High-Resolution Range Profile Classifiers Require Aspect-Angle Awareness

TL;DR

It is shown that a causal Kalman filter can estimate aspect angles online with a median error of 5{\textdegree}, and that training and inference with estimated angles preserves most of the gains, supporting the proposed approach in realistic conditions.

Abstract

We revisit High-Resolution Range Profile (HRRP) classification with aspect-angle conditioning. While prior work often assumes that aspect-angle information is incomplete during training or unavailable at inference, we study a setting where angles are available for all training samples and explicitly provided to the classifier. Using three datasets and a broad range of conditioning strategies and model architectures, we show that both single-profile and sequential classifiers benefit consistently from aspect-angle awareness, with an average accuracy gain of about 7% and improvements of up to 10%, depending on the model and dataset. In practice, aspect angles are not directly measured and must be estimated. We show that a causal Kalman filter can estimate them online with a median error of 5{\textdegree}, and that training and inference with estimated angles preserves most of the gains, supporting the proposed approach in realistic conditions.

Figures (5)

• Figure 1: HRRP structure and aspect-angle dependence: an HRRP sums the echoes of dominant scatterers within each range cell; changing $\texttt{asp}$ alters the coarse-scale signature.
• Figure 2: HRRP of a ship at multiple aspect angles.
• Figure 3: Length on Range Profile (LRP) mfn across aspect angles.
• Figure 4: Sorted MMSI class frequencies for Ship (A) and Ship (B) datasets.
• Figure 5: Kalman-based aspect-angle estimation error over 100k segments (all segments and worst 10%).