A spaceborne synthetic aperture radar (SAR) is a sensor that is mounted on a satellite and utilizes radio waves to enable high-resolution and all-weather Earth observation. However, conventional spaceborne SAR has some limitations in use because it has a trade-off relationship within the main system parameters. In addition, it obtains information from the Earth’s surface by transmitting and receiving only a single polarization.

In Earth observation using spaceborne SAR, an efficient observation requires a short revisit time by observing wide areas [1]. For this purpose, wide-swath SAR operation modes, such as ScanSAR and TOPSAR (terrain observation with progressive scans SAR), were introduced and studied [2, 3]. However, these two modes observe the sub-swaths sequentially and this operational method worsens the resolution in the azimuthal direction of the SAR system. Therefore, obtaining a wide-swath and fine resolution at the same time is difficult for a conventional wideswath SAR system. To overcome this constraint, many operation modes are currently being studied as high-resolution wideswath (HRWS) systems. Among these systems, we are interested in HRWS systems using a reflector antenna [4–6]. Generally, reflector antennas have high gain and moderate bandwidth and can be made light-weight by using composite material or with a mesh-type design. In addition, digital beam forming (DBF) can be performed by on/off operation of the feed elements. One of the operation modes analyzed as a reflector-based HRWS system is SweepSAR [4], which is a SAR operation mode of obtaining a high-resolution image using a broad beam to transmit and a narrow beam to receive based on the SCORE(scan-on-receive) method using DBF. However, this operation mode requires the use of a low pulse repetition frequency (PRF) to obtain an unambiguous wide swath. For this reason, a modified mode of SweepSAR using dual channels, which is called M-SweepSAR, was examined for a single polarization (single-pol) SAR system [6].

Generally, the SAR system employs only a single transmitting/receiving polarization and obtains one type of polarization information. In contrast, in polarimetric SAR, such as dual-polarimetric and quadrature-polarimetric (quad-pol) approaches, various information can be obtained using multiple transmitting and receiving polarizations, and it has been utilized this way in many fields [7, 8]. However, for a SAR system operating in a quad-pol mode, interleaving transmitting pulses of different polarizations impacts the design of the SAR system. Therefore, accurately analyzing this effect, as done in [9, 10], is necessary to design an appropriate quad-pol SAR system.

In the current paper, the reflector antenna and the performance of the proposed quad-pol system operating in M-SweepSAR mode are designed and analyzed, respectively. We also compared wide-swath quad-pol SAR systems in terms of system performance and antenna design.

Generally, a SAR system can be characterized by performance indicators, such as the ambiguity-to-signal ratio and resolution. These indicators are strongly related to the system parameters, antenna patterns, and SAR operation modes.

The ambiguity-to-signal ratio, which comprises the range ambiguity-to-signal ratio (RASR) and azimuth ambiguity-to-signal ratio (AASR), is the ratio of undesired signals to the desired SAR signal. it is calculated through the transmiting/receiving antenna pattern, G_t(θ)/G_r(θ); backscattering coefficient, σ⁰; slant range, ρ; range ambiguous slant range, ρ_RA; incidence angle, θ_I; Doppler processing bandwidth, PB; and the angles at which the ambiguity signal occurs in the range and azimuth directions, θ_A and θ_AZ, respectively [9, 11]. The resolution, which is the degree of distinguishing the main target from other objects, in the azimuthal direction (ρ_a), is calculated through the beam footprint velocity (V_g), Doppler bandwidth (Δf_dop) and wavelength (λ) [12, 13]. The range resolution is related to the bandwidth of the transmitted pulse and is not directly associated with other performance indicators.

We dealt with the wide-swath and quad-pol operation modes and designed the antenna, which is one of the most important parts of the SAR system, to conduct a proper performance analysis.

In this research, we selected a spaceborne SAR system with a C-band center frequency operating at an altitude of 505 km with a total swath width of 150 km for a single-pol mode at an incidence angle of 20°–35° [6]. In this case, we assumed that the SAR system performance, which involves an ambiguity ratio in each direction and the resolution should be under the conditions of −20 dB and 10 m, respectively.

The required SAR antenna beamwidth was determined by the swath width and the number of sub-swaths, which help in selecting the optimum reflector antenna depending on the operation mode. This paper suggests a reflector length for the offset reflector antennas of 4 m, 13.5 m, and 6 m for ScanSAR, SweepSAR, and M-SweepSAR, respectively. These antennas had a reflector with F/D_p= 0.409 and a feed array linearly placed in the range direction with one row. Only M-Sweep-SAR had a feed array with two rows. The radiation patterns of the reflector antenna by several elements at one row of the feed array for the M-SweepSAR system are presented in Fig. 2. These patterns were simulated using the FEKO EM simulation tool. The transmitting pattern in the elevation direction was broad enough to cover the overall swath using additional feed antennas at both ends of each row of the feed array. The receiving patterns were tilted in the elevation direction, so that each feed received echoes from each sub-swath.

The reflector antenna of each SAR system had a feed antenna according to the number of sub-swaths in the single-pol mode. However, SAR systems use fewer feed antennas in the quad-pol mode to meet the requirements (except for the M-SweepSAR system). In quad-pol M-SweepSAR, the PRF and signal acquisition areas of each channel were determined based on the timing diagram shown in Fig. 3. M-SweepSAR had 14 sub-swaths in the quad-pol mode with PRFs of 5,250 Hz and 5,400 Hz in each row of the feed array, which maintained an PB of 0.16° and had 14 sub-swaths in a single-pol mode. As shown in Fig. 3, each row of the feed array transmits the signal to sub-swath (sw) 1–14, with one row receiving echoes from sw 5–7 and 11–12, and the other row receiving echoes from sw 1–4, 8–10, and 13–14. ScanSAR had three sub-swaths in the quadpol mode, with three PRFs of 10.15, 10, and 9 kHz and a PB of 0.44°, and 6 sub-swaths in the single-pol mode. SweepSAR had 13 sub-swaths in the quad-pol mode, with a PRF of 3,400 Hz and PB of 0.26°, and 24 sub-swaths in the single-pol mode. All of the information for the three wide-swath operation modes working in the quad-pol mode is shown in Table 1.

Based on the simulation results of the three quad-pol SAR systems for wide-swath, the SAR performances, including reflector antenna size, ambiguity ratio, resolution, and swath width, were analyzed and compared with the use of the backscattering coefficients assumption, which was suggested by Ulaby and Dobson [16]. In our previous work [18], we had conducted a comparison between our calculation results and data in [9] to verify our analysis procedure on quad-pol SAR systems. Generally, the RASR for the co-pol (HH and VV) was about 10 dB lower than that for the cross-pol (HV and VH). Therefore, we present only the RASR for HV in Fig. 4.

From Table 1 and Fig. 4, the reflector length for SweepSAR was the largest in order to meet the required performance in the single-pol and quad-pol modes while maintaining the required swath width obtained in the single-pol mode because of the characteristics of the SweepSAR operation mode. Conversely, the reflector lengths of M-SweepSAR and ScanSAR were half that of SweepSAR and the smallest, respectively. Quad-pol SweepSAR had the largest reflector antenna but was best in terms of total ambiguity ratio, which means the sum of RASR and AASR. Quad-pol ScanSAR and quad-pol M-SweepSAR had similar levels in ambiguity within the incidence angle range of 20.0°–27.5°. However, it can be conjectured from Fig. 4 that quad-pol M-SweepSAR had better characteristics than quadpol ScanSAR and satisfied the requirements on the overall incidence angle range.

The azimuth resolution depends on the Doppler bandwidth, and in the case of ScanSAR, it is also affected by the number of sub-swaths. As shown by the simulation results, quad-pol ScanSAR had the largest PB with the smallest antenna size, but its resolution was the worst. In addition, quad-pol Sweep-SAR had the best resolution among the three operation modes, even though it had the largest antenna size because quad-pol M-SweepSAR had a narrower PB for the azimuth ambiguity performance.

In terms of the swath width under the conditions of less than −20 dB RASR and AASR, quad-pol ScanSAR was limited in obtaining a wide swath width because of the ambiguity performance degradation over a large incidence angle range. As the PRF increased, and the receiving interval was reduced for the quad-pol mode, quad-pol SweepSAR was also limited in obtaining a continuous total swath width because of the blind range. Therefore, the two modes had half of the total swath width in comparison with the single-pol mode. On the contrary, quad-pol M-SweepSAR could obtain a swath width almost like that of the single-pol mode within the required performance levels. In other words, M-SweepSAR used two PRFs, and this characteristic was advantageous in obtaining a wider swath width than SweepSAR. In addition, M-SweepSAR had a better ambiguity performance than ScanSAR; it was also advantageous in obtaining a wider swath width than ScanSAR.

Most of the differences in the performance indicators discussed in this chapter arose from the differences in antenna patterns, which were determined on the basis of the required performance. Therefore, if the requirements change, the antenna size, which determines the achievable range of the SAR system performance, should be changed. As a result of confirming the performance changes derived from the size of the reflector antenna in each operation mode, increasing the reflector size for the ScanSAR improves the ambiguity performance but worsens the resolution [20]. Conversely, SweepSAR and M-SweepSAR have the advantage of maintaining resolution in comparison with ScanSAR even when the reflector size increases, and M-SweepSAR can reduce the reflector size in comparison with SweepSAR, as mentioned in Chapter II. In other words, for the ScanSAR system to obtain a wider swath width in the quad-pol, a larger reflector must be required to improve the ambiguity ratios, which does not satisfy the resolution requirement. However, we can design a reflector antenna for the M-SweepSAR and achieve a wider swath width in the quad-pol, although it uses a larger reflector than ScanSAR.

In this paper, the performance analysis of the proposed quadpol M-SweepSAR system was conducted. To determine the validity of our analysis, we designed, analyzed, and compared the ScanSAR, SweepSAR, and M-SweepSAR systems. The result of the analysis and comparison of the three operation modes based on the properly designed systems indicated that M-SweepSAR is expected to be highly recommended as the HRWS spaceborne SAR system because it is advantageous in satisfying the required ambiguity ratio and resolution in both the single-pol and the quad-pol modes with an appropriate reflector antenna size and obtains a wider swath width in the quad-pol mode than the conventional wide-swath operation mode. However, as the feed array has two rows, the system must have dual channels and operate two center frequencies in the available bandwidth. Therefore, filtering for interference suppression should be considered for practical implementation.