Hierarchical Wavelet Representations of Ship Radar Returns

Abstract

In this paper we investigate the problem of efficient representations of large databases of pulsed radar returns in order to economize memory requirements and minimize search time. We use synthetic radar returns from ships as the experimental data. We motivate wavelet multiresolution representations of such returns. We develop a novel algorithm for hierarchically organizing the database, which utilizes a multiresolution wavelet representation working in synergy with a Tree Structured Vector Quantizer (TSVQ), utilized in its clustering mode. The tree structure is induced by the multiresolution decomposition of the pulses. The TSVQ design algorithm is of the "greedy" type. We show by experimental results that the combined algorithm results in data search times that are logarithmic in the number of terminal tree nodes, with negligible performance degradation (as measured by distortion-entropy curves) from the full search vector quantization. Furthermore we show that the combined algorithm provides an efficient indexing scheme (with respect to variations in aspect, elevation and pulsewidth) for radar data which is equivalent to a multiresolution aspect graph or a reduced target model. Promising experimental results are reported using high quality synthetic data.