Noncooperative target recognition (NCTR) is often based on the use of inverse synthetic aperture radar (ISAR) images. Although ISAR systems are also able to provide an image of the target when the target's motions are not known a priori, the interpretation of the obtained image is not straightforward, and therefore, the target recognition becomes less effective. This is due to the fact that the image projection plane (IPP), which depends on both the radar position and the target's motions, cannot be predicted in typical ISAR scenarios. Given some target's motions the only degree of freedom with which to force the orientation of the IPP to be a desired one is the radar position. The criteria for defining the optimal ISAR sensor position are introduced. The theoretical basis for optimising the sensor position, given a desired IPP, is also provided. The effectiveness of the proposed method is then tested on real data.