Mark T. Frankford

A Study of the Shapiro–Wilk Test for the Detection of Pulsed Sinusoidal Radio Frequency Interference

The performance of the Shapiro-Wilk (S-W) test of normality for the detection of pulsed sinusoidal radio frequency interference (RFI) in microwave radiometry is analyzed. The study is motivated by the fact that the S-W test has been shown in the statistical literature to be effective in detecting a wide variety of non-Gaussian signal types. The basic properties of the S-W test statistic are reviewed, and the implementation of the test in digital hardware is discussed. Because the properties of the test statistic are difficult to obtain analytically, Monte Carlo simulations are utilized to compute performance. Results show that the test can provide reasonable performance in detecting pulsed sinusoidal RFI and that quantization of data has only a minimal impact on the sensitivity achieved. Detection performance is also compared with that of the kurtosis test for normality. It is shown that the S-W test produces comparable but degraded sensitivity compared to that of the kurtosis test in most cases while avoiding the ldquoblind spotrdquo associated with the kurtosis test for pulsed interferers having 50% duty cycle. Test performance is also shown to be improved if a priori knowledge of expected RFI pulse lengths is incorporated.

Software defined radar studies of human motion signatures

The detection and monitoring of human motion with radar has numerous applications in surveillance, urban military operations, search-and-rescue, and other areas. Recent studies have shown that movements of humans generate unique micro-Doppler signatures that can be exploited to classify human motions. This motivates an improved understanding of human Doppler signatures. Numerous simulations and measurements of human “dismount” signatures has been performed in the past, but most have been focused on a single radar center-frequency and have not taken polarization effects into consideration. In this paper, human modeling and motion measurements using multiple radar frequencies are proposed to explore the impact of the radar frequency on human range/Doppler signatures. Furthermore, ground effects on human targets are investigated using a four path model. The OSU Software defined radar (SDR) system, which can be tuned from 2GHz to 18 GHz with 500MHz bandwidth, was used for the measurements. This radar can operate at two frequencies simultaneously, allowing for dual frequency human measurements. Also, different polarizations are considered to understand human Doppler signatures. Modeling efforts are based on a finite dielectric cylinder approximation, so that the human body is modeled as a collection of dielectric cylinders. Scattering signatures are computed neglecting scattering interactions among these cylinders.

Software-defined radar for MIMO and adaptive waveform applications

The development of a software-defined radar testbed is described. The testbed is to be used to explore advanced techniques such as multiple-input multiple-output radar and adaptive waveforms. The system features a fully programmable, dual-channel, arbitrary pulsed waveform generator with a quadrature downconverting receiver. The system can generate waveforms of up to 500 MHz instantaneous bandwidth at a center frequency tunable from 2–18 GHz. The RF front end features two independent transmit and receive channels that can be multiplexed between four dual-polarized transmit and four dual-polarized receive antennas.

Numerical and experimental studies of target detection with MIMO radar

Spatially diverse multiple-input multiple-output (MIMO) radar systems combine multistatic measurements of the target under view into a single detection algorithm and are thereby expected to alleviate the effects of fading on target radar cross sections (RCS) as the angle of observation is varied. Previous analytical studies of target detection for this case have shown that MIMO radar detection performance can exceed that of the corresponding phased array radar if both sufficient spatial diversity and signal-to-noise ratio (SNR) are achieved. These results have been based on a statistical model for the multistatic RCS of the target that is similar to the traditional Swerling models of the monostatic RCS. The degree to which these results are applicable to specific target geometries therefore remains uncertain. To address this issue, two studies of MIMO radar target detection incorporating realistic RCS properties for specific target geometries were performed. The first study utilized a numerical method to compute the multistatic RCS of a helicopter-like target observed at center frequency 200 MHz, while the second involved radar measurements of an unmanned aerial vehicle (UAV) target at 2.75 and 4.5 GHz. MIMO radar configurations having two transmitters and either three (for the radar measurements) or four (numerical simulations) receivers were used. In both cases, multistatic received fields were combined with regulated thermal noise levels in postprocessing to study target detection performance. Because in general the azimuthal orientation of a specific target with respect to the radar is uncertain, the detection performance results shown are averaged over the azimuthal orientation angle of the target. The average over target orientation can also be interpreted as similar to an average over “trials” of a statistical target description, enabling comparisons of field properties averaged over target orientation with similar ensemble averages from the statistical models of the literature. Although detection performance curves for the specific targets considered are not identical to those predicted analytically by the statistical target model, results for these targets confirm that the MIMO radar system can achieve enhanced detection performance as compared with the corresponding phased array radar system.

Including Spatial Correlations in the Statistical MIMO Radar Target Model

Previous studies of statistical MIMO radar detection performance have used a target model that consists of a large number of point scatterers located within a rectangular target area. These point scatterers have scattering amplitudes that are complex random variables and are spatially uncorrelated, so that the target is a white noise process in space. Spatial correlations are introduced into the target model in this paper, and the impact of these correlations on MIMO radar system detection performance is analyzed.

Simulation and analysis of polarimetric radar signatures of human gaits

Radar observations of human activities have a variety of applications in security, defense, and rescue operations. Range-Doppler signatures of human motions are a useful tool for retrieving information on observed activities but require an understanding of the scattering processes involved to enable interpretation. This paper presents a study of human Doppler signatures using simulations, in particular focusing on the impact of the polarization to enable an understanding of any advantages in the use of polarimetric radar. The simulation model utilized is based on an approximate scattering approach combined with a 12-cylinder description of the human body. A comparison with single polarization co-pol measurements is used to show that the model provides reasonable first-order predictions of human signatures. Further simulations for polarimetric signatures illustrate the differing contributions of individual body parts to micro-Doppler returns and suggest that multi-polarization measurements can be useful in future micro-Doppler radar systems for human observation.

Airborne L-band RFI observations in the smapvex08 campaign with the L-band interference suppressing radiometer

Radio Frequency Interference (RFI) is a major concern for microwave radiometry. In September-October 2008, an airborne campaign for observing L-band RFI was conducted and included approximately fifty hours of flight time. The campaign included test flights in Grand Junction, Colorado, observations over soil moisture ground truth sites in Iowa and Maryland, transit and return flights from Colorado to Maryland, and dedicated RFI observing missions over urban areas. In this paper, the L-band digital backend and its RFI detection and mitigation algorithms are described. RFI statistics and examples from the airborne campaign are also presented.

On the Shapiro-Wilk Test for the Detection of Pulsed Sinusoidal Radio Frequency Interference

The Shapiro-Wilk (S-W) test is a test of normality recommended in the statistical literature for its effectiveness in detecting a wide-variety of non-gaussian signals. Tests for normality can be used for the detection of Radio Frequency Interference (RFI). However, so far only the kurtosis test has been used for this purpose. The performance of the S-W test against pulsed sinusoidal sources is studied in this paper as an alternative method of RFI detection. An overview of the S-W test is provided, and implementation of the test in digital hardware is discussed. Performance of the test was analyzed using Monte Carlo simulations due to the difficulties in obtaining an analytical solution. Results show that the test can provide reasonable performance in detecting pulsed sinu-soidal RFI, and the effect of quantization is minimal in the sensitivity obtained.

Compensation of Faraday Rotation in Multipolarization Scatterometry

Spaceborne remote sensors operating at L-band and lower frequencies can significantly be affected by Faraday rotation (FR) as their signals pass through the Earths ionosphere. A method of compensating for FR in multipolarization scatterometry is introduced, which utilizes an ancillary estimate of the FR angle to retrieve corrected polarized scattered powers. Simulation results are presented to demonstrate the behavior of the FR correction process when radar speckle, instrument errors, and errors in the ancillary FR angle estimate are introduced.

Compensation of Faraday Rotation in Multi-Polarization Scatterometry

Spaceborne remote sensors operating at L-band and lower frequencies can significantly be affected by Faraday rotation (FR) as their signals pass through the Earths ionosphere. A method of compensating for FR in multipolarization scatterometry is introduced, which utilizes an ancillary estimate of the FR angle to retrieve corrected polarized scattered powers. Simulation results are presented to demonstrate the behavior of the FR correction process when radar speckle, instrument errors, and errors in the ancillary FR angle estimate are introduced.