發(fā)布時(shí)間：2019-05-06 20:12

【摘要】：傳統(tǒng)的水聲信號(hào)分類都是直接使用原信號(hào)進(jìn)行處理的,特征提取耗時(shí)長(zhǎng),數(shù)據(jù)量大,針對(duì)這兩個(gè)缺點(diǎn),提出了一種壓縮感知理論中基于稀疏表示的水聲信號(hào)特征提取方法;該方法利用了水聲信號(hào)在DCT變換域的稀疏特性,將信號(hào)的稀疏表示作為目標(biāo)特征,并采用SVM分類算法進(jìn)行分類識(shí)別。仿真結(jié)果表明,該方法不僅減少了特征向量的計(jì)算時(shí)間,還提高了目標(biāo)分類識(shí)別率,還降低了水聲信號(hào)的傳輸數(shù)據(jù)量,壓縮率可達(dá)96%,在實(shí)際工程應(yīng)用中具有較高的實(shí)用價(jià)值。
[Abstract]:The traditional classification of underwater acoustic signal is directly processed by the original signal, and the feature extraction is time-consuming and large amount of data. In view of these two shortcomings, a feature extraction method of underwater acoustic signal based on sparse representation in compression sensing theory is proposed. In this method, the sparse representation of underwater acoustic signal in DCT transform domain is used as the target feature, and the SVM classification algorithm is used to classify and recognize the underwater acoustic signal. The simulation results show that this method not only reduces the computing time of eigenvector, but also improves the recognition rate of target classification, and also reduces the amount of data transmitted by underwater acoustic signals, and the compression ratio can reach 96%. It has high practical value in practical engineering application.
【作者單位】： 西北工業(yè)大學(xué)航海學(xué)院;
【分類號(hào)】：TN911.7

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 劉曉山;付國(guó)蘭;;基于脊波變換的圖像壓縮[J];電腦與信息技術(shù);2007年02期

2 劉曉山;付國(guó)蘭;;基于脊波變換和SPIHT算法相結(jié)合的圖像壓縮[J];江西師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2007年06期

3 王華丹;劉海林;;稀疏盲源分離問(wèn)題的恢復(fù)性研究[J];廣東工業(yè)大學(xué)學(xué)報(bào);2008年02期

4 談華f，

本文編號(hào)：2470458