發(fā)布時(shí)間：2019-02-17 15:48

【摘要】：隨著無線通信和智能移動(dòng)終端技術(shù)的飛速發(fā)展,人們對(duì)位置感知服務(wù)的要求越來越高,而高精度的室內(nèi)定位技術(shù)是實(shí)現(xiàn)位置感知服務(wù)的核心。在當(dāng)前室內(nèi)環(huán)境WAN分布廣泛和運(yùn)用普遍化的今天,使得基于WLAN定位系統(tǒng)成為研究的熱點(diǎn)課題。基于位置指紋的WLAN室內(nèi)定位具有實(shí)現(xiàn)簡(jiǎn)單、成本較低,無需知道AP的位置和發(fā)射功率即可實(shí)現(xiàn)定位,對(duì)額外的硬件設(shè)備需求量少等優(yōu)點(diǎn),使其在學(xué)術(shù)界和工業(yè)界得到廣泛的關(guān)注和運(yùn)用。位置指紋定位技術(shù)主要分為兩個(gè)階段：采樣階段和定位階段,當(dāng)前位置指紋定位技術(shù)在實(shí)現(xiàn)定位時(shí),在兼顧位置指紋定位算法的精度和效率上還沒有相對(duì)完善的機(jī)制,因此本文針對(duì)這個(gè)問題對(duì)位置指紋定位技術(shù)實(shí)施優(yōu)化改進(jìn)。 本文通過分析比較幾種典型的位置指紋定位算法,KNN定位算法在時(shí)間復(fù)雜度和定位精度上都有一定的優(yōu)勢(shì),但是KNN算法在定位時(shí)需要耗費(fèi)巨大的時(shí)間與指紋數(shù)據(jù)庫進(jìn)行比對(duì)從而確定K個(gè)指紋數(shù)據(jù)的選取,且由于K的選擇是固定的而影響了某些位置處的定位精度。 針對(duì)KNN定位算法存在的不足,本文提出了一種基于模糊聚類改進(jìn)的KNN定位算法,對(duì)定位環(huán)節(jié)的龐大指紋數(shù)據(jù)庫運(yùn)用聚類分析方法實(shí)現(xiàn)模糊聚類,繼而采用KNN算法實(shí)現(xiàn)移動(dòng)終端的定位。對(duì)改進(jìn)后的位置指紋定位技術(shù)性能利用MATLAB仿真實(shí)驗(yàn)進(jìn)行測(cè)試,并在Android平臺(tái)上通過原型系統(tǒng)進(jìn)行實(shí)驗(yàn)驗(yàn)證。 最后通過仿真實(shí)驗(yàn)分別對(duì)傳統(tǒng)的KNN算法和改進(jìn)后的KNN算法進(jìn)行分析比較,在不影響定位系統(tǒng)其他性能的機(jī)制下,最終實(shí)驗(yàn)結(jié)果表明改進(jìn)后的位置指紋定位在時(shí)間性能和匹配效率上都有極大的提高。
[Abstract]:With the rapid development of wireless communication and intelligent mobile terminal technology, people demand more and more location sensing services, and high precision indoor positioning technology is the core of location sensing services. With the wide distribution and generalization of WAN in indoor environment, WLAN based positioning system has become a hot topic. WLAN indoor location based on position fingerprint has the advantages of simple realization, low cost, no need to know the location and transmitting power of AP, and less demand for additional hardware equipment. It has been widely concerned and applied in academia and industry. The location fingerprint location technology is mainly divided into two stages: sampling stage and location stage. When the current location fingerprint localization technology realizes the location, there is not a relatively perfect mechanism in taking into account the accuracy and efficiency of the location fingerprint location algorithm. Therefore, this paper optimizes and improves the location fingerprint location technology. Through the analysis and comparison of several typical location fingerprint location algorithms, the KNN localization algorithm has some advantages in time complexity and location accuracy. However, the KNN algorithm needs to spend a great deal of time comparing with the fingerprint database to determine the selection of K fingerprint data, and because the selection of K is fixed, the location accuracy of some locations is affected. In view of the shortcomings of KNN localization algorithm, this paper proposes an improved KNN location algorithm based on fuzzy clustering, which uses clustering analysis method to realize fuzzy clustering for large fingerprint database of location link. Then the location of mobile terminal is realized by KNN algorithm. The performance of the improved location fingerprint location technology is tested by MATLAB simulation experiment and verified by the prototype system on the Android platform. Finally, the traditional KNN algorithm and the improved KNN algorithm are analyzed and compared by simulation experiments. The experimental results show that the time performance and matching efficiency of the improved location fingerprint location are greatly improved.
【學(xué)位授予單位】：云南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TN925.93

【參考文獻(xiàn)】

相關(guān)期刊論文 前6條

1 房秉毅;;基于超寬帶技術(shù)的室內(nèi)定位系統(tǒng)[J];電子技術(shù)應(yīng)用;2006年07期

2 楊帆;趙東東;;基于Android平臺(tái)的WiFi定位[J];電子測(cè)量技術(shù);2012年09期

3 馮浩然;袁睿翕;慕春棣;;無線網(wǎng)絡(luò)中基于信號(hào)強(qiáng)度的定位及算法比較[J];計(jì)算機(jī)工程與應(yīng)用;2006年32期

4 周傲英;楊彬;金澈清;馬強(qiáng);;基于位置的服務(wù):架構(gòu)與進(jìn)展[J];計(jì)算機(jī)學(xué)報(bào);2011年07期

5 羅軍舟;吳文甲;楊明;;移動(dòng)互聯(lián)網(wǎng):終端、網(wǎng)絡(luò)與服務(wù)[J];計(jì)算機(jī)學(xué)報(bào);2011年11期

6 余小鵬;周德翼;;一種自適應(yīng)k-最近鄰算法的研究[J];計(jì)算機(jī)應(yīng)用研究;2006年02期

，

本文編號(hào)：2425328