Calibration is an important work to ensure the measuring accuracy of radar. All the methods of calibration in China are to compare the measure data of radar with the more accuracy position data of space target. It is a time-consuming work with a large workload of coordination problems. A self-calibration method is proposed in this paper, in which low orbit satellite is tracked by ground-based radar. The design scheme of self-calibration, which evaluates biases of radar measurements, is given with the data processing and the key algorithms. The correctness and effectiveness of the proposed method is verified by simulation verification, which uses noise-free measuring data. The interaction and coupling effects between the random errors of the measurements and the estimated orbit parameters are found through noise simulation verification, which has accumulated experience for subsequent engineering application and technology improvement.