In inverse synthetic aperture radar (ISAR) imaging, the target rotatio
nal motion with respect to the radar line of sight contributes to the
imaging ability, whereas the translational motion must be compensated
out. This paper presents a novel two-step approach to translational mo
tion compensation using an adaptive range tracking method for range bi
n alignment and a recursive multiple-scatterer algorithm (RMSA) for si
gnal phase compensation. The initial step of RMSA is equivalent to the
dominant-scatterer algorithm (DSA). An error-compensating point sourc
e is then recursively synthesized from the selected range bins, where
each contains a prominent scatterer, Since the clutter-induced phase e
rrors are reduced by phase averaging, the image speckle noise can be r
educed significantly. Experimental data processing for a commercial ai
rcraft and computer simulations confirm the validity of the approach.