本文選題：線性預(yù)測(cè)編碼　切入點(diǎn)：支持向量機(jī)　出處：《艦船科學(xué)技術(shù)》2016年02期

【摘要】：針對(duì)利用船舶輻射噪聲進(jìn)行水下目標(biāo)識(shí)別的問題進(jìn)行研究,提出一種基于線性預(yù)測(cè)編碼(LPC)倒譜系數(shù)和支持向量機(jī)(SVM)的船舶目標(biāo)識(shí)別方法。該方法通過對(duì)捕獲到船舶輻射噪聲進(jìn)行LPC倒譜分析,實(shí)現(xiàn)各信號(hào)分量及信道的分離,以提取其LPC倒譜參數(shù)。再采用支持向量機(jī)技術(shù)處理多類水下目標(biāo)的非線性、小樣本的識(shí)別分類。最后,利用仿真得到的幾種水下目標(biāo)輻射噪聲進(jìn)行本文算法試驗(yàn),證明本文算法是有效的,并取得較高的識(shí)別準(zhǔn)確率。
[Abstract]:Aiming at the problem of underwater target recognition using ship radiated noise, a ship target recognition method based on linear predictive coding (LPC) cepstrum number and support vector machine (SVM) is proposed.By analyzing the LPC cepstrum of the captured ship noise, the signal components and the channel are separated to extract the LPC cepstrum parameters.Then support vector machine (SVM) is used to deal with the nonlinear and small sample recognition and classification of multiple underwater targets.Finally, several underwater target radiated noises are simulated and tested in this paper. The results show that the proposed algorithm is effective and has a high recognition accuracy.
【作者單位】： 紹興職業(yè)技術(shù)學(xué)院信息工程學(xué)院;
【分類號(hào)】：U675.7;U661.44
，

本文編號(hào)：1687134