Title :
Optimum multi-target detection using an ANC neuro-fuzzy scheme and wideband replica correlator
Author :
Tseng, Chien-Hsun ; Cole, Marina
Author_Institution :
Sch. of Eng., Univ. of Warwick, Coventry
Abstract :
An optimum target detection algorithm is developed for underwater active wideband sonar signal in the presence of reverberation and ambient noise. The hybrid algorithm makes use of an adaptive ANC neuro-fuzzy scheme in the first instance to provide an effective fine tuned signal followed by an iterative cross correlation-based target motion estimation (TME) scheme for efficient target detection. Computer simulations based on real input data sets demonstrate capability and efficiency of the proposed hybrid algorithm in extracting even zero-Doppler target signals from a highly reverberated noisy environment.
Keywords :
fuzzy neural nets; geophysics computing; iterative methods; motion estimation; object detection; reverberation; sonar signal processing; sonar tracking; adaptive ANC neuro-fuzzy scheme; ambient noise; even zero-Doppler target signals extracting; highly reverberated noisy environment; hybrid algorithm; iterative cross correlation-based; optimum multitarget detection; reverberation; target motion estimation; underwater active wideband sonar signal; wideband replica correlator; Active noise reduction; Computer simulation; Correlators; Data mining; Iterative algorithms; Motion estimation; Object detection; Reverberation; Sonar detection; Wideband; ANC; ANFIS; echolocation; optimal wideband replica correlator; reverberation; wavelets;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2009. ICASSP 2009. IEEE International Conference on
Conference_Location :
Taipei
Print_ISBN :
978-1-4244-2353-8
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2009.4959847
