A synthetic aperture radar (SAR) system is capable of acquiring images in all kinds of weather by overcoming the shortcomings of optical technology that are affected by changes in climate. As a result, it is widely employed as a representative reconnaissance and surveillance mission system. SAR systems can observe a large area, capture high-resolution images depending on the operation mode, and can also be used on various platforms, such as satellites, aircraft, and drones. Consequently, SAR system design has been actively researched in recent times [1].

Among the SAR systems that are currently being actively studied, satellite SAR systems are more expensive to operate than drone and aircraft SAR systems. In view of this expense, it is necessary to analyze the required image quality for a given scenario before a design is implemented so as to determine whether the system is suitably prepared to meet mission performance requirements.

A SAR system should be designed by accounting for the correlation among its performance variables in specific scenarios. Typically, trade-off relationships between the incidence angle, antenna size, and pulse repetition frequency (PRF) have been observed [2]. For instance, the selection of a low PRF tends to improve the range ambiguity-to-signal ratio (RASR) but reduces the azimuth ambiguity-to-signal ratio (AASR). Meanwhile, the selection of a high PRF results in considerable improvements in AASR but reduces RASR. Therefore, considering these trade-off relationships, the performance analysis of a SAR system must factor in the effects of its design variables.

Multiple design studies on monostatic SAR systems that use only one satellite have been conducted in the past. Although the use of a single satellite has allowed for some reconnaissance and surveillance missions to be carried out, such a setup cannot cope with geometric problems in a flexible manner [3], and there are limitations to its performance. As a result, research on designing bistatic/multistatic SAR systems that can operate two or more satellites has recently gained momentum [4]. Bistatic SAR systems have been proven to not only achieve better performance than monostatic SAR systems, but also exhibit stronger resistance to jamming attacks and can flexibly cope with terrain conditions [5, 6].

Considering its many advantages over monostatic SAR systems, an in-depth analysis of the system design for a bistatic SAR system is necessary. In particular, since the transmitter and receiver are the same in a monostatic SAR system, there are fewer variables to consider when implementing platform fluctuation and signal processing algorithms. In contrast, the transmitter and receiver in a bistatic/multistatic SAR system are separate, and they are designed by considering various conditions, such as time/frequency/phase synchronization [7]. This implies that the effects of this separation of the transmitter and receiver must be accounted for when conducting image quality analysis. Therefore, in this paper, a performance analysis is conducted to account for the variable changes caused by the bistatic geometry in the design of a bistatic staring spotlight mode SAR system in which the receiver is vertically separated from the satellite movement but is located at the same altitude.

This paper is structured as follows. In Section II, a brief description of the theory behind the performance variables involved in the monostatic staring spotlight mode is provided [8]. Subsequently, a simulation is conducted based on the theory, and the results are confirmed. In Section III, the performance variables of the bistatic staring spotlight mode are examined based on the bistatic geometry and simulation results. Next, a simulation of the performance analysis is conducted using the performance variables of the bistatic staring spotlight mode. Notably, the bistatic SAR system scenario used in the simulation was constructed based on TerraSAR-X’s monostatic SAR system scenario. Finally, in Section IV, a performance evaluation of the bistatic staring spotlight mode is conducted, and the results are analyzed.

Fig. 1 displays the geometric difference between the operation of the staring spotlight mode, which is investigated in this study, and the stripmap and sliding spotlight modes, which are commonly employed in SAR systems. Notably, the stripmap mode works by setting the beam rotation center at an infinite distance. In contrast, the staring spotlight mode uses azimuth beam steering to match the beam rotation center with the target, thereby obtaining higher resolution images than the other SAR operating modes [9]. Owing to these advantages, the German Aerospace Center (Deutsches Zentrum für Luft- und Raumfahrt [DLR]) has been conducting research using the staring spotlight mode with TerraSAR-X [10, 11].

The staring spotlight mode functions by continuously observing the area of interest, which enables it to obtain higher resolution images than the other modes. To put it simply, its main difference from the stripmap mode is the use of electronic beam steering to capture high-resolution images. However, since electronic beam steering leads to variations in antenna patterns based on the steering angle, an analysis of azimuth performance variables must account for these effects. Hence, in the following sections, the performance analysis conducted for the staring spotlight mode is briefly described while also taking these effects into consideration.

Section II presented a description of the performance analysis conducted for the monostatic staring spotlight mode, following which simulations were performed and verified. Expanding on this, the performance of the bistatic staring spotlight mode is analyzed in this section. As shown in Figs. 10 and 11, the bistatic geometry adopted in this study comprises a receiver, which was positioned perpendicular to the flight direction of the satellite, at the same altitude in the vicinity of the target. Table 2 presents the geometric differences between the variables of the bistatic and monostatic SAR systems. The variables investigated for this performance analysis were the slant range between each sensor and the target, the antenna pattern, and the incidence angle [4, 19].

In Section II, a timing diagram was created to confirm the limitations pertaining to receiving timing before conducting performance analysis for the monostatic SAR system. In this section, the theory of a timing diagram for a bistatic SAR system is described. Bistatic SAR systems feature an extra limitation, in addition to the two constraints noted for the monostatic SAR system. This extra limitation is a direct signal, which is a limitation that must be addressed through the reception of a signal generated from the sidelobe of the transmit antenna, which travels by line of sight to the receiver.

Fig. 12 displays the timing diagram for a bistatic system, with the baseline set to 30 km in the TerraSAR-X scenario, featuring the parameters presented in Table 1. While the timing diagram of the monostatic SAR system comprises two types of lines, another set of lines is added in the bistatic SAR system. Although the bistatic SAR system offers many advantages, the PRF and incidence angle in the system must be determined in more detail.

The performance analysis was conducted after checking the timing diagram of the bistatic SAR system. Table 2 presents the differences between the performance variables—the slant range, incidence angle, and each antenna pattern difference—observed when applying bistatic geometry. First, in the case of the bistatic NESZ, all changes caused by the bistatic geometry were considered and analyzed using Eq. (8):

The azimuth resolution was calculated through approximation using Eq. (10) by accounting for the antenna illumination taper [4]. Since the slant range between the target and each transmitter and receiver was the same in the monostatic SAR system, it is expressed as ½ in Eq. (4). However, in the case of the bistatic SAR system, it is calculated by considering the slant range between the target and each transmitter and receiver, as expressed in Eq. (10). The range resolution was converted into ground range resolution using the sine function. Furthermore, since the transceiver and receiver were separated, the bistatic angle was converted into ground range resolution to account for each incidence angle, formulated using Eq. (11):

The AASR was also analyzed by accounting for the effects described above. First, the antenna pattern of each transceiver and receiver was considered, following which the ambiguity signal was determined by accounting for the antenna illumination taper when analyzing the azimuth resolution. Furthermore, as in the case of NESZ, RASR was analyzed by considering each antenna pattern and incidence angle. AASR and RASR can be expressed as Eqs. (12) and (13), respectively:

Simulations for the bistatic SAR system performance analysis were conducted using the above formula. A comparison of the results obtained for the proposed structure and those attained for the monostatic SAR system are depicted in Figs. 13–16.

In the case of NESZ, a slight improvement was observed for the bistatic SAR system (Fig. 13), resulting from a decrease in the slant range due to the receiver’s position being close to the target. The azimuth variable also improved for the same reason. An improvement in azimuth resolution owing to the antenna illumination taper is confirmed (Fig. 14). Drawing on the same reason, it is possible to confirm that AASR tends to improve due to the lowering of the sidelobe signal. However, it was also observed that it worsened according to the PRF because it was affected by the grating lobe (Fig. 15). As the receiver approached the target, the slant range decreased but the range difference of the ambiguity signal did not change. Consequently, the angle of receiving the ambiguity signal was relatively far away compared to the transmitter. Hence, the tendency of improvement in RASR in a bistatic SAR system is confirmed.

On the other hand, as depicted in Fig. 14(b), the ground range resolution deteriorated. As the total slant range from the target decreased, the incidence angle had to be lowered to observe the same target, resulting in deterioration in the ground range resolution. Through this performance analysis of the bistatic SAR system, the quality of the images it is capable of producing in specific scenarios was verified before moving on to modeling the system.

These simulations confirm that most performance variables tend to improve with an increase in the baseline (with the position of the receiver approaching the target), except for ground range resolution. Based on this finding, an ASR simulation was conducted to verify whether the same trend persists as the baseline increases. Fig. 17 displays the ASR results when the baseline is set to 10 km, 30 km, 50 km, 70 km, and 100 km.

The figures above show that the ASR improved with an increase in the baseline, consistent with our analysis results. However, the RASR in Fig. 17(a) shows a different tendency when the PRF is lower than 4,500 Hz. This observation can be attributed to variations in the angle at which the ambiguity signal is received, depending on the PRF. The ambiguity signal received in the given scenario was near the null of the antenna pattern. If the angle at which the ambiguity signal is received corresponds to the null of the antenna pattern, the RASR improves. In contrast, if it corresponds to the peak of the upper lip, the RASR worsens. As a result, the influence of the baseline was analyzed using the PRF band (4,500–5,000 Hz), since it is less affected by the antenna pattern.

In this study, a performance analysis of the bistatic staring spotlight mode was conducted. This study lays the theoretical foundations for designing a bistatic SAR system that involves sophisticated design variables. To ensure precision, theorization, simulation, and verification of the monostatic staring spotlight mode were first conducted, with regard to previous studies in the area. Subsequently, a simulation of the monostatic staring spotlight mode was carried out, and its results were compared to those attained by previous studies in the literature. After verifying the simulation results for the monostatic staring spotlight mode, performance analysis of the bistatic SAR system was conducted, especially accounting for its geometry. The performance variables were analyzed based on the bistatic geometry, and a performance analysis formula for the bistatic staring spotlight mode was devised. Using this formula, we analyzed a bistatic SAR system scenario considering a 30 km baseline. Notably, this scenario was created by expanding the TerraSAR- X scenario—a monostatic SAR system consisting of previously acquired images. The performance comparison results are presented in Figs. 13–16. The NESZ, ASR, and azimuth resolution all tended to improve, while the ground range resolution tended to deteriorate. This finding can be attributed to the incidence angle of the receiver being lower than that of the transceiver, since the receiver was positioned close to the target. Notably, before conducting the bistatic simulation, the results of the monostatic SAR system were confirmed to be consistent with the expected results, since the lower the incident angle, the similar the trend when the slant range is the same.

In this study, the performance of the bistatic staring spotlight mode was analyzed using a receiver at the same altitude and azimuth coordinates as the transceiver. The results of this study, in conjunction with those of previous research [20] conducted on the bistatic stripmap mode, offer basic knowledge and lay the foundation for further study on various bistatic scenarios in the future.