本文選題：WIFI定位 + 位置指紋算法�。� 參考：《武漢郵電科學(xué)研究院》2017年碩士論文

【摘要】：隨著科學(xué)技術(shù)的發(fā)展,人們對(duì)于獲取自身位置的需求越來越多,基于位置的服務(wù)獲得了越來越多的關(guān)注。目前廣泛使用的GPS定位技術(shù),可以快速精確的提供定位服務(wù),廣泛的應(yīng)用于手機(jī)導(dǎo)航等方面,但是由于GPS信號(hào)穿透墻壁時(shí)衰減嚴(yán)重,因此GPS技術(shù)不能用于室內(nèi)定位。近幾年來WIFI技術(shù)發(fā)展迅速,WIFI覆蓋率大幅上升,基于WIFI技術(shù)實(shí)現(xiàn)隧道人員定位,無論從組網(wǎng)能力、經(jīng)濟(jì)效益、還是定位精度上來說都是非常適合的�；赪IFI實(shí)現(xiàn)室內(nèi)定位,一般使用RSSI(接收信號(hào)強(qiáng)度指示)作為信號(hào)的特征值。RSSI定位算法有兩大類,一種是基于測(cè)距的算法,另一種是無需測(cè)距的算法。基于測(cè)距的定位算法是根據(jù)無線信號(hào)的自由空間中的傳輸損耗模型,對(duì)比目標(biāo)位置RSSI值來確定位置,對(duì)環(huán)境有非常大的依賴性。無需測(cè)距的RSSI定位算法即位置指紋算法,分為離線指紋采集和在線位置估計(jì)兩個(gè)階段。離線指紋采集階段需要采集每一個(gè)位置的RSSI值建立指紋數(shù)據(jù)庫(kù)。在線位置估計(jì)階段,取得目標(biāo)位置的RSSI值,與數(shù)據(jù)庫(kù)匹配,估算出目標(biāo)位置。本論文選取了隧道人員定位作為研究方向,在現(xiàn)有的WIFI定位的基礎(chǔ)上,分析了現(xiàn)有定位算法的優(yōu)缺點(diǎn),并作出相應(yīng)的改進(jìn),主要內(nèi)容可歸納如下。(1)分析了基于測(cè)距的定位算法,采用了基于傳輸損耗模型獲得待測(cè)點(diǎn)與參考點(diǎn)的距離,采用加權(quán)質(zhì)心的極大似然估計(jì)方法估算位置,提高了測(cè)距算法的定位精度。(2)為了提高傳統(tǒng)位置指紋定位算法的定位精度,在離線指紋采集階段,對(duì)于指紋數(shù)據(jù)進(jìn)行降噪處理。使用聚類分析的方法訓(xùn)練指紋數(shù)據(jù)庫(kù),降低算法的時(shí)間復(fù)雜度。在在線位置估計(jì)階段,則采用改進(jìn)的最鄰近算法去來估算位置。這種方法可以有效的提高傳統(tǒng)的位置指紋算法的定位精度。(3)測(cè)試改進(jìn)后的定位算法的可行性,比較改進(jìn)前后的算法的定位精度。分析影響定位精度的因素。
[Abstract]:With the development of science and technology, more and more people need to obtain their own location, and more and more attention has been paid to location-based services. At present, the widely used GPS positioning technology can quickly and accurately provide positioning services, widely used in mobile phone navigation and other aspects, but because the GPS signal attenuation through the wall, GPS technology can not be used for indoor positioning. With the rapid development of WIFI technology in recent years, the coverage rate of WiFi has increased rapidly. It is very suitable for tunnel personnel positioning based on WIFI technology in terms of network ability, economic benefit and positioning accuracy. RSSI (received signal intensity indication) is generally used as the eigenvalue of the signal based on WIFI. There are two kinds of localization algorithms: one is based on ranging and the other is without ranging. The location algorithm based on ranging is based on the transmission loss model in the free space of wireless signal. The location is determined by comparing the RSSI value of the target location, which is highly dependent on the environment. The RSSI location algorithm without ranging, that is, the location fingerprint algorithm, is divided into two stages: off-line fingerprint acquisition and on-line position estimation. In the off-line fingerprint acquisition stage, the RSSI value of each location should be collected to establish the fingerprint database. In the stage of online position estimation, the RSSI value of the target position is obtained, matched with the database, and the target position is estimated. This paper selects the tunnel personnel localization as the research direction, on the basis of the existing WIFI localization, analyzes the advantages and disadvantages of the existing localization algorithm, and makes the corresponding improvement, the main content can be summarized as follows: 1) analyze the location algorithm based on ranging. Based on the transmission loss model, the distance between the point to be measured and the reference point is obtained. The maximum likelihood estimation method of weighted centroid is used to estimate the location, which improves the location accuracy of the ranging algorithm. In the off-line fingerprint acquisition stage, the fingerprint data is de-noised. The method of clustering analysis is used to train the fingerprint database to reduce the time complexity of the algorithm. In the stage of online position estimation, an improved nearest neighbor algorithm is used to estimate the position. This method can effectively improve the location accuracy of the traditional location fingerprint algorithm. It can test the feasibility of the improved location algorithm and compare the location accuracy of the improved algorithm. The factors influencing the positioning accuracy are analyzed.
【學(xué)位授予單位】：武漢郵電科學(xué)研究院
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TN92

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王s檛，

本文編號(hào)：1826185