本文關(guān)鍵詞：麥克風(fēng)網(wǎng)絡(luò)中基于分布式粒子濾波的說(shuō)話(huà)人跟蹤方法研究　出處：《大連理工大學(xué)》2016年博士論文　論文類(lèi)型：學(xué)位論文

  更多相關(guān)文章： 分布式麥克風(fēng)網(wǎng)絡(luò) 說(shuō)話(huà)人跟蹤 分布式粒子濾波 一致性算法 非線(xiàn)性非高斯系統(tǒng)

【摘要】：利用麥克風(fēng)陣列對(duì)室內(nèi)說(shuō)話(huà)人進(jìn)行跟蹤,是根據(jù)分布于空間中的多個(gè)麥克風(fēng)接收的音頻信號(hào)對(duì)運(yùn)動(dòng)說(shuō)話(huà)人的位置信息進(jìn)行估計(jì)�；邴溈孙L(fēng)陣列的說(shuō)話(huà)人跟蹤技術(shù)在公共安全監(jiān)控、音視頻會(huì)議系統(tǒng)、語(yǔ)音識(shí)別、車(chē)載電話(huà)以及機(jī)器人等領(lǐng)域都有廣泛應(yīng)用。近年來(lái),隨著無(wú)線(xiàn)傳感器網(wǎng)絡(luò)、網(wǎng)絡(luò)通信、移動(dòng)計(jì)算以及集成電路技術(shù)的快速發(fā)展,生產(chǎn)小尺寸麥克風(fēng)的成本越來(lái)越低,嵌入式處理器的計(jì)算能力顯著增強(qiáng),分布式麥克風(fēng)網(wǎng)絡(luò)逐漸發(fā)展起來(lái),基于分布式麥克風(fēng)網(wǎng)絡(luò)的聲源定位和跟蹤成為語(yǔ)音處理領(lǐng)域一個(gè)新的研究熱點(diǎn)。然而,目前大多數(shù)聲源跟蹤算法主要針對(duì)傳統(tǒng)的規(guī)則麥克風(fēng)陣列,并不能直接用于分布式麥克風(fēng)網(wǎng)絡(luò)。貝葉斯濾波是解決室內(nèi)說(shuō)話(huà)人跟蹤問(wèn)題的典型方法,它采用狀態(tài)空間的方法對(duì)說(shuō)話(huà)人跟蹤問(wèn)題進(jìn)行建模。當(dāng)狀態(tài)空間模型為線(xiàn)性、高斯時(shí),卡爾曼濾波是貝葉斯濾波的最優(yōu)解；當(dāng)狀態(tài)空間模型為非線(xiàn)性、非高斯時(shí),粒子濾波是貝葉斯濾波的有效近似解。在室內(nèi)噪聲和混響條件下,說(shuō)話(huà)人狀態(tài)的后驗(yàn)分布多為非高斯分布,其觀測(cè)模型通常為非線(xiàn)性模型。據(jù)此,本文在貝葉斯濾波理論框架下,對(duì)現(xiàn)有的分布式粒子濾波算法進(jìn)行改進(jìn),提高了濾波器的跟蹤精度和魯棒性；通過(guò)深入研究粒子濾波理論,提出了一種新的分布式粒子濾波器。在此基礎(chǔ)上,將所提出的分布式粒子濾波算法應(yīng)用于麥克風(fēng)網(wǎng)絡(luò)進(jìn)行室內(nèi)說(shuō)話(huà)人跟蹤,提出了一些針對(duì)性的改進(jìn)措施。本論文的主要?jiǎng)?chuàng)新工作如下：(1)在現(xiàn)有的基于粒子權(quán)重一致性的分布式粒子濾波器中,其似然函數(shù)的計(jì)算要求各個(gè)節(jié)點(diǎn)的觀測(cè)在給定狀態(tài)的條件下相互獨(dú)立,且需要已知觀測(cè)噪聲的統(tǒng)計(jì)信息。針對(duì)該問(wèn)題,本文利用廣域相干場(chǎng)函數(shù)在某一空間位置的取值反應(yīng)了聲源在該位置處的可能性大小的特點(diǎn)構(gòu)建了一種偽似然函數(shù),進(jìn)而推導(dǎo)了一種廣域相干場(chǎng)-分布式粒子濾波器,并將其用于麥克風(fēng)網(wǎng)絡(luò)中的說(shuō)話(huà)人跟蹤問(wèn)題。該方法不要求各個(gè)節(jié)點(diǎn)的觀測(cè)條件獨(dú)立,也無(wú)需已知觀測(cè)噪聲的統(tǒng)計(jì)信息,且易于分布式計(jì)算。仿真和實(shí)際實(shí)驗(yàn)結(jié)果表明,所提出的方法在噪聲和混響環(huán)境中具有良好的跟蹤性能。(2)針對(duì)非線(xiàn)性高斯系統(tǒng),提出了一種改進(jìn)的分布式高斯粒子濾波器,并將其應(yīng)用于麥克風(fēng)網(wǎng)絡(luò)中的說(shuō)話(huà)人跟蹤問(wèn)題。該方法在預(yù)測(cè)階段采用粒子的形式對(duì)狀態(tài)的概率密度進(jìn)行預(yù)測(cè),并對(duì)各個(gè)節(jié)點(diǎn)的局部預(yù)測(cè)信息進(jìn)行融合,進(jìn)而使每個(gè)節(jié)點(diǎn)擁有狀態(tài)后驗(yàn)概率的全局預(yù)測(cè)結(jié)果；在融合階段根據(jù)一種最優(yōu)的融合規(guī)則對(duì)各個(gè)節(jié)點(diǎn)的局部估計(jì)進(jìn)行融合并去除了局部估計(jì)之間的公共先驗(yàn),最終每個(gè)節(jié)點(diǎn)都擁有關(guān)于狀態(tài)的全局估計(jì)。該方法只要求相鄰節(jié)點(diǎn)間的局部通信,且允許各個(gè)節(jié)點(diǎn)的局部估計(jì)具有一定的相關(guān)性。仿真和實(shí)際實(shí)驗(yàn)結(jié)果表明,所提出的說(shuō)話(huà)人跟蹤方法在噪聲和混響環(huán)境中能夠?qū)\(yùn)動(dòng)的說(shuō)話(huà)人進(jìn)行有效地跟蹤。(3)針對(duì)包含線(xiàn)性、高斯子結(jié)構(gòu)的非線(xiàn)性、非高斯系統(tǒng),提出了一種分布式邊緣輔助粒子濾波器。該算法利用邊緣化技術(shù)將線(xiàn)性狀態(tài)分量從狀態(tài)空間模型中分離出來(lái),并利用分布式卡爾曼濾波器來(lái)估計(jì)；而剩余的非線(xiàn)性狀態(tài)分量則采用分布式輔助粒子濾波器來(lái)估計(jì)。針對(duì)說(shuō)話(huà)人狀態(tài)空間模型包含線(xiàn)性、高斯子結(jié)構(gòu)的特點(diǎn),將分布式邊緣輔助粒子濾波器應(yīng)用于說(shuō)話(huà)人跟蹤問(wèn)題,通過(guò)邊緣化技術(shù)將說(shuō)話(huà)人的位置信息從狀態(tài)空間模型中分離出來(lái)并采用分布式的輔助粒子濾波器來(lái)估計(jì)；而其速度信息則采用分布式卡爾曼濾波器來(lái)估計(jì)。此外,利用互相關(guān)函數(shù)的幅度信息和能量比,提出了一種時(shí)間延遲選擇機(jī)制來(lái)去除不可靠的觀測(cè),提高了其跟蹤性能。仿真和實(shí)際實(shí)驗(yàn)結(jié)果表明,所提出的方法在噪聲和混響環(huán)境中具有良好的跟蹤效果。
[Abstract]:Track of indoor speakers and a microphone array is estimated according to the position information of the audio signal of multiple microphone distribution in space of movement. The speaker received speaker tracking microphone array technology in public security monitoring system based on audio and video conferencing, voice recognition, mobile phone and robotics and other fields are widely used. In recent years, with the development of wireless sensor network, network communication, the rapid development of mobile computing and integrated circuit technology, the production of small size microphone is more and more low cost, significantly enhance the computing ability of embedded processors, distributed microphone network gradually developed, sound source localization and tracking of distributed microphone network has become a new field of speech processing research based on the hot spot. However, most of the sound source tracking algorithm is mainly aimed at the traditional rules of Mike wind array, and Can not be directly used for the distributed microphone network. Bayesian filtering is a typical method to solve indoor speaker tracking problem, it uses the method of state space modeling for speaker tracking problem. When the state space model is linear, Gauss, Calman filter is the optimal Bayesian filter solution; when the state space model is nonlinear, non Gauss, particle filtering is an effective approximate solution of Bayesian filtering. In the indoor noise and reverberation, speaker of the posterior distribution of the state for the non Gauss distribution, the observation model is usually nonlinear model. Accordingly, based on the framework of Bayesian filtering theory, the distributed particle filter algorithm to improve existing, improve the tracking accuracy and robustness filter; through in-depth study of the particle filter theory, this paper presents a new distributed particle filter. On this basis, the proposed The application of distributed particle filtering algorithm to the microphone network indoor speaker tracking, puts forward some measures for its improvement. The main contributions of this thesis are as follows: (1) in the existing distributed particle filter based on particle weight consistency in calculating the likelihood function of the observation requirements of each node in a given state under the condition of mutual independent statistical information and the need of the known observation noise. To solve this problem, this paper using the coherent field function in wide area value of a spatial location reflects the characteristics of sound source possibility at the position size to construct a pseudo likelihood function, and then derive a coherent field - wide area distributed particle filter, and the microphone for speaker tracking problems in the network. This method does not require each node of the observation conditional independence, no statistical letter should known observation noise information, And easy to distributed computing. The results of simulation and actual experiments show that the proposed method has good tracking performance in noisy and reverberant environment. (2) according to the nonlinear Gauss system, proposes an improved distributed Gauss particle filter, and its application in microphone speaker tracking problems in the network. The method of prediction by using the probability density of state particles form at the stage of prediction, and local predictive information for each node and each node of the fusion prediction results have global state posterior probability; in order according to the fusion of an optimal fusion rule for each node's local estimation fusion and remove the local estimation between the public prior, each node will have on the global state estimation. This method requires only local communication between adjacent nodes, and allows each node of the Bureau The Ministry estimates with a certain correlation. The results of simulation and actual experiments show that the speaker tracking method proposed can effectively track the movement of the speaker in a noisy and reverberant environment. (3) for linear, nonlinear Gauss sub structure, non Gauss system, proposes a distributed edge auxiliary particle filter. The the algorithm using edge technique separates the linear state component from the state space model, and the use of distributed Calman filter is used to estimate the nonlinear state; and the remaining component is used for distributed auxiliary particle filter to estimate the speaker. According to the state space model consists of a linear, Gauss sub structure, distributed edge auxiliary particle filter is applied to speaker tracking the problem, the location information of the speaker from the state space model of isolated and marginalized by technology The auxiliary particle filter to estimate and distributed; the velocity information is distributed by Calman filter to estimate. In addition, the amplitude information and energy by using correlation function, we propose a time delay selection mechanism to remove the observation is not reliable, improves the tracking performance. The results of simulation and actual experiments show that the method this has a good tracking effect in noisy and reverberant environments.

【學(xué)位授予單位】：大連理工大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類(lèi)號(hào)】：TN912.3

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 景慎艷;;需求跟蹤方法探索與實(shí)踐[J];電腦與電信;2011年10期

2 呂雪立，柯有安;基于整體特征的線(xiàn)跟蹤方法[J];北京理工大學(xué)學(xué)報(bào);1995年02期

3 賈沛璋;對(duì)飛機(jī)的跟蹤方法[J];航空學(xué)報(bào);1984年04期

4 王啟忠;預(yù)測(cè)跟蹤方法及其誤差分析[J];數(shù)值計(jì)算與計(jì)算機(jī)應(yīng)用;1990年03期

5 王濤，向健勇，，劉上乾;一種新的快速相關(guān)跟蹤方法[J];紅外與激光技術(shù);1995年01期

6 尹春霞;徐德;羅楊宇;李成榮;;一種魯棒的自適應(yīng)帶寬跟蹤方法[J];控制與決策;2013年11期

7 權(quán)偉;陳錦雄;余南陽(yáng);;在線(xiàn)學(xué)習(xí)多重檢測(cè)的可視對(duì)象跟蹤方法[J];電子學(xué)報(bào);2014年05期

8 田原,譚鐵牛,孫洪贊;一種具有良好魯棒性的實(shí)時(shí)跟蹤方法[J];自動(dòng)化學(xué)報(bào);2002年05期

9 呂金剛;吳斌;劉恒;周翔宇;邵延華;;一種快速魯棒的動(dòng)態(tài)人手跟蹤方法[J];微型機(jī)與應(yīng)用;2012年01期

10 楊?,裴繼紅,楊萬(wàn)海;實(shí)時(shí)光點(diǎn)檢測(cè)與跟蹤方法研究[J];紅外與毫米波學(xué)報(bào);2001年04期

相關(guān)會(huì)議論文 前9條

1 王寧;陳常念;陳加忠;范曄斌;王冼;;標(biāo)定環(huán)境下基于粒子濾波的行人跟蹤方法[A];第七屆和諧人機(jī)環(huán)境聯(lián)合學(xué)術(shù)會(huì)議（HHME2011)論文集【oral】[C];2011年

2 劉勛;孟驤龍;毋立芳;;基于自適應(yīng)對(duì)象模型的籃球運(yùn)動(dòng)跟蹤方法[A];第十二屆全國(guó)信號(hào)處理學(xué)術(shù)年會(huì)（CCSP-2005）論文集[C];2005年

3 羅詩(shī)途;張s

本文編號(hào)：1379073