Research on RCS measurement of ship targets based on conventional radars
Publié en ligne: 23 déc. 2022
Pages: 1105 - 1116
Reçu: 21 juil. 2022
Accepté: 19 nov. 2022
DOI: https://doi.org/10.2478/amns.2022.1.00030
Mots clés
© 2022 Huatao Tang et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.
As a platform operating on the sea, the electromagnetic scattering characteristics of a ship target differ from those of other targets. Their electromagnetic scattering characteristics are related not only to those of the ship target but also to the characteristics of coupling between the ship target and the sea surface [1,2,3]. Therefore, we cannot simply measure the radar cross-section (RCS) of ship targets in the indoor compact range or outdoor ground open site. This is because these two measurement methods ignore the coupling characteristics of the sea surface, and therefore the measurement result error is large [4]. Measurements must be carried out in the open sea to get accurate results.
A number of countries have established an open measurement range to measure the RCS of water-surface ship targets and conducted RCS measurement and theoretical research [5]. For example, the U.S. Navy conducted a coupled electromagnetic scattering full polarisation measurement test of the simulated sea environment and sea-surface targets at the Naval Surface Warfare Center (NSWC) Carderock test site [6]. France performed an experimental study on the coupled electromagnetic scattering characteristics of ship targets and sea environments in different marine environments in the IMST laboratory [7]. Italy carried out RCS test research on a ship scale model and superstructure and components in a simulated sea environment at its ship RCS test site [8]. However, since the measurement conditions are limited, the above measurement targets are scaling models, not real large ships. The main purpose of the above measurement is to analyse the electromagnetic scattering characteristics of cooperative ship targets to guide the stealth design and stealth index evaluation of ships. There are few experiments to study the electromagnetic scattering characteristics of non-cooperative ship targets to establish a proper RCS measurement system and method of non-cooperative ship targets [9].
In addition, the above measurements are made by a special RCS measurement radar. However, conventional radars are still a main target detection sensor in developing radars at this stage. Providing a relatively limited amount of information compared with an RCS measurement radar, it can only rely on the detection results of conventional radars to analyse and calculate the target RCS when the ship is still running [10]. Thus, ship target RCS and target recognition are determined by calculating and analysing the sea surface RCS of the ship target based on conventional radars [11]. Practical experience shows that conventional radar can also measure RCS. In Li et al., [9] an RCS measurement system was designed based on an airborne radar. By studying the tracking algorithm, the method to track non-cooperative targets stably using the radar and optoelectronic equipment of the system was introduced. So the RCS of the non-cooperative targets can be measured accurately. Finally, the measured results were compared with the theoretical values. In Lu et al., [12] an RCS testing method based on airborne radar and measurement data analysis method were introduced. The measured results were compared with the simulation results. In Uchiyama et al., [13] RCS of road objects was measured with a 79 GHz ultra-wideband (UWB) automotive radar. In Lainer et al., [14] RCS of turbines in wind turbine sites was measured with an X-band radar.
This paper studies the impact of the coupling scattering characteristics of target and sea surface on the measurement results of target RCS. The calibration error is also analysed, and the error elimination method and the exact calculation formula of surface target RCS are given. Then, an RCS measurement system is designed based on conventional radars. First, the system measures the RCS of the Luneburg lens reflector on the lake. Combined with the deduced RCS calculation formula of a water surface target, the system performance is analysed, the radar radiation distribution is studied and the RCS measurement method of water surface targets is preliminarily proposed. The system then measures the large ship target on the sea, the RCS distribution of ships is analysed and the potential measurement uncertainty is studied.
When measuring the RCS of the target on the sea surface, there is a coupling effect between the target and the sea surface, which will alter the incident field at the target. The measurement range belongs to the GPR [15]. Under GPR conditions, there is a multipath radio wave propagation effect, [16] as shown in Figure 1.
Fig. 1
The multipath wave propagation effect.
The target to be measured is irradiated by two different incident fields, including the direct irradiation of the incident field of the transmitting antenna and irradiation of the incident field to the target after being reflected by the water surface. Thus, the vector superposition of the incident wave transmitted by multipath in the target area is formed. Similarly, the scattering field of the measured target reaches the radar receiving antenna in a similar transmission path. Consequently, if the higher-order water surface reflection is not considered, the target scattering echo consists of the echo components of the following four transmission paths: transmitting antenna – target – receiving antenna (direct path), transmitting antenna – water surface – target – receiving antenna (indirect path), transmitting antenna – target – water surface – receiving antenna, and transmitting antenna – water surface – target – water surface – receiving antenna. Therefore, given that the incident field is E0, the total scattering field can be expressed as follows:
The directivity function of an antenna must also be considered in calculations. Let the directivity function of the antenna be
The power gain factor under GPR condition can be expressed as follows:
According to the comparison method, [16] if the RCSs of the target and calibrator are
Under GPR conditions, the relationship between the target echo power
Similarly, the relationship between the calibrator echo power
Therefore, by introducing Eqs. (5) and (6) into Eq. (4), the relationship between the target RCS and the calibrator RCS under ground-level field conditions can be expressed as follows [16]:
Conventional radars are mechanical scanning radars. Given the large lateral dimension of Ship B, multiple beams will be scanned during its operation, and a series of pulse echo trains will be generated in the radar receiver. To calculate the target RCS using Eq. (7), the position of Ship B must be calculated to determine its echo power. The more commonly used methods to determine the position of Ship B include maximum signal, improved intermediate value and centroid [18, 19]. However, as shown in Figure 2, if the calculated position of Ship B is inconsistent with its real position, the calculated echo power is not equal to the real echo power of the target. Hence, the final calculated RCS will also have errors.
Fig. 2
The error in the target's position calculation.
The RCS measurement system is improved based on a conventional radar. The main improvements include the addition of transmission monitoring and linear receiving channels, RCS measurement, control software modules and data recording equipment.
In order not to affect the normal operation of the original radar, the transceiver unit of the radar is improved. The original transmitter and receiver are retained, and a linear coherent receiver is added to the transceiver unit to monitor certain parameters such as transmitted signal power, frequency and pulse width, as well as the linear reception, high-precision sampling, correction compensation and coherent processing of echo signal, to improve the accuracy of RCS measurement.
The original radar display and control console software is improved. The RCS measurement and control software module is added to confirm and track the measured target and the measurement operation control and transmit the target information and data recording control instructions to the data recording equipment.
Data recording equipment must be added to record the ship's navigation attitude information, radar working status, target echo data and target track data for analysing and solving the target RCS.
The measurement site is a lake. Clear and windless weather is selected for measurement to further eliminate the influence of clutter. Two Luneburg lens reflectors of different sizes are measured. The small Luneburg lens reflector is utilised as the calibration body, and the big Luneburg lens reflector is used as the target to be measured. The accuracy and feasibility of the RCS measurement system are verified by comparing the measurement results with the actual value. The RCS of the two Luneburg lens reflectors has already been calibrated in the microwave anechoic chamber, whose results are presented in Table 1. When measuring the RCS of the two Luneburg lens reflectors in the microwave anechoic chamber, adjust the frequency of the measuring radar to make it the same as the frequency of the conventional radar in this paper.
The parameters of the test targets.
| 1 | The small Luneburg lens reflector | 190-I | 5.6 |
| 2 | The big Luneburg lens reflector | 300-I | 41 |
RCS, radar cross-section.
At a distance of 4500 m from the radar, an open area on the water surface without foreign matter interference is selected, and the support rod with the fixing scheme is arranged, as shown in Figure 3. The height of the support rod above the water surface is 0.6 m, and the exposed part is wrapped with microwave absorbing materials (MAMs). The small Luneburg lens reflector is placed, and the working state of the radar and linear receiver is adjusted to ensure that the radar detection probability is > 80%.
Fig. 3
The installation scheme of the Luneburg lens reflector.
The linear received echo signal of the small Luneburg lens reflector is illustrated in Figure 4. The horizontal axis implies the distance, the starting point implies the starting point of the recording sector and the vertical axis implies the echo intensity. As shown, the maximum echo intensity of the small Luneburg lens reflector is close to 80 dB. Its peak is obvious, there is almost no interference around and the measurement result is ideal.
Fig. 4
The echo signal of the small Luneburg lens reflector.
The small Luneburg lens reflector needs to be measured many times and its echo intensity recorded, as shown in Figure 5. Using the same method, the big Luneburg lens reflector should be measured at the same position many times, and its echo intensity recorded, as shown in Figure 6. For this purpose, the true value of RCS and measurement results of the small Luneburg lens reflector are substituted into Eq. (7) to calculate the RCS of the big Luneburg lens reflector, as shown in Figure 7.
Fig. 5
The echo intensity of the small Luneburg lens reflector.
Fig. 6
The echo intensity of the big Luneburg lens reflector.
Fig. 7
The RCS of the big Luneburg lens reflector. RCS, radar cross-section.
The average RCS of the big Luneburg lens reflector is calculated to be 42.12 m2, which differs from its true value (41 m2) by 2.73%, indicating that the design of the RCS measurement system is reasonable. The measurement error may result from the drift of the measurement system and environmental noise.
The study of the effect of the distance and elevation between the target and the radar on the measurement results requires adjusting the position of the big Luneburg lens reflector. To this aim, the support rod is initially displaced 5000 m from the radar to measure its echo intensity. Then, the height of the support rod is increased to measure its echo intensity, as shown in Table 2. The RCS of the big Luneburg lens reflector is calculated based on the true value of the RCS of the small Luneburg lens reflector and its measurement results at 4500 m, as shown in Table 2.
The intensity and RCS of the big Luneburg lens reflector at different positions.
| 4500 | 0.6 | 85.19 | 42.12 | 2.73 |
| 5000 | 0.6 | 83.26 | 42.06 | 2.58 |
| 5000 | 1 | 90.81 | 41.77 | 1.89 |
| 5000 | 2 | 96.64 | 42.43 | 3.48 |
RCS, radar cross-section.
As shown in Table 2, when the support rod is installed at different distances and heights, the error between the RCS and the true value of the big Luneburg lens reflector obtained by the measurement system is < 4%.
After adjusting the position and height of the big Luneburg lens reflector and calculating its RCS in accordance with the measurement results of the small Luneburg lens reflector at 4500 m, there will be different calibration errors in GPR. As reflected in Table 2, the RCS values are relatively close, indicating that this measurement method eliminates the relative calibration error in GPR and yields ideal measurement results.
Moreover, at a distance of 5000 m, the echo intensity increases gradually with an increase in the height of the big Luneburg lens reflector, where the change mode is nonlinear. This is because, under GPR conditions, the multipath effect of electromagnetic wave propagation and the directional function of radar antenna jointly affect the antenna radiation field, making the intensity of antenna radiation field change irregularly, resulting in beam splitting [20]. That is, at the same distance, with an increase in height, the intensity of the radiation field will change alternately from weak to strong and then from strong to weak. Therefore, when measuring the RCS of the target in GPR, radiation field intensity should be analysed, and measurements should be calibrated in advance. Moreover, the influence of various factors needs to be fully considered, and reasonable measurement parameters must be selected to improve the accuracy of RCS measurement.
The RCS of the ship in the open sea is measured by the measurement system. This system is installed on Ship A, and the measured target is Ship B. The two ships lie at a certain distance. Ship B is a special experimental ship whose RCS value has already been measured, which can be used as the true value to compare with the measurement results.
Throughout measurements, the weather was fine, and the sea condition was grade 4–5. The standard metal ball is selected as the calibration body, fixed on a helium balloon, and released on Ship B.
The previous analysis concluded that when measuring the RCS of the target in GPR, the radiation field intensity distribution and parameters of the target should be considered simultaneously. The radiation field power gain factor of the measurement system can be calculated according to the height of the measurement system on Ship A and Eq. (3), as shown in Figure 8. The height of the first deck of Ship B is about 3 m, and that of the second deck is about 5 m, and its main reflection centre is concentrated in the range of 3–8 m. Thus, combined with Figure 3, it is appropriate to set the distance between Ship A and Ship B as 12 km. At this point, the first beam of the measurement system completely covers Ship B.
Fig. 8
The radiation field power gain factor.
As the ship belongs to a large target, it is necessary to consider whether a single radar beam can cover the longitudinal direction of the ship [12]. According to the size of the tested Ship B and 3 dB beam width, the distance between the two ships is set to be 12 km to meet the target beam coverage requirements. The starboard side of Ship B faces the measuring radar. The single measurement results are depicted in Figure 9. The target echo in Display P is clear; there is no interference around, and the echo envelope structure in Display A is stable.
Fig. 9
The echo of Ship B.
The method for processing an echo amplitude profile:
Select the maximum echo amplitude value from the multi-frame data within the scanning range of Ship B, and take it as the echo amplitude of Ship B. Read the noise power of multiple frames of the pure on-target area, and calculate its root mean square (RMS) value as the noise amplitude of Ship B echo. The echo intensity of Ship B is obtained by subtracting the noise amplitude from the echo amplitude. Substituting the echo intensity and other measurement site parameters into Eq. (7) and using the calibration data of the standard metal ball, the RCS of Ship B is calculated to be 57.01 dBsm.
To measure the circumferential RCS of Ship B, let it sail in a counterclockwise circle with a circumferential radius of 2.5 km [21]. Throughout measurements, the measurement system records the real-time information of Ship A and Ship B by means of the data recording equipment. In data analysis, timestamp synchronisation is used for multiple data alignments to solve the target's azimuth and its corresponding RCS. After calculation, the circumferential RCS of ship B is obtained, as shown in Figure 10, and compared with its true value.
Fig. 10
Comparison of the measured RCS (blue) of Ship B and its true value (red). RCS, radar cross-section.
The average RCS of Ship B is 39.89 dBsm, and the maximum RCS is located on the port side of Ship B, that is, about 58.99 dBsm. As shown in Figure 10, the mean RCS value of Ship B is about 2 dBsm smaller than its true value; however, the changing trend of the RCS curve is basically consistent, verifying the accuracy of this measurement. Error analysis: the true RCS value of Ship B is obtained in the port with a special RCS measurement radar. This measurement radar is a mechanical scanning radar with a low data rate. Although it has been improved, the measurement accuracy is still not as good as that of the special RCS measurement radar. Ship B has been moving throughout measurements, and its azimuth and corresponding RCS are calculated.
Therefore, its RCS curve is not continuous enough and contains an error. Furthermore, the two measurements do not meet the far-field conditions in the absolute sense, and there are both measurement errors.
The first part determines that the error in the target's position calculation will cause a measurement error of the target RCS. To study the impact of the error in the target's position calculation during the measurement process, the echo intensity of different beams is recorded in a scanning cycle, and their corresponding RCS is calculated, as shown in Table 3. As can be seen, Beam 3 faces the geometric centroid of Ship B, and its corresponding RCS is taken as the standard value, which is compared with the RCS of other positions to calculate its error.
The RCS error of Ship B.
| 1 | 57.04 | 9.13 |
| 2 | 57.44 | 0.35 |
| 3 | 57.46 | – |
| 4 | 57.42 | 1.00 |
| 5 | 57.48 | 0.36 |
RCS, radar cross-section.
As can be seen, the RCS corresponds to multiple beam changes within a certain range. Compared with its standard value, the measured value may be larger or smaller. As the target ship is a non-cooperative target in actual measurement, its actual position and RCS cannot be determined accurately, which can only be solved by radar echo. In case of solution errors, the size of the error and the target RCS error calculated from this cannot be determined; only the uncertainty range can be given. Thus, the uncertainty caused by the error in the target's position calculation should be considered while analysing the uncertainty of the target RCS measurement results. According to Table 3, the uncertainty range of target RCS caused by the error in the target's position calculation is −0.42 dB and 0.02 dB.
This paper considered the coupling electromagnetic reflection characteristics between the target and water surface. The influence of the multipath effect of electromagnetic wave propagation and the directivity function of radar antenna on the antenna radiation field and the resulting RCS measurement calibration error was analysed, and the RCS calculation formula, which could eliminate the error, was deduced.
An RCS measurement system was designed by improving a conventional radar. This system was employed to measure two Luneburg lens reflectors of different sizes many times in an open lake. Compared with its true value, the measurement error of the big Luneburg lens reflector was 2.73%. It was demonstrated that the RCS measurement system and calculation method were reasonable, and the relative GPR calibration error was eliminated. Variations in antenna radiation field under GPR conditions were also studied.
The system was used to measure the RCS of a ship target in the open sea. The mean RCS value of ship is about 2 dBsm smaller than its true value, but the changing trend of the two RCS curves are basically consistent. The uncertainty in RCS measurement caused by the ship's position calculation error was also examined.
In this paper, a classical theory was utilised to calculate the sea surface reflection coefficient, and the impact of the sea surface reflection coefficient on the coupling reflection characteristics of the target and the sea surface under different sea conditions was not deeply studied, which will be the focus of future research.