Yoshio YAMAGUCHI Masafumi NAKAMURA Hiroyoshi YAMADA
One of the polarimetric radar applications is classification or identification of targets making use of the scattering matrix. This paper presents a decomposition scheme of a scattering matrix into three elementary scattering matrices in the circular polarization basis. The elementary components are a sphere, a diplane (dihedral corner reflector), and a helix. Since a synthetic aperture FM-CW radar provides scattering matrix through a polarimetric measurement, this decomposition scheme was applied to the actual raw data, although the matrix is resulted from a swept frequency measurement. Radar imaging experiments at the Ku band (14.5-15.5GHz) were carried out to obtain a total of 6464 scattering matrices in an imaging plane, using flat plates, corner reflectors and wires as elementary radar targets for classification. It is shown that the decomposition scheme has been successfully carried out to distinguish these targets and that the determination of rotation angle of line target is possible if the scattering matrix is classified as a wire.
Jian YANG Yoshio YAMAGUCHI Hiroyoshi YAMADA Shiming LIN
For the completely polarized wave case, this paper presents the explicit formulae of the characteristic polarization states in the co-polarized radar channel, from which one can obtain the CO-POL Max, the CO-POL Saddle and the CO-POL Nulls in the Stokes vector form. Then the problem on the polarimetric contrast optimization is discussed, and the explicit formula of the optimal polarization state for contrast enhancement is presented in the Stokes vector form for the first time. To verify these formulae, we give some numerical examples. The results are completely identical with other authors', which shows the validity of the presented method.
A radio propagation experiment at the Okinawa Radio Observatory of the Communications Research Laboratory is investigating the feasibility of calibrating the spaceborne precipitation radar onboard the Tropical Rainfall Measuring Mission by using the path-averaged rainfall rate estimated from rain attenuation. Because this estimated rainfall rate has errors due to the spatial inhomogeneity of rainfall rate and the variability of raindrop size distribution, we used distrometer data to evaluate both of these errors by computer simulation.
Shigeki OHSHIMA Yoshikazu ASANO Kunitoshi NISHIKAWA
We propose a method for accomplishing accurate RCS (Radar Cross Section ) images of a car in a compact range. It is an improved method based on an ISAR (Inverse Synthetic Aperture Radar) technique. To obtain accurate RCS values, an idea of an image correction function for the Fourier transform used in the ISAR processing is introduced. The role of the image correction function is to compensate the difference of the propagation loss as to the different scattering points on a target. As a result, `sensitivity' of imaging in the compact range is kept uniform. Hamming window is suitable for the Fourier transform to accomplish RCS images because of its low sidelobe level and the sharpness of a mainlobe. When hamming window is adopted, the spatial resolution is approximately twice the size of granularity which is determined by the ISAR parameters. To verify the improvement of the RCS images obtained by means of our method, several numerical target models are employed. The results of the investigation show that uniformity of `sensitivity' for obtained RCS images is achieved in the compact range and accurate images with the resolution of twice the size of granularity are accomplished without blurs or distortions in the unambiguous area. RCS images for rear aspects of a passenger car are investigated with the spatial resolution of 50 mm in the 60 GHz band. The RCS image varies with the aspect angle of the car and the specular reflection occurs for the millimeter wave. When the curvature on the car edge is small, a blurred RCS image is observed. The reason is that a scattering center of the specular reflection moves so widely that it can't be regarded as a fixed point. This causes elongation of the RCS image. A peak value in the dominant area for each aspect angle is less the 0 dBsm and no remarkable areas where the RCS value exceeds-20 dBsm is found any more on the car except such the dominant area.
Chih-ping LIN Motoaki SANO Matsuo SEKINE
The millimeter wave (MMW) radar has good compromise characteristics of both microwave radar and optical sensors. It has better angular and range resolving abilities than microwave radar, and a longer penetrating range than optical sensors. We used the MMW radar to detect targets located in the sea and among sea ice clutter based on fractals, wavelets, and neural networks. The wavelets were used as feature extractors to decompose the MMW radar images and to extract the feature vectors from approximation signals at different resolution levels. Unsupervised neural classifiers with parallel computational architecture were used to classify sea ice, sea water and targets based on the competitive learning algorithm. The fractal dimensions could provide a quantitative description of the roughness of the radar image. Using these techniques, we can detect targets quickly and clearly discriminate between sea ice, sea water, and targets.
Tamio SAITO Teruhisa NINOMIYA Osamu ISAJI Tominaga WATANAME Hiroshi SUZUKI Naofumi OKUBO
An important aspect of traffic safety is the development of aids that extend the driver's time and motion perception. One promising candidate is the compact, lightweight millimeter-wave FM-CW radar now being widely studied. Although the homodyne FM-CW radar is well known form its simplicity, it has a relatively low S/N ratio. This paper describes the principles behind our newly-developed heterodyne FM-CW radar and it's evaluation results. The heterodyne FM-CE radar generates sidebands by switching a front-end amplifier and also uses the heterodyne detection technique for gaining sensor sensitivity. The heterodyne FM-CW radar's signal to noise ratio was 19.5 dB better than previously designed homodyne FM-CW radar.
Yoshikazu ASANO Shigeki OHSHIMA Kunitoshi NISHIKAWA
This paper presents a method for simply estimating characteristics of signals received by a millimeter wave car radar. In this method, the substitution of a radar target with a set of scattering points is introduced to take account of the phenomenon that only a part of the target is irradiated with the radio wave from the radar antenna with a sharp beam; the phenomenon is peculiar to the car radar which operates in a compact range. The positions of these scattering points and the RCS values for the scattering points are appropriately determined on the basis of a measured RCS image for the target. The RCS image means a spatial distribution of RCS values on the surface of the target. In addition, influence of the ground, which is a dominant clutter in car radar environments, and characteristics of the car radar hardware can be included in the estimation method. The estimated characteristics of the signal received by the car radar are compared with the measured ones under typical cases in the car radar environments. The comparison verifies not only that the received signal characteristics are well estimated even when the range is rather short but also that the substitution of the target with scattering points is valid. The proposed method can realize the estimation of the received signal characteristics. Furthermore, the method can be developed into a computer simulation for evaluating the target detection performance of the car radar.
RCS fluctuation of targets such as ships can be reduced by the high-resolution radar. The high-resolution radar resolves the total radar echo into several parts which do not interfere each other. The reduction of interference gives stable target RCS. A simple model of the RCS fluctuation reduction is presented. Typical data for ships taken by an experimental radar which has range resolution about 0.75 m, are also shown. The analysis results show that the RCS fluctuation reduction agree with the simple model well.
A maximum diameter (Dmax) of raindrop should be assumed when rainfall rate (R) is estimates from the differential reflectivity (ZDR) and the horizontal reflectivity (ZH) measured with dual-polarization radar. If the assumed Dmax is different from actual Dmax, the estimated R contains errors. Using distrometer data, it was found that ZDR correlates with Dmax, and it was verified that when Dmax is adaptively determined by an empirical relationship between ZDR and Dmax, errors in estimated R can be reduced.
Toru SATO Toshihiro TERAOKA Iwane KIMURA
The MU radar of Japan is one of important candidates for providing accurate ground truth for the TRMM precipitation radar. It can provide the dropsize distribution data together with the background atmospheric wind data with high accuracy and high spatial resolution. Special observation scheme developed for TRMM validation using the MU radar is described, and preliminary results from its test experiment are shown. The high-resolution MU radar data are also used in numerical simulations to validate the rain retrieval algorithm for the TRMM PR data analysis. Among known sources of errors in the rain retrieval, the vertical variability of the dropsize distribution and the partial beam-filling effect are examined in terms of their significance with numerical simulations based on the MU radar data. It is shown that these factors may seriously affect the accuracy of the TRMM rain retrieval, and that it is necessary to establish statistical means for compensation. However, suggested means to improve the conventional α-adjustment method require careful treatment so that they do not introduce new sources of errors.
A method to correct the path-integrated attenuation derived from spaceborne radar measurement for the non-uniform beam filling (NUBF) effect is studied . A preliminary test using the data obtained from shipborne and ground-based radars is performed. It is found that the relation between the coarse-scale variability (radar-measurable quantity, σL) and the fine-scale variability (a quantity necessary for the NUBF correction, σH) of rain depends somewhat upon the rain cases studied and there still remains some underestimation in the corrected results. Nevertheless, the test result demonstrates the potential of utilizing the "local" statistical properties of rain in order to decrease the bias error in rain rate estimation caused by the NUBF.
Hua JIANG Motoaki SANO Matsuo SEKINE
We have compared the various raindrop-size distributions (DSD) with the recent experimental data collected by the distrometer. It is shown that the Weibull distribution is the best fit to the experimental data for drizzle, widespread and thunderstorm rain cases. By using this Weibull DSD, we obtained a new expression of the radar reflectivity factor (Z) and the rainfall rate (R) relation, that is Z=285R1.48, which gives few errors comparing to some measurements in TRMM frequency of 14GHZ.
Weather clutter was measured using an L-band long-range (200 nmi.) air-route surveillance radar (ARSR), from the stored data of five scans. It is shown that weather clutter obeys a Weibull distribution, and moreover such Weibull-distributed weather clutter also obeys a Weibull distribution after passing through a moving target indicator (MTI) processor realized by a double canceller. Some adaptive methods for suppressing Weibull-distributed clutter are summarized.
Yasuyoshi HORIBATA Hiroshi OIKAWA
Several major aircraft accidents have been attributed to low-altitude wind shears, which are normally caused by microbursts or gust fronts. Terminal Doppler Weather Radar (TDWR) systems are being installed near major airports for the detection of low-altitude wind shears. In order to develop a TDWR system further, low-altitude wind shears were numerically simulated in this study. The basic equations, which contain prognostic equations for air velocity, pressure, temperature, water vapor, and rainwater, were solved using a finite difference scheme. A terrain-following coordinate transformation was employed to simulate terrain effects. The simulation results are presented in this paper.
Yoshikazu KANAI Kazuhiro YASUDA Hiroshi OIKAWA Masaaki SUGANO Yuko SATO
Winter thundercloud on the coast of the Sea of Japan is difficult for the detection and forecast. This paper reports the new method of thundercloud detection using C-Band weather radar data and meteorological data from rawin sonde. This method includes the thunderclouds echo tracking and chronological evaluation of thundercloud life-cycle stages. The detection probability of critical area for cloud-to-ground lightning is over 90% on winter cloud echo data. This method is workable on realtime processing in work-station computer.
Hiroshi KUMAGAI Kenji NAKAMURA Hiroshi HANADO Ken'ichi OKAMOTO Naoki HOSAKA Noriaki MIYANO Toshiaki KOZU Nobuhiro TAKAHASHI Toshio IGUCHI Hiroshi MIYAUCHI
A new airborne rain radar named CAMPR (CRL Airborne Multiparameter precipitation Radar) has been developed for the major purpose of calibrating PR (Precipitation Radar) onboard TRMM (Tropical Rainfall Measuring Mission; scheduled to be launched in 1997) in orbit by observing the same rain with both CAMPR and TRMM satellite. CAMPR operates as a coherent radar at 13.8 GHz, the same frequency as TRMM-PR, and has polarimetric and Doppler capabilities. It is installed on a relatively small aircraft and can scan the antenna over a wide angle range, from the nadir to the near-horizon. These functions have been verified to work well and it is shown that the radar system is accurately calibrated. Examples of measurement data show CAMPR's high capability to extract various quantities relating to precipitation and cloud physics. Before the TRMM launch, CAMPR is being used to obtain TRMM-PR simulation data to help its algorithm development as well as to obtain data concerning precipitation and cloud physics.
Ysutaka MAKIHARA Naotaka UEKIYO Akira TABATA Yoshiyuki ABE
A verification is made on the accuracy of Radar-AMeDAS precipitation, which represents hourly precipitation over the Japanese Islands and the surrounding sea area with a spatial resolution of 5km using data from 5cm conventional radars, 10cm Fujisan Radar, and Automated Meteorological Data Acquisition System (AMeDAS) raingauge network. By comparing with data from a very dense raingauge network of the Tokyo Metropolitan Government, it is found that 1) Radar-AMeDAS precipitation shows good agreement if a positioning error of one pixel of 5km square is allowed 2) Radar-AMeDAS precipitation represents almost the average of raingauge measurements in the 5km square for most of the precipitation caused by a large scale disturbance, and 3) Radar-AMeDAS precipitation is close to the maximum raingauge measurement in the pixel when precipitation is extremely localized such as thunderstorms or showers. Radar-AMeDAS precipitations are compared also with AMeDAS measurements statistically with respect to the appearance rates, that is (total number of pixels where specific intensity is observed) / (total number of all pixels), for different precipitation intensities. The rate of Radar- AMeDAS precipitation shows excellent agreement with that of AMeDAS if radar echoes are observed at the altitude lower than 2km. Since Radar- AMeDAS precipitation on land sometimes represents the maximum of precipitation in a pixel for the purpose of unfailingly detecting extremely localized severe precipitation, it shows a high appearance rate at high precipitation intensity than AMeDAS, which is considered to represent statistically the average of a pixel. As a result, in estimating areal rainfall amounts, Radar- AMeDAS precipitation overestimates AMeDAS measurement by 8% at 5mm/h and by 12% at 40mm/h. Radar- AMeDAS precipitation over the sea, with no local calibration by AMeDAS and with little influence of orography, is 2% weaker in intensity than AMeDAS at 10mm/h, and 12% at 40mm/h.
Sevtap SAPMAZ Kazuya KOBAYASHI Alinur BUYUKAKSOY Gokhan UZGOREN
The E-polarized plane wave diffraction by a perfectly conducting strip located at the plane interface between two different media is analyzed by the Wiener-Hopf technique. Applying the boundary conditions to the integral representations for the unknown scattered field, the problem is formulated in terms of the modified Wiener-Hopf equation(MWHE), which is reduced to a pair of simultaneous integral equations via the factorization and decomposition procedure. The integral equations are solved asymptotically for large strip width via the method of successive approximations leading to the first, second and third order solutions, which are valid at high frequencies. The scattered far field expression is derived by taking the inverse Fourier transform and applying the saddle point method. It is shown that the high-frequency scattered far field comprises the geometrical optics field, the singly, doubly and triply diffracted fields and the lateral waves. Numerical examples of the radar cross section(RCS) and the lateral waves are presented, and the far field scattering characteristics discussed in detail.
Kiyonobu ABE Kazuhiro HIRASAWA Hideaki WATANABE
High power interference rejection characteristics of a sidelobe canceller which have not been well discussed yet are investigated through computer simulation and experiment in the real radio wave environment. To improve the high power interference rejection performance, a new method is considered. The performance of the method is also analyzed through computer simulation and experiment.
Yasutaka OGAWA Nobuyoshi KIKUMA
Signal processing antennas have been studied not only for interference suppression but also for high-resolution estimation of radio environment such as directions-of-arrival of incident signals. These two applications are based on the common technique, that is, null steering. This tutorial paper reviews the MUSIC algorithm which is one of the typical high-resolution techniques. Examining the eigenvector beam patterns, we demonstrate that the high-resolution capability is realized by steering nulls. The considerations will be useful for understanding the high-resolution techniques in the signal processing antennas. We then describe a modified version of MUSIC (Root MUSIC). We show the performance and robustness of the method. Furthermore, we introduce radar target identification and two-dimensional radar target imaging. These study fields are new applications of the signal processing antennas, to which a great deal of attention has been devoted recently.