Application of Discrete Wavelet Transform to Broadband Interference Rejection

• Main Authors: Hsa, Chuan-Chien, 徐傳健
• Other Authors: Chen, Yuan-Hwang
• Format: Others
• Language: zh-TW
• Published: 1997
• Online Access: http://ndltd.ncl.edu.tw/handle/48304854734020399593

碩士 === 國立中山大學 === 電機工程研究所 === 85 === In this paper the multiresolution analysis of the discrete wavelet transfom (DWT) is applied to an adaptive filtering structure for broadband interference rejection. The M-dimensional time-domain input vector is mapped to N wavelet levels (N<M) with individual frequency band and time resolution by DWT. The adaptive processor is then implemented in these wavelet levels domain. The performance of the DWT domain-based adaptive normalized LMS (least mean square) filter is investigated by the help of learning curves and power spectral density of system output for narrowband and broadband interference. It is shown for broadband interference that the faster convergence rate and higher signal to noise ratio are achieved for the DWT-based NLMS filter as compared to those in DFT (discrete Fourier transform) domain and time domain, respectively. In addition, the use of the DWT-based adaptive signal processing further reduce the number of adaptive weights with comparable result. This advantage gives us the expe-ctation that the DWT-based adaptive filter outperform well in broadband interference rejection with adaptive weights limited. Finally, this technique is applied to adaptive beamfonning for array sidelobe canceling.