Robust Multiple-Measurement Sparsity-Aware STAP with Bayesian Variational Autoencoder
Due to the shortage of independent and identically distributed (i.i.d.) training samples, space−time adaptive processing (STAP) often suffers remarkable performance degradation in the heterogeneous clutter environment. Sparse recovery (SR) techniques have been introduced into STAP for the benefit of...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/15/3800 |
| _version_ | 1827601718344417280 |
|---|---|
| author | Chenxi Zhang Huiliang Zhao Wenchao Chen Bo Chen Penghui Wang Changrui Jia Hongwei Liu |
| author_facet | Chenxi Zhang Huiliang Zhao Wenchao Chen Bo Chen Penghui Wang Changrui Jia Hongwei Liu |
| author_sort | Chenxi Zhang |
| collection | DOAJ |
| description | Due to the shortage of independent and identically distributed (i.i.d.) training samples, space−time adaptive processing (STAP) often suffers remarkable performance degradation in the heterogeneous clutter environment. Sparse recovery (SR) techniques have been introduced into STAP for the benefit of the drastically reduced training requirement, but they are incompletely robust for involving the tricky selection of hyper−parameters or the undesirable point estimation for parameters. Given this issue, we incorporate the <b>M</b>ultiple−measurement <b>C</b>omplex−valued <b>V</b>ariational relevance vector machines (MCV) to model the space−time echoes and provide a Gibbs−sampling−based method to estimate posterior distributions of parameters accurately. However, the Gibbs sampler require quantities of iterations, as unattractive as traditional Bayesian type SR−STAP algorithms when the real−time processing is desired. To address this problem, we further develop the <b>B</b>ayesian <b>A</b>utoencoding MCV for STAP (BAMCV−STAP), which builds the generative model according to MCV and approximates posterior distributions of parameters with an inference network pre−trained off−line, to realize fast reconstruction of measurements. Experimental results on simulated and measured data demonstrate that BAMCV−STAP can achieve suboptimal clutter suppression in terms of the output signal to interference plus noise ratio (SINR) loss, as well as the attractive real−time processing property in terms of the convergence rate and computational loads. |
| first_indexed | 2024-03-09T05:02:26Z |
| format | Article |
| id | doaj.art-c4b63c77f3634699bacdde42ee9d3496 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T05:02:26Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-c4b63c77f3634699bacdde42ee9d34962023-12-03T12:59:06ZengMDPI AGRemote Sensing2072-42922022-08-011415380010.3390/rs14153800Robust Multiple-Measurement Sparsity-Aware STAP with Bayesian Variational AutoencoderChenxi Zhang0Huiliang Zhao1Wenchao Chen2Bo Chen3Penghui Wang4Changrui Jia5Hongwei Liu6National Lab of Radar Signal Processing, Xidian University, Xi’an 710071, ChinaNational Lab of Radar Signal Processing, Xidian University, Xi’an 710071, ChinaNational Lab of Radar Signal Processing, Xidian University, Xi’an 710071, ChinaNational Lab of Radar Signal Processing, Xidian University, Xi’an 710071, ChinaNational Lab of Radar Signal Processing, Xidian University, Xi’an 710071, ChinaThe 38th Research Institute of China Electronics Technology Corporation, Hefei 230088, ChinaNational Lab of Radar Signal Processing, Xidian University, Xi’an 710071, ChinaDue to the shortage of independent and identically distributed (i.i.d.) training samples, space−time adaptive processing (STAP) often suffers remarkable performance degradation in the heterogeneous clutter environment. Sparse recovery (SR) techniques have been introduced into STAP for the benefit of the drastically reduced training requirement, but they are incompletely robust for involving the tricky selection of hyper−parameters or the undesirable point estimation for parameters. Given this issue, we incorporate the <b>M</b>ultiple−measurement <b>C</b>omplex−valued <b>V</b>ariational relevance vector machines (MCV) to model the space−time echoes and provide a Gibbs−sampling−based method to estimate posterior distributions of parameters accurately. However, the Gibbs sampler require quantities of iterations, as unattractive as traditional Bayesian type SR−STAP algorithms when the real−time processing is desired. To address this problem, we further develop the <b>B</b>ayesian <b>A</b>utoencoding MCV for STAP (BAMCV−STAP), which builds the generative model according to MCV and approximates posterior distributions of parameters with an inference network pre−trained off−line, to realize fast reconstruction of measurements. Experimental results on simulated and measured data demonstrate that BAMCV−STAP can achieve suboptimal clutter suppression in terms of the output signal to interference plus noise ratio (SINR) loss, as well as the attractive real−time processing property in terms of the convergence rate and computational loads.https://www.mdpi.com/2072-4292/14/15/3800clutter suppressionvariational autoencoder (VAE)space−time adaptive processing (STAP)clutter plus noise covariance matrix (CCM)sparse recovery (SR) |
| spellingShingle | Chenxi Zhang Huiliang Zhao Wenchao Chen Bo Chen Penghui Wang Changrui Jia Hongwei Liu Robust Multiple-Measurement Sparsity-Aware STAP with Bayesian Variational Autoencoder Remote Sensing clutter suppression variational autoencoder (VAE) space−time adaptive processing (STAP) clutter plus noise covariance matrix (CCM) sparse recovery (SR) |
| title | Robust Multiple-Measurement Sparsity-Aware STAP with Bayesian Variational Autoencoder |
| title_full | Robust Multiple-Measurement Sparsity-Aware STAP with Bayesian Variational Autoencoder |
| title_fullStr | Robust Multiple-Measurement Sparsity-Aware STAP with Bayesian Variational Autoencoder |
| title_full_unstemmed | Robust Multiple-Measurement Sparsity-Aware STAP with Bayesian Variational Autoencoder |
| title_short | Robust Multiple-Measurement Sparsity-Aware STAP with Bayesian Variational Autoencoder |
| title_sort | robust multiple measurement sparsity aware stap with bayesian variational autoencoder |
| topic | clutter suppression variational autoencoder (VAE) space−time adaptive processing (STAP) clutter plus noise covariance matrix (CCM) sparse recovery (SR) |
| url | https://www.mdpi.com/2072-4292/14/15/3800 |
| work_keys_str_mv | AT chenxizhang robustmultiplemeasurementsparsityawarestapwithbayesianvariationalautoencoder AT huiliangzhao robustmultiplemeasurementsparsityawarestapwithbayesianvariationalautoencoder AT wenchaochen robustmultiplemeasurementsparsityawarestapwithbayesianvariationalautoencoder AT bochen robustmultiplemeasurementsparsityawarestapwithbayesianvariationalautoencoder AT penghuiwang robustmultiplemeasurementsparsityawarestapwithbayesianvariationalautoencoder AT changruijia robustmultiplemeasurementsparsityawarestapwithbayesianvariationalautoencoder AT hongweiliu robustmultiplemeasurementsparsityawarestapwithbayesianvariationalautoencoder |