Periodic-Filtering Method for Low-SNR Vibration Radar Signal

Radar is a non-contact, high-precision vibration measurement device and an important tool for bridge vibration monitoring. Vibration information needs to be extracted from the radar phase, but the radar phase information is sensitive to noise. Under low signal-to-noise ratio (SNR) data acquisition c...

Full description

Bibliographic Details
Main Authors: Yun Lin, Linghan Zhang, Hongwei Han, Yang Li, Wenjie Shen, Yanping Wang
Format: Article
Language:English
Published: MDPI AG 2023-07-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/15/14/3461
_version_ 1797587676453404672
author Yun Lin
Linghan Zhang
Hongwei Han
Yang Li
Wenjie Shen
Yanping Wang
author_facet Yun Lin
Linghan Zhang
Hongwei Han
Yang Li
Wenjie Shen
Yanping Wang
author_sort Yun Lin
collection DOAJ
description Radar is a non-contact, high-precision vibration measurement device and an important tool for bridge vibration monitoring. Vibration information needs to be extracted from the radar phase, but the radar phase information is sensitive to noise. Under low signal-to-noise ratio (SNR) data acquisition conditions, such as low radar transmission power or a long observation distance, differential phase jump errors occur and clutter estimation becomes difficult, which leads to inaccurate inversion of vibration deformation. Traditional low-pass filtering methods can filter out noise to improve SNR, but they require oversampling. The sampling rate needs to be several times higher than the Doppler bandwidth, which is several times higher than the vibration frequency. This puts high data acquisition requirements on radar systems and causes large data volumes. Therefore, this paper proposes a novel vibration signal filtering method called the periodic filtering method. The method uses the periodicity feature of vibration signals for filtering without oversampling. This paper derives the time-domain and frequency-domain expressions for the periodic filter and presents a deformation inversion process based on them. The process involves extracting the vibration frequency in the Doppler domain, suppressing noise through periodic filtering, estimating clutter using circle fitting on the data complex plane, and inverting final deformation with differential phase. The method is verified through simulation experiments, calibration experiments, and bridge vibration experiments. The results show that the new periodic filtering method can improve the SNR by five times, resolve differential phase jumps, and accurately estimate clutter, thus getting submillimeter-level vibration deformation at low SNR.
first_indexed 2024-03-11T00:42:16Z
format Article
id doaj.art-ab4a6d0166a648efa27857a81c604c22
institution Directory Open Access Journal
issn 2072-4292
language English
last_indexed 2024-03-11T00:42:16Z
publishDate 2023-07-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj.art-ab4a6d0166a648efa27857a81c604c222023-11-18T21:11:07ZengMDPI AGRemote Sensing2072-42922023-07-011514346110.3390/rs15143461Periodic-Filtering Method for Low-SNR Vibration Radar SignalYun Lin0Linghan Zhang1Hongwei Han2Yang Li3Wenjie Shen4Yanping Wang5Radar Monitoring Technology Laboratory, School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaRadar Monitoring Technology Laboratory, School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaRadar Monitoring Technology Laboratory, School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaRadar Monitoring Technology Laboratory, School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaRadar Monitoring Technology Laboratory, School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaRadar Monitoring Technology Laboratory, School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaRadar is a non-contact, high-precision vibration measurement device and an important tool for bridge vibration monitoring. Vibration information needs to be extracted from the radar phase, but the radar phase information is sensitive to noise. Under low signal-to-noise ratio (SNR) data acquisition conditions, such as low radar transmission power or a long observation distance, differential phase jump errors occur and clutter estimation becomes difficult, which leads to inaccurate inversion of vibration deformation. Traditional low-pass filtering methods can filter out noise to improve SNR, but they require oversampling. The sampling rate needs to be several times higher than the Doppler bandwidth, which is several times higher than the vibration frequency. This puts high data acquisition requirements on radar systems and causes large data volumes. Therefore, this paper proposes a novel vibration signal filtering method called the periodic filtering method. The method uses the periodicity feature of vibration signals for filtering without oversampling. This paper derives the time-domain and frequency-domain expressions for the periodic filter and presents a deformation inversion process based on them. The process involves extracting the vibration frequency in the Doppler domain, suppressing noise through periodic filtering, estimating clutter using circle fitting on the data complex plane, and inverting final deformation with differential phase. The method is verified through simulation experiments, calibration experiments, and bridge vibration experiments. The results show that the new periodic filtering method can improve the SNR by five times, resolve differential phase jumps, and accurately estimate clutter, thus getting submillimeter-level vibration deformation at low SNR.https://www.mdpi.com/2072-4292/15/14/3461radarvibrationfilteringclutter suppression
spellingShingle Yun Lin
Linghan Zhang
Hongwei Han
Yang Li
Wenjie Shen
Yanping Wang
Periodic-Filtering Method for Low-SNR Vibration Radar Signal
Remote Sensing
radar
vibration
filtering
clutter suppression
title Periodic-Filtering Method for Low-SNR Vibration Radar Signal
title_full Periodic-Filtering Method for Low-SNR Vibration Radar Signal
title_fullStr Periodic-Filtering Method for Low-SNR Vibration Radar Signal
title_full_unstemmed Periodic-Filtering Method for Low-SNR Vibration Radar Signal
title_short Periodic-Filtering Method for Low-SNR Vibration Radar Signal
title_sort periodic filtering method for low snr vibration radar signal
topic radar
vibration
filtering
clutter suppression
url https://www.mdpi.com/2072-4292/15/14/3461
work_keys_str_mv AT yunlin periodicfilteringmethodforlowsnrvibrationradarsignal
AT linghanzhang periodicfilteringmethodforlowsnrvibrationradarsignal
AT hongweihan periodicfilteringmethodforlowsnrvibrationradarsignal
AT yangli periodicfilteringmethodforlowsnrvibrationradarsignal
AT wenjieshen periodicfilteringmethodforlowsnrvibrationradarsignal
AT yanpingwang periodicfilteringmethodforlowsnrvibrationradarsignal