^{1}

^{1}

^{1}

One of the important applications of high frequency-ground wave radar (HFGWR) is to detect offshore ships. A proper method should be used to obtain the ship radar cross section (RCS), which is a key parameter of the ship. This paper proposes a method based on an automatic information system (AIS). The relationship of the ship RCS versus bearing for different frequencies is analyzed by processing multifrequency HFGWR data. With this new method, bearing information is taken into consideration, which is not the case in traditional empirical formulas. The results provide prior knowledge for ship detection and tracking; therefore, the probability of detection is significantly improved.

Over-the-horizon radar, HF ground-wave radar (HFGWR), has emerged in the past few decades for both monitoring ocean kinetic parameters (e.g., current, wave, and wind) and detecting moving targets on the ocean surface [

There is no simple formula applicable for calculating the RCS in the resonance region. The RCS is typically obtained by numerical simulation and experimental measurements. However, due to experimental conditions and cost constraints, it is challenging for HFGWR to obtain a large number of ship RCSs with experimental measurements. Certain numerical simulation methods, for example, the method of moments (MOM) and the finite difference time domain method (FDTD), have been applied to analyze RCSs in the HF bands [

By analyzing experimental data of HFGWR, Barrick proposed that the target echo is proportional to the sixth power of the height of the mast [

Because of the limitations of previous RCS studies, a more convenient and comprehensive method should be developed. This paper proposes a method based on experimental HFGWR data with the help of automatic information system (AIS) information. A large number of targets are analyzed, and the relationship of the RCS versus the bearing for a variety of frequencies is obtained.

The radar equation for HFGWR is

When the HFGWR operates in steady state,

The attenuation factor

Both the echo power of ships and Bragg waves decline when distance increases. In the case of a fully developed sea surface, Barrick proposed that the RCS of Bragg wave echo per unit area is invariant [

Based on this equation, the RCS of the Bragg wave echo at different ranges is invariant. From (

The AIS information provides characteristics of offshore ships including location, velocity, and heading. By matching AIS information and HFGWR data, some matched targets can be obtained. The bearings to the radar site of these targets can be calculated with the location of the radar site, the targets, and the heading of the targets (defining the bearing to be 0° when the bow is facing the radar site and 180° when the stern is facing the radar site), as illustrated in Figure

Bearing of ship to radar site.

The HFGWR used in this paper is a wide beam radar. Therefore, a digital beamforming (DBF) algorithm should be implemented before calculating the power of target echoes. When the receiving antenna is determined, the DBF weighting values at different bearings are identical; thus, there is no need to consider the problem of weight compensation at different bearings.

As described above, the RCS of the matched targets can be expressed as

Although the absolute value of RCS is not obtained, the relationship of RCS versus bearing can be characterized using the target power. To illustrate this relationship more clearly, it is normalized as (

The experimental data used in this paper is collected with a multifrequency HFGWR system. This system operates at several different carrier frequencies by time-division multiplexing, and its transmitted waveform is a linear frequency-modulated interrupted continuous wave (FMICW). The sweep bandwidth can be preset so that the distance resolution is adjustable from 1 km to 5 km. The radar operates at two frequencies, frequency 1 (10.8 MHz) and frequency 2 (8.2 MHz). The average transmission power is 300 W, and the farthest detection range designed is 150 km. The transmitting antenna is a 3-element Yagi antenna. The receiving antenna is a linear array antenna of eight 2.5 m passive monopole helical antennas. The antenna system used in the experiment has a wide beam width of 35° (corresponding to operating frequency 8.2 MHz) and 24° (corresponding to 10.8 MHz). The bearing of the target is calculated by the super resolution algorithm multiple signal classification (MUSIC). The coherent integration time of the HFGWR is sufficiently long that the velocity resolution is relatively high. The targets can be distinguished by their bearing, distance, and velocity differences.

During the field experiment, a set of AIS receivers was used to collect ship information at sea (within approximately 60 km), including location, heading, and velocity. The AIS information can be used as auxiliary information for signal processing [

The following results were obtained from HFGWR field experimental data collected at the Zhujiajian radar site (122.4275 E, 29.8931 N), Zhejiang Province, China, on August 29, 2010. After detecting and matching, the relationship of the RCS versus the bearing can be obtained from the echo power of the ships and the bearing to the radar site.

Figure

Vessel’s Details

Ship Type: Crude oil tanker

Year Built: 1991

Length × Breadth: 247 m × 42 m

Gross Tonnage: 53892, Dead Weight: 97002t

Speed recorded (Max/Average): 10.3/7.1 knots

Flag: China [CN]

Call Sign: BLAG8

IMO: 8908222, MMSI: 413398000

Results of target detection and matched AIS information.

The corresponding dynamic information, such as location, heading, and velocity, can be received and decoded by the AIS receiver.

Figure

RCS versus bearing at two frequencies for one target.

RCS ratio of two frequencies for one target.

To obtain the overall characteristics of ship RCSs, the influence of the ship size on the RCS must be decoupled and eliminated. Fifteen matched ships with similar sizes are chosen, all approximately 250 m in length. The bearings of these ships range from 40° to 160°. Other ships are beyond the detection range or have an undetectably low RCS.

The RCSs of these ships are calculated by the proposed method above (for frequency 1). For each ship and bearing, the RCS is plotted as blue circles in Figure

RCS versus bearing for

This paper proposed a method to obtain the relationship of the RCS versus bearing based on HFGWR experimental data. The result is similar to certain numerical simulations [

The authors declare that there are no conflicts of interest regarding the publication of this paper.

This work was supported by the National Natural Science Foundation of China (61671331) and the National Key Scientific Instruments and Equipment Development Project of China (2013YQ160793).