Fraunhofer Institute for High Frequency Physics and Radar Techniques
Fraunhofer Institute for High Frequency Physics and Radar Techniques
Wagner S.A.,Fraunhofer Institute for High Frequency Physics and Radar Techniques
IEEE Transactions on Aerospace and Electronic Systems | Year: 2016
A combination of a convolutional neural network, which belongs to the deep learning research field, and support vector machines is presented as an efficient automatic target recognition system. Additional training methods that incorporate prior knowledge to the classifier and further improve its robustness against imaging errors and target variations are also presented. These methods generate artificial training data by elastic distortion and affine transformations that represent typical examples of image errors, like a changing range scale dependent on the depression angle or an incorrectly estimated aspect angle. With these examples presented to the classifier during the training, the system should become invariant against these variations and thus more robust. For the classification, the spotlight synthetic aperture radar images of the moving and stationary target acquisition and recognition database are used. Results are shown for the ten class database with a forced decision classification as well as with rejection class. © 1965-2011 IEEE.
News Article | January 4, 2016
Abandoned items of luggage are frequently found at airports and train stations. This is a case for the emergency services, who have to assume that these items might contain bombs. They must assess the potential threat quickly, avert any possible danger, and preserve evidence for criminal proceedings. In the future, police will have the support of a remote-controlled sensor system as they go about their duties. Fraunhofer researchers are developing this sensor suite in cooperation with industry partners and criminal investigation authorities. Anyone who forgets their luggage in public places, airports or train stations will spark off a large-scale police operation. Time and again, suitcases, bags or backpacks left lying around unsupervised cause a bomb alert. Admittedly, most abandoned luggage items turn out to be harmless. But in the first instance the emergency services have to proceed on the assumption of possible danger and check whether they are dealing with an improvised explosive device (IED) that might blow up at any time. This involves getting up close to the luggage to inspect it. A system that makes it possible to assess the danger of the situation quickly – and also records 3D images of the contents and shape of the luggage as well as of the surrounding area – would make the specialists' work considerably easier, speed up the reconnaissance process, and minimize the risk for the emergency personnel. Since November 2014, researchers at the Fraunhofer Institute for High Frequency Physics and Radar Techniques FHR in Wachtberg have been developing such a system together with the North Rhine-Westphalia State Office of Criminal Investigation, the Leibnitz University in Hannover, ELP GmbH and Hentschel System GmbH. The German Federal Office of Criminal Investigation in Wiesbaden and the German Federal Police Force are supporting the project as additional expert consultants. The German Federal Ministry of Education and Research is funding the USBV Inspector project with a grant of two million euros as part of its Research for Civil Security program. Emergency services do not have to enter the danger zone The system the researchers have developed comprises a multimodal sensor suite consisting of a millimeter wave scanner, a high-resolution digital camera, and a 3D environment monitoring system. The components are contained in a housing and mounted on a robot platform. Bomb disposal engineers remotely control the robot from a safe distance. Its swiveling 3D sensors make a three-dimensional survey of the crime scene, and the digital camera provides high-resolution images for later optical evidence preservation. Meanwhile the millimeter wave sensor scans the source of danger and creates an image of what's inside. A built-in embedded PC on the robot collects the data and sends it to the investigators, where it will be merged on the computer by means of sensor data fusion. "Up to now our techniques have not allowed us to form a 3D outline of suitcase bombs, and it has been impossible – or only partially possible – to make a spatial map of the contents. With the sensor suite we can visualize in three dimensions what's inside a luggage item, and so determine the composition of the bomb and how the parts are arranged in the luggage," explains Stefan A. Lang, team leader at the FHR and the project's coordinator. This lets the explosives experts quickly assess the threat, and going forward they will also be able to preserve as much evidence as possible about the bomb. Until now, specialists were often forced to destroy suitcase bombs – making it difficult to identify the perpetrators. Other advantages of the contact-free detection system: it is light, compact, and platform independent, which means it can be mounted on any robot. Within the project, the FHR researchers are developing the millimeter wave scanner (also referred to as a radar sensor) for fast reconnaissance. This scanner allows a very high depth resolution. "For the radar we make use of the synthetic aperture radar, or SAR, principle, by which the sensor is moved along a trajectory, a kind of track – from left to right in front of the case, for example – and the Doppler information generated in the process is used to create an image," explains Lang. Apart from the research work on the sensor, the expert and his team are also looking into ways of determining the optimum trajectory for surveying an object. This depends on the shape of the luggage item or container, its position in the environment, and the position of the robot. A radar sensor demonstrator will be ready in April 2016. Extensive field tests of the remote-controlled sensor suite begin in the middle of 2017, with the multimodal sensor suite set to be launched in 2019.
Gierull C.H.,Defence Research and Development Canada |
Sikaneta I.,Defence Research and Development Canada |
Cerutti-Maori D.,Fraunhofer Institute for High Frequency Physics and Radar Techniques
IEEE Transactions on Geoscience and Remote Sensing | Year: 2013
Ground moving target indication (GMTI) from space has currently raised interest from wide area traffic monitoring as well as for military surveillance activities. This paper presents theoretical and real data GMTI results of RADARSAT-2's Moving Object Detection Experiment based on data gathered during the commissioning phase in February 2008. The proposed constant false alarm rate target detection is based on a novel and thorough analysis of the multilook test statistics. The practical and relevant case in which a target, for instance a passenger car, occurs in only l-out-of-n ground resolution multilook cells is analyzed. This general case is analyzed for the determination of adequate detection thresholds as well as the anticipated probability of detection P d particularly with regard to a varying degree of terrain heterogeneity and target characteristics. The overall false alarm rate P fa is significantly reduced by a complementary detection step using the along-track interferometric phase. The validity of these theoretical findings are corroborated by real two-channel space-based synthetic aperture radar-GMTI data of civilian vehicles of opportunity, whose main parameters have been estimated and compared with the derived Cramér-Rao Bounds. © 2012 IEEE.
Cerutti-Maori D.,Fraunhofer Institute for High Frequency Physics and Radar Techniques |
Cerutti-Maori D.,Defence Research and Development Canada |
Sikaneta I.,Defence Research and Development Canada
IEEE Transactions on Geoscience and Remote Sensing | Year: 2013
This paper generalizes the well-known displaced-phase-center antenna (DPCA) method for efficient ground moving target indication (GMTI) with a two-channel synthetic aperture radar (SAR) to any multichannel SAR/GMTI radars independent of the number of receive channels. This processing method called extended DPCA (EDPCA) is derived in this paper and is applied to data acquired with the Canadian RADARSAT-2 satellite. The expected GMTI performance of RADARSAT-2 after EDPCA processing is compared to results achieved with measured RADARSAT-2 data recorded during several trials in order to validate the developed theory. © 2012 IEEE.
Ender J.H.G.,Fraunhofer Institute for High Frequency Physics and Radar Techniques
Signal Processing | Year: 2010
Compressive sensing (CS) techniques offer a framework for the detection and allocation of sparse signals with a reduced number of samples. Today, modern radar systems operate with high bandwidths-demanding high sample rates according to the Shannon-Nyquist theorem-and a huge number of single elements for phased array antennas. Often only a small amount of target parameters is the final output, arising the question, if CS could not be a good mean to reduce data size, complexity, weight, power consumption and costs of radar systems. There is only a small number of publications addressing the application of CS to radar, leaving several open questions. This paper addresses some aspects as a further step to CS-radar by presenting generic system architectures and implementation considerations. It is not the aim of this paper to investigate numerically efficient algorithms but to point to promising applications as well as arising problems. Three possible applications are considered: pulse compression, radar imaging, and air space surveillance with array antennas. Some simulation results are presented and enriched by the evaluation of real data acquired by an experimental radar system of Fraunhofer FHR. © 2009 Elsevier B.V. All rights reserved.
Rosebrock J.,Fraunhofer Institute for High Frequency Physics and Radar Techniques
IEEE Transactions on Geoscience and Remote Sensing | Year: 2011
The Tracking and Imaging Radar (TIRA) System is used to determine orbits and to generate Inverse Synthetic Aperture Radar (ISAR) images of satellites. In certain situations such as damage analysis, a satellite to be observed is not stabilized, and its attitude is not known in advance. In order to assess the situation, the intrinsic motion is important to know. In this paper, several algorithms are derived and tested which estimate a constant rotational motion from 3-D features obtained from 2-D ISAR images by a modified stereo vision algorithm. © 2011 IEEE.
Ender J.,Fraunhofer Institute for High Frequency Physics and Radar Techniques
Proceedings International Radar Symposium | Year: 2013
In this paper we give a brief review of compressive sensing (CS) applied to radar. Though CS theory has been introduced only a few years ago (in 2006, see e. g. ), it today manifests as a kind of revolution in signal processing and sensor systems. We will discuss some properties of CS radar and present a few examples. It is also a concern of the author to point to some limitations and shortcomings if CS is 'blindly' applied with great enthusiasm to any radar problem. The time has come to deliberate when CS will be an advantage and when 'oldfashioned' methods should better be applied. © 2013 German Inst of Navigation.
Ribalta A.,Fraunhofer Institute for High Frequency Physics and Radar Techniques
IEEE Geoscience and Remote Sensing Letters | Year: 2011
In this letter, we develop time-domain reconstruction algorithms for frequency-modulated continuous wave synthetic aperture radar (FMCW-SAR). The algorithms considered here are the time-domain correlation algorithm, and two versions of the backprojection algorithm: the standard one based on the start-stop approximation, and a modified version that takes into account the movement of the sensor during the transmission of the pulse. Numerical simulations illustrate the performance of the algorithms, showing that the start-stop approximation may not be valid for FMCW-SAR, whereas the modified backprojection algorithm works very well here. © 2010 IEEE.
Kuschel H.,Fraunhofer Institute for High Frequency Physics and Radar Techniques
2013 International Conference on Radar - Beyond Orthodoxy: New Paradigms in Radar, RADAR 2013 | Year: 2013
The history of passive radar dates back to the early days of radar in 1935 when the Daventry experiment was conducted in the UK. It continues in WW II with the German Klein Heidelberg passive radar and receives new interest today, as passive covert radar (PCR) systems like Silent Sentry, Homeland Alerter 100, Aulos and PARADE are ready for operation. The future of PCR will strongly depend on the availability of transmitters of opportunity such as FM-radio and digital broadcast networks. © 2013 IEEE.
Cristallini D.,Fraunhofer Institute for High Frequency Physics and Radar Techniques |
Burger W.,Fraunhofer Institute for High Frequency Physics and Radar Techniques
IEEE Transactions on Signal Processing | Year: 2012
In this paper, a novel approach for direct data domain space time adaptive processing (STAP) is presented. As already described in past literature, direct data domain STAP (also known as deterministic STAP) has several advantages compared to traditional stochastic STAP. In particular, being implicitly a single snapshot interference cancellation technique, deterministic STAP generally outperforms stochastic STAP in fast varying interference scenarios. On the other hand, in its classical derivation, target detection performances of deterministic STAP are severely deteriorated in case of uncertainty in the knowledge of exact target parameters as direction of arrival (DOA) and Doppler frequency. To overcome this problem, we propose a robust implementation of deterministic STAP in order to take into account a possible mismatch between the nominal and the actual target parameters. The proposed approach reformulates the deterministic STAP problem in the context of convex problem optimization. A detailed analysis of the maximum acceptable target parameters error is conducted, which ensures the existence of a numerical solution for the convex problem optimization. The proposed robust deterministic approach is defined for both the one dimensional (spatial-only) and the two dimensional (space-time) case. The effectiveness of the proposed approach is shown both in simulated scenarios and by direct application to real data taken from the experimental multichannel radar system PAMIR developed at Fraunhofer FHR. © 2011 IEEE.