發(fā)布時間：2018-04-06 06:03

  本文選題：室內(nèi)無線定位　切入點：低功耗藍牙　出處：《西安電子科技大學(xué)》2015年碩士論文

【摘要】：隨著智能移動終端的普及和互聯(lián)網(wǎng)技術(shù)的飛速發(fā)展,基于位置的服務(wù)越來越受到人們的關(guān)注,它在醫(yī)療健康、安全監(jiān)管、推送信息等領(lǐng)域具有巨大的潛力�，F(xiàn)如今人們的工作、生活大部分時間在室內(nèi)進行,而傳統(tǒng)的GPS定位系統(tǒng)并不能適應(yīng)室內(nèi)復(fù)雜多變的環(huán)境,所以利用無線技術(shù)實現(xiàn)室內(nèi)定位成為了當(dāng)今研究的熱點。由于傳統(tǒng)藍牙技術(shù)的功耗高、穿透性差、信號傳輸距離短等原因在室內(nèi)定位領(lǐng)域沒有得到廣泛的推廣。低功耗藍牙4.0通信標準的發(fā)布使得其應(yīng)用在室內(nèi)定位中成為可能。它在功耗、傳輸距離、信號強度等方面都有了明顯的改善,因此本文在低功耗藍牙技術(shù)的基礎(chǔ)上對室內(nèi)定位算法進行了研究。本文分析了室內(nèi)環(huán)境中藍牙信號在室內(nèi)的分布特征以及影響藍牙信號強度分布特征的因素,測試了距離、傳輸路徑、設(shè)備對接收信號強度的影響程度。由于室內(nèi)環(huán)境復(fù)雜多變,采用信號傳播模型的定位方法會產(chǎn)生較大的誤差,位置指紋定位算法更適用于室內(nèi)環(huán)境中。但是傳統(tǒng)位置指紋定位在時間復(fù)雜度和精確度上存在不足。所以針對傳統(tǒng)位置指紋定位算法的缺陷,本文采取了改進的措施。在建立位置指紋庫時,采用k-means聚類和模糊c均值聚類的分析初始的位置指紋庫,并劃分成多個子類。實時數(shù)據(jù)不再和初始數(shù)據(jù)庫中的所有數(shù)據(jù)匹配,而是將它歸屬到某一個子類中,并和子類中的指紋數(shù)據(jù)匹配。這樣有效地縮減了搜索空間,減少匹配時間的同時還削弱了指紋庫對定位結(jié)果的影響。在位置估算時,對原有的權(quán)重系數(shù)進行了改進,使估算類與類間的邊界時的精確度得到了提高,類內(nèi)的點估計不會受到影響,從而提高了整體的定位精度。最后在搭建的測試平臺基礎(chǔ)上采集實際室內(nèi)環(huán)境中的低功耗藍牙信號并在matlab上對提出的算法進行了實驗驗證,結(jié)果顯示在1m的誤差范圍內(nèi)基于模糊c均值的FWKNN算法能夠達到80%的概率,證明了改進后的位置指紋定位算法與傳統(tǒng)算法相比,定位的精度得到了有效地提升。
[Abstract]:With the popularity of intelligent mobile terminals and the rapid development of Internet technology, location-based services have attracted more and more attention. It has great potential in the fields of health care, safety supervision, push information and so on.Nowadays, people work and live most of the time indoors, but the traditional GPS positioning system can not adapt to the complex and changeable indoor environment, so using wireless technology to achieve indoor positioning has become a hot topic.Because of the high power consumption, poor penetration and short transmission distance of the traditional Bluetooth technology, it has not been widely used in the field of indoor positioning.The release of low power Bluetooth 4.0 communication standard makes its application in indoor positioning possible.It has obvious improvement in power consumption, transmission distance and signal intensity, so this paper studies indoor location algorithm based on low power Bluetooth technology.In this paper, the distribution characteristics of Bluetooth signal in indoor environment and the factors influencing the intensity distribution of Bluetooth signal are analyzed. The influence of distance, transmission path and equipment on the intensity of received signal is tested.Because the indoor environment is complex and changeable, the localization method based on the signal propagation model will produce large errors, and the location fingerprint location algorithm is more suitable for indoor environment.However, the traditional location fingerprint location has some shortcomings in time complexity and accuracy.Therefore, aiming at the defects of the traditional location fingerprint location algorithm, this paper takes some improved measures.When the location fingerprint database is established, the initial location fingerprint database is analyzed by k-means clustering and fuzzy c-means clustering, and is divided into several subclasses.The real-time data is no longer matched with all the data in the initial database, but belongs to a subclass and matches the fingerprint data in the subclass.In this way, the search space is reduced effectively, the matching time is reduced, and the influence of fingerprint database on the location results is weakened.In position estimation, the original weight coefficient is improved to improve the accuracy of estimating the boundary between the class and the class, and the point estimation within the class will not be affected, thus improving the overall positioning accuracy.Finally, the low-power Bluetooth signal in the indoor environment is collected on the basis of the test platform, and the proposed algorithm is tested on matlab.The results show that the probability of FWKNN algorithm based on fuzzy c-means can reach 80% in the error range of 1m. It is proved that the improved location fingerprint location algorithm can improve the accuracy of location effectively compared with the traditional algorithm.
【學(xué)位授予單位】：西安電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：TN925

【參考文獻】

相關(guān)期刊論文 前10條

1 朱中一;;基于WiFi的室內(nèi)定位技術(shù)在博物館的應(yīng)用[J];軟件產(chǎn)業(yè)與工程;2013年03期

2 錢志鴻;王義君;;物聯(lián)網(wǎng)技術(shù)與應(yīng)用研究[J];電子學(xué)報;2012年05期

3 吳靜;吳曉燕;高忠長;;客觀多因素權(quán)重分配方法及其應(yīng)用[J];上海航天;2011年03期

4 徐玉濱;鄧志安;馬琳;;基于核直接判別分析和支持向量回歸的WLAN室內(nèi)定位算法[J];電子與信息學(xué)報;2011年04期

5 羅瑋;;一種新興的藍牙技術(shù)——超低功耗藍牙技術(shù)[J];現(xiàn)代電信科技;2010年10期

6 楊元喜;;北斗衛(wèi)星導(dǎo)航系統(tǒng)的進展、貢獻與挑戰(zhàn)[J];測繪學(xué)報;2010年01期

7 李文杰;李文明;;基于k-近鄰算法的定位方法設(shè)計和仿真[J];計算機仿真;2009年04期

8 李昊;;位置指紋定位技術(shù)[J];山西電子技術(shù);2007年05期

9 程曉暢;蘇紹景;王躍科;杜金榜;;類GPS超聲定位系統(tǒng)中幾種定位算法比較[J];傳感技術(shù)學(xué)報;2007年03期

10 黎俊鋒;朱鋒峰;;基于樣本密度的FCM改進算法[J];科學(xué)技術(shù)與工程;2007年04期

，

本文編號：1718268