In this paper we propose a new approach for applying Compressed Sensing to Stepped Frequency Continuous Wave SAR measurements. The proposed technique allows the sensor to decide autonomously, while scanning, on the number of samples needed, while assuring a reconstruction quality chosen by the operator. With the online reconstruction algorithm sampling rates far below the bound fixed by the Nyquist-Shannon theorem are achieved. Moreover, to improve further on minimizing the frequency sampling rates, the measurements obtained from previous sensor positions are added as online weighted side information into the reconstruction algorithm. The applicability and excellent performance of the approach is illustrated by a series of experiments on sparsified Through-the-Wall Imaging radar data.
Becquaert, M, Cristofani, E, Vandewal, M, Stiens, J & Deligiannis, N 2018, Online Sequential Compressed Sensing with Weighted Multiple Side Information for Through the Wall Imaging. in URSI Benelux Forum 2018. URSI, pp. 1-5, URSI Benelux Forum 2018, Delft, Netherlands, 25/01/18.
Becquaert, M., Cristofani, E., Vandewal, M., Stiens, J., & Deligiannis, N. (2018). Online Sequential Compressed Sensing with Weighted Multiple Side Information for Through the Wall Imaging. In URSI Benelux Forum 2018 (pp. 1-5). URSI.
@inproceedings{fafa4e54fe2848d09700702d6fee0e67,
title = "Online Sequential Compressed Sensing with Weighted Multiple Side Information for Through the Wall Imaging",
abstract = "In this paper we propose a new approach for applying Compressed Sensing to Stepped Frequency Continuous Wave SAR measurements. The proposed technique allows the sensor to decide autonomously, while scanning, on the number of samples needed, while assuring a reconstruction quality chosen by the operator. With the online reconstruction algorithm sampling rates far below the bound fixed by the Nyquist-Shannon theorem are achieved. Moreover, to improve further on minimizing the frequency sampling rates, the measurements obtained from previous sensor positions are added as online weighted side information into the reconstruction algorithm. The applicability and excellent performance of the approach is illustrated by a series of experiments on sparsified Through-the-Wall Imaging radar data.",
author = "Mathias Becquaert and Edison Cristofani and Marijke Vandewal and Johan Stiens and Nikolaos Deligiannis",
year = "2018",
month = jan,
day = "25",
language = "English",
pages = "1--5",
booktitle = "URSI Benelux Forum 2018",
publisher = "URSI",
note = "URSI Benelux Forum 2018 ; Conference date: 25-01-2018 Through 25-01-2018",
}