本文選題：無(wú)線傳感器網(wǎng)絡(luò)　切入點(diǎn)：時(shí)鐘同步　出處：《華北電力大學(xué)(北京)》2017年碩士論文

【摘要】：目標(biāo)跟蹤技術(shù)一直以來(lái)都是探索的熱點(diǎn),其應(yīng)用范圍涉及軍事、工業(yè)、商業(yè)、醫(yī)療等各個(gè)領(lǐng)域。目標(biāo)跟蹤技術(shù)出現(xiàn)至今已有五十余年,帶動(dòng)了大量的理論與相關(guān)技術(shù)的進(jìn)步。隨著傳感器技術(shù)近些年的飛速發(fā)展,無(wú)線傳感器網(wǎng)絡(luò)出現(xiàn)得到了各方的高度關(guān)注,由于其覆蓋范圍廣泛,成本低廉以及檢測(cè)方式的多樣性的特點(diǎn),無(wú)線傳感器網(wǎng)絡(luò)將成為二十一世紀(jì)最重要的技術(shù)之一�；跓o(wú)線傳感器網(wǎng)絡(luò)的目標(biāo)跟蹤是當(dāng)今研究熱點(diǎn)之一。本文首先介紹了無(wú)線傳感器網(wǎng)絡(luò)的體系結(jié)構(gòu)、節(jié)點(diǎn)定位技術(shù)和目標(biāo)跟蹤的技術(shù),然后總結(jié)了無(wú)線傳感器網(wǎng)絡(luò)定位與跟蹤需考慮的一些因素包括目標(biāo)運(yùn)動(dòng)狀態(tài)的建模、常用的運(yùn)動(dòng)模型,并且指出傳統(tǒng)的目標(biāo)跟蹤算法在無(wú)線傳感器網(wǎng)絡(luò)跟蹤中面臨的問(wèn)題包括難以確定目標(biāo)和濾波器的初始狀態(tài)、間歇性目標(biāo)丟失以及在雜波與誤測(cè)量較多的情形下其濾波性能較差。針對(duì)上述缺陷,本文基于隨機(jī)集理論(RFS)和有限集統(tǒng)計(jì)(FISST)方法,提出了一種針對(duì)無(wú)線傳感器網(wǎng)絡(luò)多目標(biāo)跟蹤的粒子PHD濾波器,并且針對(duì)傳統(tǒng)PHD濾波器的計(jì)算復(fù)雜度較高的問(wèn)題,設(shè)計(jì)了一種分布式粒子PHD濾波器改善其跟蹤的實(shí)時(shí)性能。在基于超聲波測(cè)距的無(wú)線傳感網(wǎng)絡(luò)中,本文首先設(shè)計(jì)了一種基于HC-SR04陣列的360°超聲波傳感器硬件節(jié)點(diǎn),而后針對(duì)粒子濾波分布式計(jì)算的同步需求,提出了一種基于AODV路由協(xié)議的時(shí)鐘同步算法TS-AODV,隨后在無(wú)線傳感器網(wǎng)絡(luò)時(shí)鐘同步的基礎(chǔ)上,研究并實(shí)現(xiàn)了分布式粒子PHD濾波。通過(guò)實(shí)驗(yàn)證明,粒子PHD濾波算法在傳感器網(wǎng)絡(luò)目標(biāo)跟蹤應(yīng)用中,能夠有效的估計(jì)出多目標(biāo)運(yùn)動(dòng)的目標(biāo)狀態(tài)、目標(biāo)數(shù)量,并且在雜波較多的情形下性能穩(wěn)定,分布式粒子PHD濾波算法與集中式粒子PHD濾波算法相比,其單步計(jì)算時(shí)間較短,從而提高了算法的實(shí)時(shí)性。
[Abstract]:Target tracking technology has always been a hot topic of exploration. Its application covers military, industrial, commercial, medical and other fields. It has been more than 50 years since target tracking technology emerged. With the rapid development of sensor technology in recent years, the emergence of wireless sensor networks has been highly concerned by all parties, because of its wide coverage, Low cost and diversity of testing methods, Wireless sensor network (WSN) will become one of the most important technologies in the 21 century. Target tracking based on WSN is one of the hot research topics. Firstly, the architecture of WSN is introduced in this paper. Node location technology and target tracking technology, then summarizes some factors that need to be considered in wireless sensor network localization and tracking, including the modeling of moving state of the target, the commonly used motion model, It is pointed out that the traditional target tracking algorithms face some problems in wireless sensor network tracking, including the difficulty in determining the initial state of the target and the filter. The intermittent target loss and the poor filtering performance in the case of more clutter and mismeasurement. In view of the above defects, this paper is based on the stochastic set theory (RFS) and the finite set statistical analysis (fish) method. A particle PHD filter for multi-target tracking in wireless sensor networks is proposed, and the computational complexity of the traditional PHD filter is high. A distributed particle PHD filter is designed to improve its real-time tracking performance. In the wireless sensor network based on ultrasonic ranging, a 360 擄ultrasonic sensor hardware node based on HC-SR04 array is designed in this paper. Then a clock synchronization algorithm TS-AODV based on AODV routing protocol is proposed to meet the synchronization requirement of particle filter distributed computing. Then it is based on clock synchronization in wireless sensor networks. The distributed particle PHD filter is studied and implemented. The experimental results show that the particle PHD filter algorithm can effectively estimate the target state and the number of targets moving in the sensor network. And the performance of distributed particle PHD filter is stable in the case of more clutter. Compared with the centralized particle PHD filter, the single-step computation time of the distributed particle PHD filter algorithm is shorter, thus improving the real-time performance of the algorithm.
【學(xué)位授予單位】：華北電力大學(xué)(北京)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP212.9;TN929.5

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 陳輝;韓崇昭;;機(jī)動(dòng)多目標(biāo)跟蹤中的傳感器控制策略的研究[J];自動(dòng)化學(xué)報(bào);2016年04期

2 楊峰;王永齊;梁彥;潘泉;;基于概率假設(shè)密度濾波方法的多目標(biāo)跟蹤技術(shù)綜述[J];自動(dòng)化學(xué)報(bào);2013年11期

3 童慧思;張顥;孟華東;王希勤;;PHD濾波器在多目標(biāo)檢測(cè)前跟蹤中的應(yīng)用[J];電子學(xué)報(bào);2011年09期

4 司海飛;楊忠;王s，

本文編號(hào)：1674934