IRCI-Free MIMO SAR
This thesis presents new configurations that (i) utilise the available bandwidth to the maximum efficiency in multiple subband multiple-input multiple-output (MIMO) synthetic aperture radar (SAR) and (ii) employ all of the phase centres in orthogonal waveform encoding MIMO SAR. These configurations enable us to image a wider swath with a higher cross-range resolution compared to the conventional orthogonal waveform encoding MIMO SAR. Two different multiple subband MIMO SAR configurations are proposed. The first one makes use of multiple contiguous narrow receiving beams with different phase centres which permits the use of a pulse repetition frequency (PRF) lower than the total Doppler bandwidth. Echoes corresponding to different transmitted subband waveforms are processed jointly without separating them at the receiver using a bank of bandpass filers (BPFs) to utilise the bandwidth to the maximum efficiency (i.e. there is no need to add guard bands between the adjacent subband spectra). Digital beamforming (DBF) on receive in elevation is proposed to mitigate the effect of interbeams overlapping on the azimuth ambiguity characteristics. The second proposed multiple subband MIMO SAR configuration has an advantage over the first in that the beamwidths of all transmitters and receivers are the same. The beams simultaneously illuminate the same imaging area which overcome the receiving interbeams overlapping problem without employing DBF on receive in elevation. This reduces the implementation complexity. The proposed orthogonal waveform encoding MIMO SAR configuration employs multiple contiguous azimuth beams. It uses all of the phase centres including the spatially overlapping ones to reduce the minimum operating PRF that should be satisfied to avoid aliasing in the azimuth dimension. The received signals in all proposed configurations are processed as the solution to system identification problems using the principle of displaced phase centres (DPC). This, in turn, facilitates the use of linear frequency modulated (LFM) waveforms for transmission and, hence, gains all the inherent benefits of these waveforms. The impulse response in the range dimension is identified using a proposed frequency domain system identification (FDSI) estimation algorithm instead of a matched filter. The length of the transmitted waveform is not a function of the channel impulse response length in the range dimension which makes the proposed algorithm suitable for a stripmap SAR application. The estimated range profile obtained using the FDSI-based algorithm has ideally a zero sidelobes level which is the property interrange cell interference (IRCI) free.