Microwave Radiometer Calibration Using Deep Learning With Reduced Reference Information and 2-D Spectral Features

The accuracy of geophysical retrievals from radiometers relies on calibration quality, encompassing both absolute radiometric accuracy and spectral consistency. Radiometers have employed various calibration techniques, including external targets, vicarious sources, and internal calibrators like nois...

Full description

Bibliographic Details
Main Authors: Ahmed Manavi Alam, Mehmet Kurum, Mehmet Ogut, Ali C. Gurbuz
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10318952/
_version_ 1827593011142328320
author Ahmed Manavi Alam
Mehmet Kurum
Mehmet Ogut
Ali C. Gurbuz
author_facet Ahmed Manavi Alam
Mehmet Kurum
Mehmet Ogut
Ali C. Gurbuz
author_sort Ahmed Manavi Alam
collection DOAJ
description The accuracy of geophysical retrievals from radiometers relies on calibration quality, encompassing both absolute radiometric accuracy and spectral consistency. Radiometers have employed various calibration techniques, including external targets, vicarious sources, and internal calibrators like noise diodes or matched reference loads. Calibration techniques face challenges like frequency dependence, instrumental effects, environmental influences, drift, aging, and radio frequency interference. Recent hardware advancements enable radiometers to collect raw samples containing both temporal and spectral information. Leveraging advanced modeling techniques like deep learning (DL) enables detecting subtle correlations, non-linear dependencies, and higher-order interactions within the data extracting valuable information that may have been challenging with conventional methods. This study utilizes NASA's Soil Moisture Active Passive (SMAP) satellite's level 1A and level 1B data products to develop a DL-based radiometer calibrator to estimate antenna temperature. Spectrograms of second raw moments equivalent to power carrying the 2-D spectral features serve as primary input in a supervised convolutional neural network-based architecture. DL-based calibrator has demonstrated high correlation and low root mean square error when incorporating spectral information from both reference and noise diodes and when not considering this information. Findings suggest that the ancillary features such as internal thermistor temperature and loss elements exhibit sufficient accuracy in estimating antenna temperature to compensate for variations in receiver noise temperature and short-term gain fluctuations in the absence of the reference load and noise diode power. The proposed calibration technique with reduced reference information might enable radiometers for a higher number of antenna scene observations within a footprint.
first_indexed 2024-03-09T02:03:59Z
format Article
id doaj.art-a1571414c43e4d6ba320b7ff8cba9a4c
institution Directory Open Access Journal
issn 2151-1535
language English
last_indexed 2024-03-09T02:03:59Z
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj.art-a1571414c43e4d6ba320b7ff8cba9a4c2023-12-08T00:02:03ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-011774876510.1109/JSTARS.2023.333326810318952Microwave Radiometer Calibration Using Deep Learning With Reduced Reference Information and 2-D Spectral FeaturesAhmed Manavi Alam0https://orcid.org/0000-0003-2022-9761Mehmet Kurum1https://orcid.org/0000-0002-5750-9014Mehmet Ogut2https://orcid.org/0000-0002-7142-6899Ali C. Gurbuz3https://orcid.org/0000-0001-8923-0299Department of Electrical and Computer Engineering, and Information Processing and Sensing Lab, Mississippi State University, Mississippi State, MS, USASchool of Electrical and Computer Engineering, University of Georgia, Athens, GA, USAJet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USADepartment of Electrical and Computer Engineering, and Information Processing and Sensing Lab, Mississippi State University, Mississippi State, MS, USAThe accuracy of geophysical retrievals from radiometers relies on calibration quality, encompassing both absolute radiometric accuracy and spectral consistency. Radiometers have employed various calibration techniques, including external targets, vicarious sources, and internal calibrators like noise diodes or matched reference loads. Calibration techniques face challenges like frequency dependence, instrumental effects, environmental influences, drift, aging, and radio frequency interference. Recent hardware advancements enable radiometers to collect raw samples containing both temporal and spectral information. Leveraging advanced modeling techniques like deep learning (DL) enables detecting subtle correlations, non-linear dependencies, and higher-order interactions within the data extracting valuable information that may have been challenging with conventional methods. This study utilizes NASA's Soil Moisture Active Passive (SMAP) satellite's level 1A and level 1B data products to develop a DL-based radiometer calibrator to estimate antenna temperature. Spectrograms of second raw moments equivalent to power carrying the 2-D spectral features serve as primary input in a supervised convolutional neural network-based architecture. DL-based calibrator has demonstrated high correlation and low root mean square error when incorporating spectral information from both reference and noise diodes and when not considering this information. Findings suggest that the ancillary features such as internal thermistor temperature and loss elements exhibit sufficient accuracy in estimating antenna temperature to compensate for variations in receiver noise temperature and short-term gain fluctuations in the absence of the reference load and noise diode power. The proposed calibration technique with reduced reference information might enable radiometers for a higher number of antenna scene observations within a footprint.https://ieeexplore.ieee.org/document/10318952/Calibrationdeep learning (DL)machine learningmicrowave radiometryneural networkradio frequency interference (RFI)
spellingShingle Ahmed Manavi Alam
Mehmet Kurum
Mehmet Ogut
Ali C. Gurbuz
Microwave Radiometer Calibration Using Deep Learning With Reduced Reference Information and 2-D Spectral Features
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Calibration
deep learning (DL)
machine learning
microwave radiometry
neural network
radio frequency interference (RFI)
title Microwave Radiometer Calibration Using Deep Learning With Reduced Reference Information and 2-D Spectral Features
title_full Microwave Radiometer Calibration Using Deep Learning With Reduced Reference Information and 2-D Spectral Features
title_fullStr Microwave Radiometer Calibration Using Deep Learning With Reduced Reference Information and 2-D Spectral Features
title_full_unstemmed Microwave Radiometer Calibration Using Deep Learning With Reduced Reference Information and 2-D Spectral Features
title_short Microwave Radiometer Calibration Using Deep Learning With Reduced Reference Information and 2-D Spectral Features
title_sort microwave radiometer calibration using deep learning with reduced reference information and 2 d spectral features
topic Calibration
deep learning (DL)
machine learning
microwave radiometry
neural network
radio frequency interference (RFI)
url https://ieeexplore.ieee.org/document/10318952/
work_keys_str_mv AT ahmedmanavialam microwaveradiometercalibrationusingdeeplearningwithreducedreferenceinformationand2dspectralfeatures
AT mehmetkurum microwaveradiometercalibrationusingdeeplearningwithreducedreferenceinformationand2dspectralfeatures
AT mehmetogut microwaveradiometercalibrationusingdeeplearningwithreducedreferenceinformationand2dspectralfeatures
AT alicgurbuz microwaveradiometercalibrationusingdeeplearningwithreducedreferenceinformationand2dspectralfeatures