Research On Environment Simulation And Moving Target Indication Approach For Space-Based Radar

Spaced-Based Radar (SBR) is a new generation of radar system for various military and civilian applications. Military applications include wide-area surveillance, threat defense, and disarmament functions. Civilian ones include remote sensing, air-traffic control, space exploration, and law enforcement. Two key advantages of SBR are a greatly increased field-of-view, and global coverage without the political, strategic and geographic issues associated with land-based and airborne radars. This paper mainly discussed the issues of environment simulation and signal processing for the Moving Target Indicator (MTI) function of the space-based surveillance radar.This paper first discussed the environmental factors of SBR, including the RCS criterion, noise, attenuation, ionosphere, land and weather clutter, and jamming signal. Based on the discussion, some systemic issues of SBR are then discussed, such as the Field of View, power-aperture product, satellite orbit and the costs. Finally, further analysis about the influence of the earth rotation and range fold-over is performed as well as the thermal mechanism distortion of the antenna.In order to evaluate the performance of the Space-Time Adaptive Process (STAP) algorithm for SBR, the high fidelity simulation is performed on the land-clutter scenes. Firstly, the instantaneous reflectivity scene of specific site is constructed according to corresponding Land Cover Data, Terrain Data and radar parameters. Secondly, the interaction between reflectivity scene and radar signal processing algorithms is achieved based on the effects of range fold-over, earth rotation, and array pattern. Finally, the effect of Internal Clutter Motion (ICM) is proposed according to Billingsley's spectrum model. In order to simulate the clutter received by radar which is linear frequency-modulated (LFM) in each pulse and agile from pulse to pulse (LFM-FA radar), the power spectrum of the clutter of Pulse-Doppler (PD) radar is firstly calculated by means of the Range-Doppler Grid Procedure. Then the influence of the linear modulated signal to the power spectrum and amplitude characteristics is analyzed in view of sub-pulses'accumulation. And then the declining of the correlation among pulses caused by frequency agility and the decrease of the time random sequence variance is explained. Based on the previous research, the following conclusions are achieved: the shape of the power spectrum about the LFM-FA radar clutter is generally similar to that of PD radar's; the change of their power ratio is relevant to the amplitude changes; the clutter random sequence of SBR is Rayleigh distributed and its variety results from the variety of the radar parameters.This paper applies the signal processing system of the three-channel DPCA/ monopulse in order to eliminate the clutter of SBR, estimate the angle under high resolutions and to perform wide-area surveillance. Because of the radar motion and the earth rotation, a Doppler frequency component is generated on the clutter of each point on the earth--range walking, which is discussed in this paper according to the SBR-earth geometrical relations. It is shown that range walking together with range fold-over significantly degrades the clutter suppression performance of the DPCA algorithms. Effects of range walking are eliminated through appropriate inverse compensation during coherent processing and linear interpolations in non-coherent processing. And the effects of rang fold-over is amended with waveform diversity.The knowledge-aided signal processing structure, which focuses on the real-time, compatibleness and extensibility of the system, is explored in this paper. Its function contains knowledge storing, accessing and integrating into the signal processing chain and radar system. The clutter covariance matrix model is obtained according to the space-time property of the clutter. This model includes two portions. One portion represents the effect of the clutter heterogeneity and the other represents the influence of the slight errors, such as ICM and mismatch of array cells.According to the clutter covariance matrix model, two classes of knowledge-aided signal processing method of SBR are proposed. The first one, based on the Expectation-Maximization (EM) technique, proposes the clutter reflect characteristic map, employing the previous multiple dimensional CPI data, the experience and the mini-squared error rule. In order to enhance the robustness of the algorithm, Expectation-Maximization (EM) algorithm is applied to estimate the correction matrix according to the slight or unknown errors (such as Internal Clutter Motion). The second method is based on colored-loading technique. Scatter points are firstly located according to the same CPI data cube and previous knowledge. The clutter returning strengths are then estimated by maximum likelihood estimator. The clutter reflectivity characteristics map is then achieved by normalizing the strengths and averaging the results from several CPIs. The clutter covariance correction matrix of the current CPI is calculated by the map and previous knowledge. Finally, the correction matrix is loaded to clutter covariance estimated matrix of the current CPI.