Ingolf Willms

On polarization diversity gain in short range UWB-Radar object imaging

In this paper a 2D bi-static, fully-polarimetric Ultrawideband (UWB) imaging system is presented. The exploitation of polarization diversity provides supplemental information compared to mono-polarized sensing techniques. This polarization diversity gain enhances the efficiency of target classification according to information contained in the back-scattered signal. Hence, additional characteristics of targets such as form, details of surface structure and orientation are gathered, which may remain invisible for mono-polarized systems. An experimental validation is carried out based on complex test objects, an M-sequence Radar device and new dual-orthogonal polarized antennas.

UWB material characterisation and object recognition with applications in fire and security

By means of UWB radar sensors the tasks of material characterisation and object recognition are performed on the basis of a previous imaging of the whole environment. A UWB version of the microwave ellipsometry method is applied for estimating the permittivity of homogenous objects. The object recognition task is performed using bistatic sensor nodes on the basis of Radar measurements. The simulation-based performance evaluations show a very robust behavior due to suitable pre-processing of Radar data. The applications comprise the detection of fire sources, the detection of metallic object hidden under clothing and the recognition of building structures.

Millimeter Wave Propagation through Dust

It is well known that radar sensors are capable to propagate through smoke and dust with only very little attenuation. The higher the operating frequency, the smaller is the necessary antenna diameter for a required geometrical resolution on the ground. Consequently millimeter waves would be the choice for this type of sensor system. For an optimum system design the question of atmospheric attenuation at different bands within the millimeter wave region due to losses by dust and sand has to be answered. As only little data exist on this topic, respective measurements were done under reproducible laboratory conditions. Different approaches were used to cover a broad range of sand and dust types. The contribution describes the experimental set-ups and gives typical results for calibrated samples of sand and dust, which were derived from the lab measurements. A perspective is given for further investigations upon this topic.

A mobile security robot equipped with UWB-radar for super-resolution indoor positioning and localisation applications

In this paper an autonomous mobile security robot named CoLORbot is introduced which operates with UWB-Radar technology. The objective is the 2D inspection of an unknown indoor area with randomly distributed unknown solitary objects. Environmental imaging by means of electro-magnetic wave propagation in the mm range is performed. Different sensor configurations (mono-static vs. bi-static) and different antenna designs (omni-directional pattern vs. narrow beam directive pattern) are evaluated and analysed. The CoLORbot which is fully equipped with professional motion units, UWB-Radar devices, antennas and a laptop for data processing is introduced and evaluated. Strategies to perform this kind of multi disciplinary purpose containing challenges in robotics, UWB-Radar sensing, imaging algorithms and antenna handling are revised and in every case experimentally validated.

Performance enhancement of UWB material characterization and object recognition for security robots

By means of UWB Radar sensors the tasks of material characterisation and object recognition can be performed on the basis of a previous imaging of the whole environment. A UWB version of the microwave ellipsometry method is applied for estimating the permittivity of homogenous objects. The object recognition task is performed using bistatic sensor nodes on the basis of Radar measurements. The simulation-based performance evaluations show a very robust behavior due to suitable preprocessing of Radar data. The applications comprise the detection of fire sources, the detection of metallic object hidden under clothing, and the recognition of building structures.

Super-resolution feature extraction imaging algorithm for complex objects

For the goal of an Object Recognition (OR) Radar system, a feature extraction algorithm is proposed in this paper. Conventional radar imaging method based on Kirchhoff Migration and the revised range point migration method are known to obtain fast and accurate images. However, these methods are not suitable for feature extraction as the Kirchhoff Migration method processes the whole evolution of the Radar data and the revised range point migration method extracts the coordinates of the target contour not features. Furthermore, the new proposed feature extraction algorithm is designed for a circular scanning trajectory, or a rotating target and a bi-static antenna configuration. The proposed algorithm calculates the target points from the a priori extracted wavefronts of the Object Under Test (OUT). A Polarimetric Dynamic Correlation Method (PDCM) is employed in the proposed algorithm for the extraction of the wavefronts. Experimental validations are performed with two complex OUT, a M-sequence Radar device (4.5 GHz-13.5 GHz) and compact dual-polarized Ultra-Wideband antennas.

3D imaging of a manmade target with weak scattering centres by means of UWB-radar

In this paper a 3D bi-static, fully polarimetric Ultra-Wideband (UWB) imaging system which satisfies superresolution conditions is presented. The objective is to inspect a target with weak scattering centres and to prove the performance of recent imaging methods under these conditions. This issue is handicapped further by spanning a suboptimal synthetic array, i.e. the main beam of the antennas is not aligned with the orientation of the weak scatterers. An opened laptop with a knife fixed at the back of the display is used as the target for an experimental validation. The radar cross section of the keyboard in this position can be assumed to be very small which reveals very weak multibounce scattering mechanisms. The evaluation of the imaging is performed by the real-time capable revised range point migration (RRPM) and the conventional Kirchhoff migration. The experimental validation is carried out with a pair of two tapered slot line Vivaldi antennas both integrated in a conical shaped teflon rod and an M-sequence Radar device with 100% fractional bandwidth at a carrier of 9 GHz.

UWB-radar based surface permittivity estimation in hostile and pathless security scenarios

Ultra-wideband radar analysis provides a high range resolution, which is not only important for ranging, but also for the material analysis of objects. This paper introduces a novel approach for a nondestructive and in-situ estimation of material characteristics. The developed technique was inspired by the already well established optical ellipsometry and is best suited for the remote inspection of hostile indoor scenarios with robots. The paper gives an overview about the technique and its current status of research and focuses on the realisation using only few mechanical components.

Exploitation of polarimetry in short range 3D UWB-radar object imaging

This paper deals with a 3D bi-static, fully-polarimetric Ultrawideband (UWB) imaging system for short range applications. The exploitation of polarization diversity provides valuable supplemental information compared to mono-polarized sensing techniques. This polarization diversity gain is utilized for a quad-polarized UWB-Radar system for the imaging of a complex shaped 3D Object with corners as well as edges. For a full perspective of the object the sensors provide circumnavigation to extract entirely all stereoscopic distributed polarimetric scattering centres. The experimental validation is carried out with a pair of two tapered slot line Vivaldi antennas both integrated in a conical shaped teflon rod and an M-sequence Radar device.

Robust UWB Radar object recognition

This paper focuses on an emergency situation in which a network of ultra-wide-band (UWB) sensor nodes mounted on moveable platforms is moving in a room for purposes of recognition of objects. The recognition is accomplished between 2D canonical objects, which are scanned by UWB radar and a set of reference objects with same size which were analyzed by a ray-tracing method. The object recognition task is performed on the basis of RCS (radar cross section) measurements leading to corresponding radar-gram data. Here the objects characteristic geometric features are examined by using polar Fourier descriptors which are extended in a way that also provides for multiple reflections.