Multicarrier SAR Image Reconstruction Using Integrated MUSIC-LSE Algorithm

In synthetic aperture radar (SAR) applications, high-resolution images and effective estimation processes are vital for the reconstruction of any targets. This can be achieved by using multicarrier waveforms such as orthogonal frequency division multiplexing (OFDM) with the help of appropriate signal processing algorithms. However, the quality of the reconstructed image degrades in low signal-to-noise ratio (SNR) environments during SAR data acquisition. In this paper, an integrated multiple signal classification (MUSIC) assisted least square estimation (LSE) algorithm (MUSIC-LSE) is proposed to enhance the quality of the reconstructed SAR image in a low-SNR environment. Simulation results measured and evaluated the quality of the reconstructed image using three performance indicators of root-mean-square-error, main lobe width and cumulative side lobe levels. These indicators are also used to investigate the effect of OFDM subcarrier selection on the reconstructed image for a different number of subcarriers. Experimental validation of the approach is carried out using two steel pipes to image and detect the curvature of the steel pipes. The results show that the proposed MUSIC-LSE approach produces better-reconstructed images compared with the existing linear frequency modulated (LFM) chirp and OFDM-LSE approaches in low-SNR (-10 dB) environments and enables the radar to distinguish and detect the curvature of the pipes even below the radar range and cross-range resolution.