Investigation of high-frequency propagation channels through pipes and ducts for building interior reconnaissance

Repository

Investigation of high-frequency propagation channels through pipes and ducts for building interior reconnaissance

Show simple record

dc.contributor.advisor Ling, Hao
dc.creator Whitelonis, Nicholas John, 1984-
dc.date.accessioned 2012-07-12T15:05:37Z
dc.date.available 2012-07-12T15:05:37Z
dc.date.created 2012-05
dc.date.issued 2012-07-12
dc.date.submitted May 2012
dc.identifier.uri http://hdl.handle.net/2152/ETD-UT-2012-05-5651
dc.description.abstract Recently, there is strong interest in the through-wall sensing capabilities of radar for use in law enforcement, search and rescue, and urban military operations. Due to the high attenuation of walls, through-wall radar typically operates in the low GHz frequency region, where resolution is limited. It is worthwhile to explore other means of propagating radar waves into and back out of a building’s interior for sensing applications. One possibility is through duct-like structures that are commonly found in a building, such as metal pipes used for plumbing or air conditioning ducts. The objective of this dissertation is to investigate techniques to acquire radar images of targets through a pipe. First, using the pipe as an electromagnetic propagation channel is studied. A modal approach previously developed for computing the radar cross-section of a circular duct is modified to compute the transmission through a pipe. This modal approach for transmission is validated against measured data. It is also shown that a pipe is a high-pass propagation channel. The modal analysis is then extended to two-way, through-pipe propagation for backscattering analysis. The backscattering from a target is observed through a pipe in simulation and measurement. Next, methods to form two-dimensional radar images from backscattering data collected through a pipe are explored. Four different methods previously developed for free-space imaging are applied to the problem of imaging through a pipe: beamforming, matched filter processing, MUSIC, and compressed sensing. In all four methods it is necessary to take into account the propagation through the pipe in order to properly generate a focused radar image. Each method is demonstrated using simulation and validated against measurement data. The beamforming and matched filter methods are found to suffer from poor cross-range resolution. To improve resolution, the MUSIC algorithm is applied and shown to give superior resolution at the expense of more complicated data collection. The final method, compressed sensing, is shown to achieve good cross-range resolution with simpler data collection. A comparison of the tradeoffs between the four methods is summarized and discussed. Two additional extensions are studied. First, a method for computing the transmission through an arbitrary pipe network using the generalized scattering matrix approach is proposed and implemented. Second, a new method for computing joint time-frequency distributions based on compressed sensing is applied to analyze the backscattering phenomenology from a pipe.
dc.format.mimetype application/pdf
dc.language.iso eng
dc.subject Electromagnetics
dc.subject Waveguides
dc.subject Radar
dc.subject Through wall radar
dc.title Investigation of high-frequency propagation channels through pipes and ducts for building interior reconnaissance
dc.date.updated 2012-07-12T15:05:51Z
dc.identifier.slug 2152/ETD-UT-2012-05-5651
dc.contributor.committeeMember Neikirk, Dean
dc.contributor.committeeMember Wilson, Preston
dc.contributor.committeeMember Powers, Edward
dc.contributor.committeeMember Alu, Andrea
dc.description.department Electrical and Computer Engineering
dc.type.genre thesis
dc.type.material text
thesis.degree.department Electrical and Computer Engineering
thesis.degree.discipline Electrical and Computer Engineering
thesis.degree.grantor University of Texas at Austin
thesis.degree.level Doctoral
thesis.degree.name Doctor of Philosophy

Files in this work

Size: 4.243Mb
Format: application/pdf

This work appears in the following Collection(s)

Show simple record


Advanced Search

Browse

My Account

Statistics

Information