无源毫米波实时成像理论与关键技术-ieeegrss.pptVIP

The purpose of our work is to find a nonlinear two-dimensional RCM compensation method for SA-FBSAR in frequency domain. The theme of our work is as following: 1. Set up SA-FBSAR response spectrum model; 2.Deduce nonlinear RCM analytic formula; 3.Propose SA-FBSAR nonlinear RCM compensation method First, we set up the geometry model for SA-FBSAR. It is shown in Fig.1. the indexes ‘S’ and ‘P’ denote spaceborne transmitter and airborne receiver, respectively. Both platforms move along the same direction. The imaging scene is in xoy plane. At slow time t=0 , the location of transmitter is(xs0, ys0, zs0) , and that of receiver is(0,yP0,zP0) ; and other notations to be used in this paper are listed . The SA-FBSAR two-dimensional frequency response after range compression is shown in formula (1). It is achieved by Air-Phase method. In formula (1), f and fd represent range frequency and doppler frequency. tb is air-phase time piont, it varies with target’s two-dimensional position and two-dimensional frequency. The space-variant spectrum of the system response is achieved by multiplying the reference target’s spectrum, and it is listed in formula (2). (Just according to the slides in this page) According to the signal spectrum model shown in formula (1)-(3), we design a frequency domain imaging algorithm for SA-FBSAR. Its flowing chart is shown in Fig.2. In the algorithm, nonlinear two-dimensional RCM play an important role. It includes two steps: the first step is the azimuth-dependent RCM(AD-RCM) compensation, it is realized by NUFFT operation with the non-uniform frequency fNU; the second step is nonlinear range-dependent RCM(RD-RCM) compensation, it is realized by iterporlation and phase multiplication operations. The simulation parameters are listed in this table. Signal from a scene with 15 point targets is simulated. Their azimuth extension is 0.2km and range extension is 1.0km. Their location relationships are shown in (a). Imaging results without RCM compensat