Linear Frequency Modulated Reverberation Suppression Using Non-Negative Matrix Factorization Methods, Dechirping Transformation and Modulo Operation
| Autor: | Seokjin Lee, Geunhwan Kim, Kyun-Kyung Lee |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
Reverberation
Modulo operation General Computer Science Computer science General Engineering Active sonar Sonar reverberation suppression sparse constraints Non-negative matrix factorization Pulse (physics) Matrix decomposition non-negative matrix factorization dechirping transformation Transformation (function) General Materials Science False alarm lcsh:Electrical engineering. Electronics. Nuclear engineering modulo operation Algorithm lcsh:TK1-9971 |
| Zdroj: | IEEE Access, Vol 8, Pp 110720-110737 (2020) |
| ISSN: | 2169-3536 |
| Popis: | In this study, we propose a reverberation suppression algorithm for linear frequency-modulated (LFM) pulse sonar systems using a non-negative matrix factorization (NMF) method. Because conventional NMF-based reverberation suppression algorithms are only applicable to continuous wave reverberation, we propose two pre-processing methods, namely dechirping transformation and modulo operation, to facilitate application of the NMF method to LFM reverberation. Moreover, we impose additional sparse constraints on the NMF method to improve its performance. To evaluate the proposed algorithm, an experiment involving simulated LFM reverberation is performed. The results thereof show improved detection performance at several signal-to-reverberation ratios and false alarm conditions. Moreover, the proposed algorithm is also applied to sea experiment data. According to the sea experiment analysis, the algorithm is able to suppress the LFM reverberation effectively and improve detection performance in practical LFM pulse sonar systems. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|