skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


  • NSF Public Access
  • Search Results
  • Detection and Mitigation of Ground Clutter in Polarimetric Phased Array Radar Measurements Using Machine Learning and Physics-Based Discriminants
Title: Detection and Mitigation of Ground Clutter in Polarimetric Phased Array Radar Measurements Using Machine Learning and Physics-Based Discriminants
This paper presents clutter detection and mitigation for polarimetric phased array weather radar measurements using machine learning. The following three approaches are analyzed for clutter detection in the cylindrical polarimetric phased array radar measurements, including naive Bayes classifier (NBC), multilayer perceptron (MLP), and convolutional neural network (CNN). Results show that CNN achieves the best performance in clutter detection, followed by MLP and NBC. This is because CNN utilizes spatial information of the input images, which has different features for clutter from that for weather. It is also shown that the combination of physics-based discriminants of power ratio and raw radar measurements is more effective in clutter detection than the direct use of raw radar measurements. In addition, CNN is employed for clutter mitigation and its performance is compared with the traditional speckle filter technique. It is demonstrated that CNN outperforms the speckle filter and incorporation of power ratio in the training process could further improve CNN’s performance in clutter mitigation.  more » « less
Award ID(s):
2136161
PAR ID:
10552866
Author(s) / Creator(s):
; ;
Publisher / Repository:
IEEE
Date Published:
Journal Name:
IEEE Transactions on Geoscience and Remote Sensing
Volume:
61
ISSN:
0196-2892
Page Range / eLocation ID:
1 to 10
Subject(s) / Keyword(s):
Clutter detection and mitigation, machine learning, multilayer perceptron, convolutional neural network, polarimetric phased array radar.
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; (, 2023 IEEE Radar Conference (RadarConf23))
    Reconfigurable intelligent surface (RIS) technology is a promising approach being considered for future wireless communications due to its ability to control signal propagation with low-cost elements. This paper explores the use of an RIS for clutter mitigation and target detection in radar systems. Unlike conventional reflect-only RIS, which can only adjust the phase of the reflected signal, or active RIS, which can also amplify the reflected signal at the cost of significantly higher complexity, noise, and power consumption, we exploit hybrid RIS that can configure both the phase and modulus of the impinging signal by absorbing part of the signal energy. Such RIS can be considered as a compromise solution between conventional reflect-only and active RIS in terms of complexity, power consumption, and degrees of freedoms (DoFs). We consider two clutter suppression scenarios: with and without knowledge of the target range cell. The RIS design is formulated by minimizing the received clutter echo energy when there is no information regarding the potential target range cell. This turns out to be a convex problem and can be efficiently solved. On the other hand, when target range cell information is available, we maximize the received signal-to-noise-plus-interference ratio (SINR). The resulting non-convex optimization problem is solved through fractional programming algorithms. Numerical results are presented to demonstrate the performance of the proposed hybrid RIS in comparison with conventional RIS in clutter suppression for target detection. 
    more » « less
  2. ; ; (, IEEE Transactions on Geoscience and Remote Sensing)
    Accurate estimation of surface precipitation with high spatial and temporal resolution is critical for decision making regarding severe weather and water resources management. Polarimetric weather radar is the main operational instrument used for quantitative precipitation estimation (QPE). However, conventional parametric radar QPE algorithms such as the radar reflectivity (Z) and rain rate (R) relations cannot fully represent clouds and precipitation dynamics due to their dependency on local raindrop size distributions and the inherent parameterization errors. This article develops four deep learning (DL) models for polarimetric radar QPE (i.e., RQPENetD1, RQPENetD2, RQPENetV, RQPENetR) using different core building blocks. In particular, multi-dimensional polarimetric radar observations are utilized as input and surface gauge measurements are used as training labels. The feasibility and performance of these DL models are demonstrated and quantified using U.S. Weather Surveillance Radar - 1988 Doppler (WSR-88D) observations near Melbourne, Florida. The experimental results show that the dense blocks-based models (i.e., RQPENetD1 and RQPENetD2) have better performance than residual blocks, RepVGG blocks-based models (i.e., RQPENetR and RQPENetV) and five traditional Z-R relations. RQPENetD1 has the best quantitative performance scores, with a mean absolute error (MAE) of 1.58 mm, root mean squared error (RMSE) of 2.68 mm, normalized standard error (NSE) of 26%, and correlation of 0.92 for hourly rainfall estimates using independent rain gauge data as references. These results suggest that deep learning performs well in mapping the connection between polarimetric radar observations aloft and surface rainfall. 
    more » « less
  3. ; (, IEEE)
    A novel sensing approach for the single radio frequency (RF) chain millimeter wave systems is proposed. The sensing strategy is inspired from synthetic aperture radar systems, and synthesizes a virtual array manifold vector using a single phased array over temporal measurements. The geometry of the virtual array can be controlled enabling synthesis of both a virtual uniform linear array (ULA) and a virtual sparse linear array starting from a large physical array. Moreover, the proposed sensing can be realized using conventional phased array/analog combiner. Several design parameters of the sensing scheme provide flexibility and are discussed including their impact on initial alignment. The proposed sensing approach allows for a rich set of options for inference. Candidate detection and estimation algorithms are presented and their performance is evaluated. 
    more » « less
  4. ; ; ; ; (, Monthly Weather Review)
    Abstract Properties of 7488 thunderstorms are summarized for June–September 2022 during the Tracking Aerosol Convection Interactions Experiment (TRACER) field campaign Houston, Texas, using polarimetric weather radar and VHF 3D Lightning Mapping Array data. Automated tracking of storms linked each instrument’s measurements to a data-defined, time-evolving storm footprint. Within each storm, the depth and magnitude of episodic columns of radar differential reflectivity and specific differential phase quantified the prevalence of updrafts that activated mixed-phase precipitation pathways. Lightning measurements further distinguished the degree of rimed precipitation formation: the fraction of tracks with lightning varied from day to day and cells with lightning had stronger polarimetric columns. Track-level correlation of the lightning flash rate with radar polarimetric measures had substantial spread, showing that lightning provides an additional signal of mixed-phase precipitation processes that can complement future studies of thermodynamic and aerosol controls on cloud microphysics in the Houston region. 
    more » « less
  5. ; ; (, Journal of Atmospheric and Oceanic Technology)
    Abstract The intrinsic uncertainty of radar-based retrievals in snow originates from a large diversity of snow growth habits, densities, and particle size distributions, all of which can make interpreting radar measurements of snow very challenging. The application of polarimetric radar for snow measurements can mitigate some of these issues. In this study, a novel polarimetric method for quantification of the extinction coefficient and visibility in snow, based on the joint use of radar reflectivity at horizontal polarization Z and specific differential phase K DP , is introduced. A large 2D-video-disdrometer snow dataset from central Oklahoma is used to derive a polarimetric bivariate power-law relation for the extinction coefficient, . The relation is derived for particle aspect ratios ranging from 0.5 to 0.8 and the width of the canting angle distribution ranging from 0° to 40°, values typical of aggregated snow, and validated via theoretical and analytical derivations/simulations. The multiplier of the relation is sensitive to variations in particles’ densities, the width of the canting angle distribution, and particles’ aspect ratios, whereas the relation’s exponents are practically invariant to changes in the latter two parameters. This novel approach is applied to polarimetric S-band WSR-88D data and verified against previous studies and in situ measurements of the extinction coefficient for four snow events in the eastern United States. The polarimetric radar estimates of the extinction coefficient exhibit smaller biases in comparison to previous studies concerning the ground measurements. The results indicate that there is good potential for reliable radar estimates of visibility from polarimetric weather radars, a parameter inversely proportional to the extinction coefficient. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email