Denoising complex background radar signals based on wavelet decomposition thresholding
The echo signals of the radar in complex backgrounds are often very unstable and thus require effective noise cancellation. In this paper, according to the characteristics of continuous wavelet variation and discrete wavelet variation, the decomposition effect of multi-resolution analysis and orthog...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Sciendo
2024-01-01
|
| Series: | Applied Mathematics and Nonlinear Sciences |
| Subjects: | |
| Online Access: | https://doi.org/10.2478/amns.2023.2.00535 |
| _version_ | 1797340744672870400 |
|---|---|
| author | Qiu Feng Yuan Kee |
| author_facet | Qiu Feng Yuan Kee |
| author_sort | Qiu Feng |
| collection | DOAJ |
| description | The echo signals of the radar in complex backgrounds are often very unstable and thus require effective noise cancellation. In this paper, according to the characteristics of continuous wavelet variation and discrete wavelet variation, the decomposition effect of multi-resolution analysis and orthogonal Mallat algorithm on low-frequency and high-frequency non-smooth signals is studied, and the selection method of wavelet bases is explored. Then, the noise characteristics affecting the pulsed LIDAR system are analyzed, and the LIDAR pulse signal is simulated by MATLAB, while Gaussian white noise is introduced to obtain the noise-added echo signal, and then multiple wavelet threshold denoising methods are applied to denoise the echo signal. For the input signal-to-noise ratio of −10.57 dB, the output signal-to-noise ratios of db8, db9, db10, and bior3.5 wavelet bases under forced thresholding are −1.971, −2.178, −2.173, and −1.032, respectively. For different input signal-to-noise ratios, the average root mean square error of db8, db9, db10, and bior3.5 wavelet bases under default thresholding is 1.51. The denoising methods for radar signals using the properties of wavelet decomposition have obvious superiority compared to traditional filters, and the wavelet transforms threshold denoising methods have wide adaptability. |
| first_indexed | 2024-03-08T10:07:45Z |
| format | Article |
| id | doaj.art-6d05c80eae7f45919f86bd7bc2e44ffe |
| institution | Directory Open Access Journal |
| issn | 2444-8656 |
| language | English |
| last_indexed | 2024-03-08T10:07:45Z |
| publishDate | 2024-01-01 |
| publisher | Sciendo |
| record_format | Article |
| series | Applied Mathematics and Nonlinear Sciences |
| spelling | doaj.art-6d05c80eae7f45919f86bd7bc2e44ffe2024-01-29T08:52:33ZengSciendoApplied Mathematics and Nonlinear Sciences2444-86562024-01-019110.2478/amns.2023.2.00535Denoising complex background radar signals based on wavelet decomposition thresholdingQiu Feng0Yuan Kee11Key Laboratory of Atmospheric Optics, Anhui Institute of Optics and Fine Mechanics, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, Anhui, 230031, China.1Key Laboratory of Atmospheric Optics, Anhui Institute of Optics and Fine Mechanics, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, Anhui, 230031, China.The echo signals of the radar in complex backgrounds are often very unstable and thus require effective noise cancellation. In this paper, according to the characteristics of continuous wavelet variation and discrete wavelet variation, the decomposition effect of multi-resolution analysis and orthogonal Mallat algorithm on low-frequency and high-frequency non-smooth signals is studied, and the selection method of wavelet bases is explored. Then, the noise characteristics affecting the pulsed LIDAR system are analyzed, and the LIDAR pulse signal is simulated by MATLAB, while Gaussian white noise is introduced to obtain the noise-added echo signal, and then multiple wavelet threshold denoising methods are applied to denoise the echo signal. For the input signal-to-noise ratio of −10.57 dB, the output signal-to-noise ratios of db8, db9, db10, and bior3.5 wavelet bases under forced thresholding are −1.971, −2.178, −2.173, and −1.032, respectively. For different input signal-to-noise ratios, the average root mean square error of db8, db9, db10, and bior3.5 wavelet bases under default thresholding is 1.51. The denoising methods for radar signals using the properties of wavelet decomposition have obvious superiority compared to traditional filters, and the wavelet transforms threshold denoising methods have wide adaptability.https://doi.org/10.2478/amns.2023.2.00535wavelet decompositionthreshold denoisingradar echo signalsignal-to-noise ratiogaussian white noise68t05 |
| spellingShingle | Qiu Feng Yuan Kee Denoising complex background radar signals based on wavelet decomposition thresholding Applied Mathematics and Nonlinear Sciences wavelet decomposition threshold denoising radar echo signal signal-to-noise ratio gaussian white noise 68t05 |
| title | Denoising complex background radar signals based on wavelet decomposition thresholding |
| title_full | Denoising complex background radar signals based on wavelet decomposition thresholding |
| title_fullStr | Denoising complex background radar signals based on wavelet decomposition thresholding |
| title_full_unstemmed | Denoising complex background radar signals based on wavelet decomposition thresholding |
| title_short | Denoising complex background radar signals based on wavelet decomposition thresholding |
| title_sort | denoising complex background radar signals based on wavelet decomposition thresholding |
| topic | wavelet decomposition threshold denoising radar echo signal signal-to-noise ratio gaussian white noise 68t05 |
| url | https://doi.org/10.2478/amns.2023.2.00535 |
| work_keys_str_mv | AT qiufeng denoisingcomplexbackgroundradarsignalsbasedonwaveletdecompositionthresholding AT yuankee denoisingcomplexbackgroundradarsignalsbasedonwaveletdecompositionthresholding |