A new ISAR range alignment method via minimum tsallis entropy

摘要

Motion compensation is very important for Inverse synthetic aperture radar (ISAR) imaging while the imaging quality is directly dependent on its performance. Range alignment (envelope alignment) is the key procedure of motion compensation. The nonextensive entropy, known as Tsallis entropy, is introduced into ISAR range alignment because of the correlation between adjacent one-dimensional range profile (or long time memory characteristic). Firstly, the principle of range alignment method via minimum Shannon entropy (RAMSE) was summarized. Based on this, a new adjacent range profile alignment method was given via minimum Tsallis entropy (RAMTE). Further, in order to reduce the accumulation of alignment error and jump error, the modified RAMTE (mRAMTE) was presented using accumulation range profile alignment and smooth filter. Based on the criterion of minimum Tsallis entropy, mRAMTE could reduce the alignment error into 1/8 range resolution cell by data interpolation. Simulation results and experiments show that mRAMTE can effectively compensate range walk caused by target motion. And it was also shown that the alignment performance of mRAMTE is better than RAMSE.