| Being more versatile compared to monostatic system, bistatic SAR(BiSAR) systemincreases difficulties in imaging for transmitter and receiver with different velocities andflight path are in separate flying platforms. Research on BiSAR imaging algorithm is ofgreat significance and value.Based on the analysis of spatial geometry and signal model of BiSAR, this thesisstudies on properties of bistatic slant range and two-dimensional resolution cell.Analysis shows that, echo signal of BiSAR is spatially variable at the azimuth.Differences in velocity vectors of transmitting and receiving platform lead to divergentimaging geometry of point target at distinct azimuth. In addition, the two-dimensionalresolution of BiSAR is also spatially variable.This thesis studies a new approximate echo signal model of BiSAR in frequencydomain. Using principle of orthogonal polynomial approximation, this modelapproximates a fourth-order fitting of bistatic slant range history and obtaines precisesolution of the bistatic stationary phase point by solving equation. A new point targetmodel in frequency domainis is established and compared with several current existingmodels. Results of the comparison show that, the obtained model is of accuracy andwide applicability, as well as simple.Two processing methods of BiSAR data imaging algorithm are demonstrated onthe basis of the two-dimensional bistatic spectrum. Stolt transformation is analysed andrange-frequency mapping function is derived. The scope of the bistatic approximationω-k algorithm and linear processing is dicussed through the linearization of thetwo-dimensional spectrum and angle approximation. To enhance the applicability of thealgorithm to the scene and flight mode, an improved algorithm is presented. Simulationresults show that, little difference in focusing performances lie between both algorithmswith medium bistatic structure, while improved algorithm is superior in larger scene andextreme bistatic structure. |
PDF Full Text Request