Si-Ho Kim

Micro-Doppler Extraction and Analysis of the Ballistic Missile Using RDA Based on the Real Flight Scenario

Micro-Doppler (MD) caused by the motion of the ballistic missile can contribute to successful recognition of the ballistic missile. Considering the real observation scenario. This paper proposes a method to derive the MD image of the ballistic missile by applying the range-Doppler algorithm (RDA) based on the real ∞ight scenario and analyzes the factor for the real-time MD imaging. Simulation results using the ∞ight trajectory constructed using the real target parameter demonstrate that we need a new cost function for phase adjustment and a new method for range alignment. In addition, matched-flltering needs to be performed in the baseband, and a su-cient PRF is required to prevent discontinuity of the MD image. Dechirping of MD and flltering of the random movement are also needed for a clear MD image. Among the various weapons used in the modern battlefleld, the ballistic missile in∞icts the biggest threat due to its high maneuvering speed and low radar cross section (RCS), and thus defending against the ballistic missile is a major issue. Recently, a theory has been developed to explain micro-Doppler (MD) efiect caused by the micro motion of the target and applied for radar target recognition purpose (1,2). In the case of the ballistic missile, three motion components | spinning, conning and nutation | cause MD, and they can be utilized for target recognition in combination with the motion parameter (1,3,4). However, very little research has been reported on its application to the real ∞ight scenario and the factor that needs to be considered for the real-time MD imaging. In this paper, considering the real observation scenario by a radar, we propose a method to extract an MD signature of a ballistic missile engaged in the real ∞ight scenario by applying the range-Doppler algorithm (RDA), which is generally used to form the inverse synthetic aperture radar (ISAR) image (5{8), and analyze various factors for the real-time high quality MD image. For this purpose, we constructed the ∞ight trajectory by using the real motion parameters of a 500km range scud missile conducted a translation motion compensation (TMC). Then, the time-varying MD image was formed by applying the time-frequency transform (TFT). Various simulations were performed by using the obtained MD image to study the requirement for the real-time MD imaging. Simulation results obtained by using a target composed of the point scatter demonstrate that the MD signature can be successfully constructed by using the range-Doppler algorithm. However, a new method for TMC is required for real-time high-quality MD imaging. In addition, matched-flltering (MF) in the baseband is required to form a focused image, and a su-cient PRF is needed to remove discontinuity. The re∞ected signal needs to be dechirped to reduce the required PRF, and a fllter needs to be designed to remove the random movement.

An Efficient ISAR Imaging Method for Multiple Targets

This paper proposes an e-cient method to obtain ISAR images of multiple targets ∞ying in formation. The proposed method improves the coarse alignment and segmentation of the existing method. The improved coarse alignment method models the ∞ight trajectory using a combination of a polynomial and Gaussian basis functions, and the optimum parameter of the trajectory is found using particle swarm optimization. In the improved segmentation, the binary image of the bulk ISAR image that contains all targets is constructed using a two-dimensional constant false alarm detector, then the image closing method is applied to the binary image. Finally, the connected set of binary pixels is used to segment each target from the bulk image. Simulations using three targets composed of point scattering centers and the measured data of the Boeing747 aircraft prove the efiectiveness of the proposed method to segment three targets ∞ying in formation.

MICRO-DOPPLER ANALYSIS OF KOREAN OFFSHORE WIND TURBINE ON THE L-BAND RADAR

To cope with the energy shortage and the rising cost of the fossil fuel, many wind farms are being constructed under the supervision of Korean government along the coasts of Korean peninsula to generate clean and renewable energy. However, construction of these wind farms may cause negative efiect on various L-band radars in operation. This paper presents the result of the micro-Doppler (MD) analysis of the in∞uence of the wind turbine on the L-band radar using the point scatterer model and the radar cross section of the real turbine predicted by the method of physical optics. The simulation results obtained at three observation angles show that the range of MD occupies a considerable portion of the helicopter MD range, and thus, the operations using helicopters need to be avoided in the wind farm region, and additional radars are required for the recognition of helicopter-like objects.

Efficient classification of ISAR images using 2d fourier transform and polar mapping

This paper proposes an efficient method to classify inverse synthetic aperture radar (ISAR) images. The proposed method achieves invariance to translation and rotation of ISAR images by using two-dimensional (2D) Fourier transform (FT) of ISAR images, polar mapping of the 2D FT image, and a simple nearest-neighbor classifier. In simulations using ISAR images measured in a compact range, the proposed method yielded high classification ratios with small-sized data regardless of the location of the rotation center, whereas the existing method was very sensitive to the location of it.

Classification of Targets Improved by Fusion of the Range Profile and the Inverse Synthetic Aperture Radar Image

The range proflle (RP) and the inverse synthetic aperture radar (ISAR) image are the useful radar signatures for classifying unknown targets because they can be used regardless of day- night and weather conditions. Since classiflcation that uses RP and ISAR is heavily dependent on ∞ight conditions, however, much more study is required on this topic. This paper proposes an e-cient method of classifying targets by using a classifler-level fusion of RP and ISAR as well as a scenario-based construction method of the training database. Simulation results using the flve targets composed of point scatterers prove that the proposed method yields high classiflcation results when the targets are ∞ying in a variety of directions at both short and long ranges.

EFFICIENT RANGE ALIGNMENT ALGORITHM FOR REAL-TIME RANGE-DOPPLER ALGORITHM

When deriving a range-Doppler image or a time-frequency image of a fast-maneuvering target at long range, existing range alignment methods yield poor results due to the large numbers of range profiles (RPs) and range bins that are required for this task. This paper proposes a threestep range alignment method to overcome the problems of these existing methods and to yield focused images: (1) coarse alignment using the interpolated center of mass of each RP, (2) fine alignment with an integer step using an entropy cost function, and (3) fine-tuning using particle swarm optimization. Compared to existing methods, the proposed method is computationally more efficient and provides better image focus.

Estimation of the micro-motion parameters of a missile warhead using a micro-Doppler profile

This paper deals with the problem of estimating micro-motion parameters. The proposed algorithm uses independent-component analysis to decompose the received signals into individual scatterer signals, and uses particle-swarm optimization to estimate the micro-motion parameters in a micro-Doppler profile. Simulation results show that the proposed method can successfully estimate micro-motion parameters of a warhead model consisting of many point scatterers.

An Efficient Method to Extract the Micro-Motion Parameter of the Missile Using the Time-Frequency Image

It is very difficult to intercept the missiles because of the small radar cross-section and the high maneuverability. In addition, due to the decoy with the similar motion parameters, additional features other than those of the translation motion parameters need to be developed. In this paper, for the successful recognition of missiles, we propose an efficient method to extract micro-motion parameters and scatterers of the missile engaged in the micro motion. The proposed method extracts motion parameters and scatterers by using the matching score between the modeled micro-Doppler function and the time-frequency binary image as a cost function. Simulation results using a target composed of the point scatterer show the parameters and the scatterers were accurately extracted.

Efficient Measurement System to Investigate Micro-Doppler Signature of Ballistic Missile

Micro-Doppler (MD) shift caused by the micro-motion of a ballistic missile (BM) can be very useful to identify it. In this paper, the MD signatures of three scale-model BMs are investigated using a portable measurement system. The measurement system consists of an X-band 2-by-2 phase comparison mono-pulse radar, and a mechanical device that can impart controlled spinning and coning motions simultaneously to a model to yield the MD signature that replicates the characteristic of each target and the corresponding micro-motion. The coning motion determined the overall period of MD, and the spinning motion increased its amplitude. MD was also dependent on aspect angle. The designed system is portable, and can implement many micro-motions; it will contribute to analysis of MD in various situations.