Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Clutter Identification Based on Kernel Density Estimation and Sparse Recovery

Wang, Haokun (2018) Clutter Identification Based on Kernel Density Estimation and Sparse Recovery. Master's Thesis, University of Pittsburgh. (Unpublished)

[img] PDF
Restricted to University of Pittsburgh users only until 11 June 2023.

Download (509kB) | Request a Copy

Abstract

Many existing radar algorithms are developed under the hypothesis that the environment (clutter) is stationary. However, in real applications, the statistical characteristics of the clutter might change immensely in space, time, or both, depending on the radar-operational scenarios. If unaccounted for, these non-stationarities may extremely hamper the radar performance. Therefore, to overcome such performance degradations, we have developed a cognitive radar framework to dynamically detect changes in the clutter characteristics, and to adapt to these changes by identifying the new clutter distribution. In this work, we present a sparse-recovery based clutter identification technique. In this technique, we build a dictionary matrix of well-known clutter statistics such that each column of the matrix is a kernel density estimation of a specific clutter distribution. When radar measurements arrive, sparse recovery, more specifically, orthogonal matching pursuit (OMP) algorithm is used to identify the distribution of the radar measurements by matching the kernel density estimation of the measurements to one of the columns of the dictionary matrix. We analyze the effect of different kernels and distance measures between the kernel density estimations on the clutter identification accuracy. With numerical examples, we demonstrate that the sparse-recovery based method provides high accuracy in clutter identification and this technique is robust to changes in the training and test sample sizes.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Wang, Haokunhaw86@pitt.eduhaw860000-0001-9764-0896
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairAkcakaya, Muratakcakaya@pitt.edu
Committee MemberMao, Zhi-Hongzhm4@pitt.edu
Committee MemberMiskov-Zivanov, Natasanmzivanov@pitt.edu
Thesis AdvisorAkcakaya, Muratakcakaya@pitt.edu
Date: 11 June 2018
Date Type: Publication
Defense Date: 5 April 2018
Approval Date: 11 June 2018
Submission Date: 5 April 2018
Access Restriction: 5 year -- Restrict access to University of Pittsburgh for a period of 5 years.
Number of Pages: 42
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Electrical and Computer Engineering
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: Clutter identification, Sparse recovery, KED, OMP, Ozturk
Date Deposited: 11 Jun 2018 18:02
Last Modified: 11 Jun 2018 18:02
URI: http://d-scholarship.pitt.edu/id/eprint/34125

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item