A Procedure for the Estimation and Compensation of Phase Delays in Spotlight SAR Data Focusing
Zamparelli, Virginia1; Fornaro, Gianfranco2
1IREA-CNR/UNICAS, ITALY; 2IREA-CNR, ITALY

Stripmap and spotlight are two different operating modes in which a Synthetic Aperture Radar (SAR) system can image an area on the ground. During the spotlight mode data collection, the sensor steers its antenna beam to continuously illuminate the terrain patch being imaged. Instead, in the stripmap mode, antenna points along a fixed direction with respect to the flight platform path and the antenna footprint covers a strip on the imaged surface as the platform moves. [1]
The spotlight mode is a practical choice when the mission objective is to collect fine-resolution data from one or more localized areas. The stripmap mode is most efficient when used for coarse-resolution mapping of large regions. The two main differences between spotlight and stripmap are the following. First, spotlight mode allows efficient imaging of multiple smaller scenes whereas stripmap mode images a long strip of terrain. Second, spotlight mode offers finer azimuth resolution than that achievable in stripmap mode using the same physical antenna.[2],[3]
SAR azimuth resolution improves with the length of the synthetic aperture. In stripmap mode, antenna pointing is fixed relative to the flight line; azimuth antenna beamwidth limits the available synthetic aperture length because it determines the transit distance over which the sensor illuminates each scatterer. Spotlight mode avoids this limitation on synthetic aperture length by steering the antenna illumination pattern to illuminate the scene for a long period. Consequently, antenna beamwidth does not limit azimuth resolution in spotlight mode.
In literature has been proposed an intermediate mode between the stripmap and the pure spotlight (staring spotlight) modes, the so-called sliding spotlight mode, also known as hybrid mode. The main characteristic of this system is that the radar antenna beam is steered about a point under the ground in such a way to achieve a ''slide'' (with respect to the staring case) of the footprint on the ground. Accordingly it is possible to generate an image whose azimuth extension is greater than that achieved in the staring spotlight mode and with an azimuth resolution improved with respect to the stripmap case [4],[5]. Therefore the stripmap and the staring spotlight modes can be seen as two limiting cases of the hybrid mode: the latter allows trading off the azimuth resolution and the extension of the imaged area. [6]
For all three modes imaging, different geometries may be possible: the so called boresight imaging geometry or the squinted imaging geometry. The first one, which is known also as zero Doppler, has an average pointing of the antenna perpendicular to the flight direction. In the second one, which is known also as non-zero Doppler, the antenna has an average pointing with an offset, forward or behind, with respect to the boresight direction.[7] When the system is operating in the spotlight (staring or sliding) mode an electronic beam steering of the antenna is adopted: the antenna radiation pattern is changed each fixed number of transmitted pulses (burst). Either due to possible spurious delays in the radiation HW, or to the introduction of delays due to the propagation through the atmosphere, or to phase delays due to the orbit curvature, phase offsets can be present on the received data from burst to burst.
In this paper we suggest a processing strategy aimed at estimating and eliminating such phase shifts. The procedure is based on the burst to burst focusing of produce different looks of the scene corresponding to each burst: phase offsets between bursts are then estimated by maximizing the image contrast during the look combination. The image contrast function represents the normalised effective power of the image intensity and gives a measure of the image focusing. In fact, when the image is focused correctly, it is composed of several pronounced peaks (one for each scatterer), which enhance the contrast. Whereas, when the image is defocused, the image intensity levels are concentrated around the mean value and the contrast is low. [8]
From this parameter we achieve the estimation of phase offsets between the looks, The estimated differential phase offsets are then integrated and phase subtracted from each burst prior to the final image focusing.
The algorithm has been successfully tested on a COSMO SKYMED image provided within the framework of the COSMO-SKYMED AO Project 2103.
In fig. 1 it is shown a zoom over an industrial area in the north of Matera (South Italy) before the correction of the phase shift. Azimuth is horizontal and on the corner reflectors, especially the brithest on at the center of the image a high level of sidelobes can be detected. In Fig. 2 it is shown the image of the same area after the estimation and correction of phase shift by using the proposed image contrast based technique: it is evident that a significant reduction of the level of sidelobes is achieved.



Figure 1 - Result of the focusing of real spotlight SAR data acquired by the COSMO/SKYMED constellation acquired over the city of Matera (South Italy), before the correction of phase shifts.



Figure 2 - As figure 1, but after the compensation of the phase shifts

[1] G. Franceschetti, and R. Lanari, Synthetic Aperture Radar Processing, CRC PRESS, New York, 1999.
[2] W. G. Carrara, R. S. Goodman, R. M. Majewski, Spotlight Synthetic Aperture Radar - Signal Processing Algorithms, Artech House, Norwood, MA, 1995.
[3] R. Lanari, M. Tesauro, E. Sansosti and G. Fornaro, ''Spotlight SAR Data Focusing Based on a Two-Step Processing Approach'', IEEE Trans. Geosci. Remote Sens., vol.39, no. 9, pp. 1993- 2004, Sep. 2001.
[4] D.P. Belcher, and C.J. Backer, ''High resolution processing of hybrid stnpmap/spotlight mode SAR'', IEE Proc., Radar Sonar Navig., 1996.
[5] R. Lanari, S. Zoffoli, E. Sansosti, G. Fornaro and F. Serafino, ''New approach for Hybrid stripmap/spot light SAR data focusing'', IEE Proceedings on Radar, Sonar and Navigation, Vol. 148, No. 6, pp. 363-372, December 2001.
[6] V. Zamparelli, G. Fornaro, R. Lanari, S. Perna, D. Reale, ''Processing of Sliding Spotlight SAR data in presence of squint'', in Proc. IGARSS, Munich, Germany, Jul. 2012.
[7] G. Fornaro, E. Sansosti, R. Lanari, M. Tesauro, ''Role of processing geometry in SAR raw data focusing'', IEEE Trans. Aerosp. Electron. Syst., vol. 38, 2002.
[8] M. Martorella, F. Berizzi and B. Haywood, ''Contrast maximisation based technique for 2-D ISAR autofocusing'', IEE Proc.-Radar Sonar Navig., Vol. 152, No. 4, August 2005