KEYWORDS: Radar, Motion models, Target detection, Signal detection, Monte Carlo methods, Motion estimation, Error analysis, Optimization (mathematics), 3D modeling, Ray tracing
This paper describes the application of the Deep Target Extractor (DTE), developed from the Maximum Likelihood Probabilistic Multiple Hypothesis Tracker (ML-PMHT), to measurements from an over-the-horizon radar (OTHR) observing a highly maneuvering target (HMT). We describe the motion model for HMTs that start at very large speeds and makes extremely sharp turns. We then present a measurement model for OTHR that uses ray-tracing software to produce detections across multiple signal paths resulting from ionospheric refractions. Next, we describe the DTE operation in the multiple signal path framework of the OTHR. Finally, we present a test scenario where the DTE performance shows low root-mean-square error (RMSE) for HMT motion parameter estimation.
This paper discusses the theoretical considerations for direction-of-arrival (DOA) estimation using antenna arrays in the presence of mutual coupling. In arrays, the relative proximity of antenna elements results in some manner of near-field mutual coupling that can negatively impact the array performance. In particular, mutual coupling can degrade the quality of DOA estimations and reduce the ability of the array to perform high-quality correlation processing and direction finding. The expected variance of an array performing DOA estimation is inversely related to the Fisher information matrix of the system. Theoretical radiated fields of a linear antenna array were compared to more realistic behavior of the equivalent architectures produced in electromagnetics simulation software. The mutual coupling between all the elements in an array can be a difficult phenomenon to precisely define, as it is an iterative process with many higher-order effects. To circumvent this, a mutual coupling matrix is defined as the relation between the theoretical radiation characteristic of an array and its simulated counterpart. An inverse solution method was used to solve for the mutual coupling interactions. The expected impact of mutual coupling in a DOA estimation context was then mathematically established by propagating the mutual coupling matrix through calculation of the Fisher information matrix and compared to the case of no mutual coupling. It was found that taking mutual coupling into consideration yields a higher Cramer-Rao Lower Bound and as a result a greater RMS angle error in a DOA estimation context. Mutual coupling was also found to have a somewhat greater impact on the Cramer-Rao Bound at S-band than at X-band.
The performance of different random array geometries is analyzed and compared. Three phased array geometries are considered: linear arrays with non-uniform randomized spacing between elements, circular arrays with non-uniform element radii, and ad hoc sensor networks with elements located randomly within a circular area. For each of these array geometries, computer simulations modeled the transmission, reflection from an arbitrary target, and reception of signals. The effectiveness of each array’s beamforming techniques was measured by taking the peak cross-correlation between the received signal and a time-delayed replica of the original transmitted signal. For each array type, the correlation performance was obtained for transmission and reception of both chirp waveforms and ultra-wideband noise signals. It was found that the non-uniform linear array generally produced the highest correlation between transmitted and reflected signals. The non-uniform circular and ad hoc arrays demonstrated the most consistent performance with respect to noise signal bandwidth. The effect of scan angle was found to have a significant impact on the correlation performance of the linear arrays, where the correlation performance declines as the scan angle moves away from broadside to the array.
The use of noise waveforms for radar has been popular for many years; however, not much work has been done to extend their use to long range applications. To understand the practicality of using noise for this work, the correlation values between transmitted and received signals were investigated as well as the ratio of reflected to transmitted power. This was done for both ground clutter and simple shapes representing targets of interest. Reflections from these different surfaces are dependent on the frequency of operation, polarization, angle of incidence, and target material. To act as a direct comparison to the noise waveform, a chirp signal was also reflected from these surfaces and correlated with the originally transmitted signal. For terrain, it was found that the noise offers similar correlation patterns as the chirp waveform but slightly larger reflected power for certain cases. Additionally, noise waveforms have decreased correlation values compared to chirp waveforms at low angles. For the simple shaped targets, the noise and chirp signals had similar correlation patterns, values, and power ratios.
Passive radar systems, which utilize broadcast and navigation signals as sources of opportunity, have attracted significant interests in recent years due to their low cost, covertness, and the availability of different illuminator sources. In this paper, we propose a novel method for synthetic aperture imaging in multi-static passive radar systems based on a group sparse Bayesian learning technique. In particular, the problem of imaging sparse targets is formulated as a group sparse signal reconstruction problem, which is solved using a complex multi- task Bayesian compressive sensing (CMT-BCS) method to achieve a high resolution. The proposed approach significantly improves the imaging resolution beyond the range resolution. Compared to the other group sparse signal reconstruction methods, such as the block orthogonal matching pursuit (BOMP) and group Lasso, the CMT-BCS provides significant performance improvement for the reconstruction of sparse targets in the redundant dictionary with high coherence. Simulations results demonstrate the superior performance of the proposed approach.
This paper proposes efficient target localization methods for a passive radar system using bistatic time-of-arrival
(TOA) information measured at multiple synthetic array locations, where the position of these synthetic array
locations is subject to random errors. Since maximum likelihood (ML) formulation of this target localization
problem is a non-convex optimization problem, semi-definite relaxation (SDR)-based optimization methods in
general do not provide satisfactory performance. As a result, approximated ML optimization problems are
proposed and solved with SDR plus bisection methods. For the case without position errors, it is shown that the
relaxation guarantees a rank-one solution. The optimization problem for the case with position errors involves
only a relaxation of a scalar quadratic term. Simulation results show that the proposed algorithms outperform
existing methods and provide mean square position error performance very close to the Cramer-Rao lower bound
even for larger values of noise and position estimation errors.
In this paper, we consider resolving over-the-horizon radar (OTHR) Doppler returns. A high-resolution time-frequency
(TF) representation of the received signal is obtained by using the local polynomial Fourier transform
(LPFT). From the optimally concentrated LPFT, multicomponent Doppler signatures, which are only several
frequency bins apart, are extracted using an instantaneous frequency estimation method based on the Viterbi
algorithm. The performance of the proposed method is validated using real data.
This paper addresses the issue of spatial diversity in radar applications. The has been an increased need for information via radio frequency (RF) detection of airborne and ground targets while at the same time the electromagnetic spectrum available for commercial and military applications has been eroding. Typically, information concerning ground and air targets is obtained via monostatic radar. Increased information is often equated with increased bandwidth in these monostatic radar systems. However, geometric diversity obtained through multistatic radar operation also affords the user the opportunity to obtain additional information concerning these targets. With the appropriate signal processing, this translates directly into increased probability of detection and reduced probability of false alarm. In the extreme case, only discrete Ultra Narrow Band (UNB) frequencies of operation may be available for both commercial and military applications. As such, the need for geometric diversity becomes imperative.
GPR data collection in the frequency range of 3 to 66 MHz has been performed by the U.S. Air Force Research Laboratory (AFRL). To process this data into images, the relative permittivity of the medium is a key parameter since it determines the propagation velocity through the medium. Image formation using phase information returned from buried scatterers can be improved by matching the processing to the environment in which radar measurements were taken. Considering the permittivity to be an unknown scalar parameter, several images can be formed as a function of the unknown parameter leading to a choice which maximizes the sensitivity and resolution of the resulting image. In this paper we extend upon this concept to include variations of the medium that are nonhomogeneous with respect to the physical locations of receiving sensors.
In this paper, the effects of receive antenna placement on image formation in a surface contact synthetic aperture radar are investigated from an experimental perspective via the analysis of high frequency ground penetrating radar data. A receive array containing 121 measurements is used to form a baseline image of subsurface features including horizontal tunnels or drifts to depths approaching 200 feet. This result is compared to images formed using far fewer spatial measurements (as few as 37 samples out of 121 measurements available) collected randomly from throughout the same field of view.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.