE Antennas and Propagation Society International Symposium 1997. Digest.

[3]  D. Donoho,et al.  Basis pursuit , 1994, Proceedings of 1994 28th Asilomar Conference on Signals, Systems and Computers.

[4]  Bhaskar D. Rao,et al.  Sparse signal reconstruction from limited data using FOCUSS: a re-weighted minimum norm algorithm , 1997, IEEE Trans. Signal Process..