Study On Array Calibration And Direct Wave Suppression For High Frequency Hybrid Sky-Surface Wave Radar

High frequency hybrid sky-surface wave radar utilizes ionosphere reflection and surface diffraction propagation modes to detect targets beyond the horizon such as vessels on the sea and low altitude flying targets. Compare with traditional HF sky wave and surface wave radar, HF hybrid sky-surface wave radar is more flexible to build shore-based and shipborne receiver. The receiver of the HF hybrid sky- surface wave radar that can be classified as bistatic radar system receives radar echoes in silence, which improves invisibility and survival of the radar in wartime. What's more, it is more convenient to build TR-RN multistatic HF radar net in purpose of sharing sky wave radar and surface wave radar resources. As a result, it is becoming a hot topic in HF OTHR field. Since HF hybrid sky-surface wave radar is a new radar system, there are considerable difficulties in signal processing. This thesis focuses on array calibration and direct wave suppression in HF hybrid sky-surface wave radar system.This thesis mainly studies shored-based radar system. First it analyses operation mode of HF hybrid sky-surface wave radar, and illustrates the signal processing of this radar system. In order to suppress strong interferences, clutters, and noise, HF hybrid sky-surface wave radar exploits range-azimuth-doppler processing method, illustrating its process in detail. This thesis analyses the characteristics of direct wave in the radar system. Direct wave is the high frequency radio waves reflected by the ionosphere directly to the receiver, not diffracting along ocean surface. According to analysis of experimental data, direct wave has a strong energy, stable azimuth and contains information of transmission station, zero Doppler frequency, and multipath propagation characteristics. Therefore, direct wave is the ideal reference signal, calibration source in the signal processing, but strong interference at the same time, which needs to be cancelled. Thesis also introduces some key technologies, such as transmitter and receiver systems synchronizing, array calibration, direct wave suppression, ionosphere phase distortion, and target positioning.In order to achieve good array response, HF radar array calibration is studied. This thesis proposes a novel array calibration method based on spatial correlation matrix (SCM) for HF hybrid sky-surface wave radar. According to characteristics of HF hybrid sky-surface wave radar, direct wave is chosen to be calibration source. It is utilized to construct spatial correlation matrix, which contains array errors. Then, phase errors could be estimated by phase matrix of SCM, while gain errors could be estimated from the main diagonal elements of SCM. SCM could be estimated in temporal domain and frequency domain, and performances of this proposed algorithm are studied under different conditions such as signal-to-noise ratios (SNRs) and snapshots. The results show that SCM method still has a good performance in case of large phase errors. That means it is suitable for large array errors. In low SNR case, frequency method has a better performance.Next, this thesis discusses robust adaptive beamforming. Thesis analyses conventional beamforming (CBF), Capon beamforming, and their problems in practical. Then, the robust adaptive beamforming is introduced. This thesis mainly studies diagonal loading (LSMI), robust minimum variance beamforming (RMVB), and robust Capon beamforming (RCB), analyzing the principles of improving robustness of the beamformers in detail. Thesis discusses about choosing parameters, array beam patterns, and beamforming output signal-to-interference-plus-noise ratios (SINRs) of the three methods. Compared these three algorithms, LSMI is simple, but the disadvantage is that the diagonal load cannot be determined exactly. Further, the signal power estimated by LSMI is lower than the truth. RMVB has the strongest ability to suppress interferences. But it has the deepest nulls for the interferences at a cost of worse noise gain, resulting in high sidelobes. The estimated signal power in RMVB method is much closer to the truth. RCB has the highest SINR compared to other two methods at the same situation. It could suppress interferences and noise at the same time. What's more, the signal power estimated by RCB method is the most accurate.Finally, a system of HF hybrid sky-surface wave radar is simulated, considering the array errors, which are calibrated by SCM method. Since orthogonal projection method is sensitive to array error, robust adaptive beamforming methods are applied to suppress direct wave. In order to evaluate the performance of the direct wave suppression, the improvement factor is defined to be the ratio of adaptive bemforming output SINR and CBF output SINR. Thesis analyzes the conditions that could affect the performances of direct wave suppression. When the array has errors and the signal is nonstationary simultaneously, the performances of direct wave suppression are affected seriously. Simulation and experimental data results show that three methods all can suppress direct wave partly, and cannot cancel direct wave completely. And target detection can be improved.