A robust detection method for inverse synthetic aperture radar (ISAR) target using high resolution range profiles (HRRP) is proposed. The proposed detector
improved robust generalized likelihood rate test (IRGLRT)
is derived by estimating signal from every range cell through an improved second order cone programming (SOCP). Furthermore
lower computational complexity is analyzed
constant false alarm rate (CFAR) property is proved and the detection threshold is estimated. Finally
experimental results for measured data of three planes illustrate that IRGLRT achieves a visible performance improvement comparing with conventional generalized likelihood rate test (GLRT)
especially in the presence of severe steering vector mismatches.