Research On Stepped Frequency SAR/ISAR Imaging Algorithm

In recent decades, with the development of hardware condition and digital signal processing technique, synthetic aperture radar(SAR) and inverse synthetic aperture radar(ISAR) have been rapidly developed and widely applied, and high resolution has become an inevitable trend of imaging radar. The range resolution of SAR/ISAR is inversely proportional to the signal bandwidth, thus in the conventional signal system, high resolution requires a large bandwidth of the transmitted signal, which calls for an A/D device with a large instantaneous bandwidth and high sampling rate, and greatly increases the cost of hardware.Stepped-frequency waveform is a waveform widely used in high-resolution radar in recent years. It emits a set of pulse signal with different carrier frequencies, and obtains high range resolution by range synthesis while reducing the instantaneous bandwidth of the system. However, since the stepped-frequency signal doesn't meet the “stop-go-stop” assumption in one burst, the main problem in both SF-SAR and SF-ISAR is how to remove the range difference in one burst. Therefore, this thesis carries out research on the imaging problems of SF-SAR and SF-ISAR.As to the compensation of range profile in SF-SAR, this thesis establishes the geometry model and signal model of SF-SAR, and the two-dimensional and three-dimensional matrix form of SF signal is given. Then the expression of the range difference compensation factor can be deduced by the principle of stationary phase(POSP). Finally, the expression of high range resolution profile(HRRP) can be obtained.In terms of SF-SAR imaging algorithm, two different imaging processes are analyzed. One of the SF-SAR imaging processes usually synthesizes the HRRP before imaging compensation, and the other one is opposite. In the first imaging process, the expression of HRRP is used to deduce two improved RD algorithms which can be applied to the SF-SAR imaging, and the compensation error is analyzed. Then, aiming at the second imaging process, a SAR imaging algorithm based on compensation in two-dimensional frequency domain is proposed. This algorithm is simpler than the first imaging process, and it can simultaneously compensate the range difference and range migration in the two-dimensional frequency domain.Aiming at SF-ISAR imaging algorithm, firstly the signal model of SF-ISAR is established, and the motion compensation problem of SF-ISAR in one burst is described. Then an ISAR imaging method based on the decoupling of motion parameters and recursive estimation is proposed. In this method, the two-dimensional search problem can be turned into a one-dimensional search problem by the two-dimensional correlation in time domain and coupling equations. Afterwards, parameter recursion is applied to precisely estimate the motion parameter. In the whole imaging process, this method only needs a one-dimensional searching, which greatly reduces the amount of calculation. Simulations have verified the validity of this method.