Dave Tahmoush

Radar micro-doppler for long range front-view gait recognition

We seek to understand the extraction of radar micro-Doppler signals generated by human motions at long range and with a front-view to use them as a biometric. We describe micro-Doppler algorithms used for the detection and tracking, and detail the gait features that can be extracted. We have measurements of multiple human subjects in outdoor but low-clutter backgrounds for identification and find that at long range and front-view, the probability of correct classification can be over 80%. However, the micro-Doppler signals are dependent on the direction of motion, and we discuss methods to reduce the effect of the direction of motion. These radar biometric features can serve as identifying features in a scene with multiple subjects. Ground truth using video and GPS is used to validate the radar data.

Angle, elevation, PRF, and illumination in radar microDoppler for security applications

We discovered several important characteristics that should be considered in the development of radar systems for the measurement of human micro-Doppler. First, radar which illuminates parts of the human body separately is capable of distinguishing more structure in the micro-Doppler because of the decreased interference from other body parts. Second, a PRF of even as low as 400 Hz is sufficient for measuring micro-Doppler, though higher is recommended. Third, the direction of motion of the target can be critical, but a higher elevation angle can be used to avoid the difficulties experienced when the motion of the targets is perpendicular to the beam of the radar.

Radar microDoppler for security applications: Modeling men versus women

We created a model of human motion that was built upon extracted gait data, and calculated the spectrogram that would result from the radar measurements. We verified the model with several measurements. We modeled the radar Doppler signature of the component parts of the human body that build up the spectrogram, and verified the components. We then modeled the spectrograms of men versus women and compared these with some early measurements. The results show that this approach could be promising since small though measurable differences are apparent, though a larger study is necessary in order to adequately explore the issue.

Stride rate in radar micro-doppler images

We extract gait information from the radar micro-Doppler signals generated by human motion. We demonstrate the extraction of information associated with gait, especially the stride rate, from simulated and measured radar data. We describe micro-Doppler algorithms used for the extraction of stride rate, the radar sensors used for the measurements, and detail the gait features that can be extracted. We make measurements of human subjects in realistic outdoor clutter backgrounds. These features help identify subjects in a scene. We gather ground truth using video to validate the radar data. We conclude that although we can extract gait features like stride rate from radar data, more features need to be extracted before reliable identification or classification can be determined.

Micro-Doppler phenomenology of humans at UHF and Ku-band for biometric characterization

Extracting biometric characteristics using radar requires a detailed understanding of the RF scattering phenomenology associated with humans. The gross translational Doppler signals associated with walking are well documented in the literature. The work reported in this paper seeks to understand the micro-Doppler signals generated by human motion associated with ancillary activities such as breathing, heartbeat, and speech. We will describe procedures for anechoic chamber and outdoor measurements at UHF and Ku-band of humans engaged in a range of activities, such as lying, sitting, standing, speaking, and walking. In addition, we will analyze and discuss the various biometric signatures that we collected.

Radar Stride Rate Extraction

We seek to understand the extraction of radar signals generated by human motion and extract information from them. We demonstrate the extraction of information associated with gait, especially the stride rate, from radar data. We describe algorithms used for detection, the radar sensors used for the measurements, and detail some of the features that can be extracted. We make measurements of human subjects in outdoor clutter backgrounds for identification. These features can serve as identifying information in a scene with multiple subjects. We gather ground truth using video and accelerometers to validate the radar data.

Simplified model of dismount microDoppler and RCS

Our goal is to be able to detect and classify dismounts, but we were lacking a quick way to estimate dismount parameters, especially with respect to angle of motion and depression angle of the radar. Micro-Doppler models have been developed which attempt to predict the human micro-Doppler response, and here we present a simplified model to quickly estimate dismount RCS and some micro-Doppler characteristics across a range of angles of motion. This model was extracted from measured radar data. We focus on modeling and measuring the characteristics of human walking parameters to determine response of dismounts to radar signals. We determine a simple closed form for RCS as a function of angle for walking dismounts as well as several rules of thumb, and we also determine a closed form for front-view micro-Doppler.

Time-integrated range-Doppler maps for visualizing and classifying radar data

A time-integrated range-Doppler map shows the micro-Doppler characteristics of targets in radar images that enable an operator to classify different target types and to classify different activities being done by the targets. A time-integrated range-Doppler map is a compilation of range-Doppler maps over time that results in a spectrogram-like characterization of Doppler while maintaining the range information as well. These are compiled from the range-Doppler maps by taking the maximum value for each pixel over a time range. The time resolution is overlapped onto the range resolution, which is in effect a rotation of the traditional spectrogram which compresses range. This type of radar imaging also allows multiple subjects to be viewed simultaneously and avoids tracking issues in spectrogram creation.

Micro-range micro-doppler for dismount classification

This paper presents a processing technique that can be used to detect and classify pedestrians group based on the micro- Doppler signature gathered with a millimeter wave radar. The evaluation of the number of pedestrians moving in a group can be a difficult task using a traditional micro-Doppler spectrogram because of a tendency for people to partially synchronize their steps when walking together. The new approach, based on multi-range variation as well as the micro-Doppler variations, provides promising results. The range-spectrogram processing technique was developed and tested using a database composed of hundreds of pedestrian and vehicle signatures gathered in an urban test site over a two year period in a variety of weather conditions. We associate image detections with radar detections through motion extracted from both radar and imagery. We also explain how radar and video together can produce an inexpensive alternative to 3-D imaging.

Radar surveillance in urban environments

Radar surveillance in difficult environments like urban areas can be challenging due to large amounts of both multipath and clutter. Additionally, buildings and clutter like parked vehicles can produce shadowed areas where the line-of-sight is broken. We analyzed urban materials to determine how to utilize multipath to see into the shadows of urban environments, which polarization has the least loss, and which frequencies performed the best across a range of environments. Urban canyons were analyzed to determine whether there was more of a multipath effect on the measurements or a waveguide effect. The detection and tracking of non-line-of-sight moving objects using multiple bounces was attempted across a variety of urban building materials with an urban radar surveillance system. We demonstrate the detection and tracking of subjects using multipath returns, but there was no disambiguation of real or multipath sources. We also discuss the challenges of classification in an urban environment.