【摘要】：隨著移動互聯網行業(yè)的高速發(fā)展,人們對于室內定位的需求日益迫切,作為位置信息服務最重要的組成部分,室內定位技術逐漸受到越來越多的商業(yè)關注。由于技術的進步,人們對于室內定位的精度要求也越來越高,但是由于室內環(huán)境的反射和多徑傳播,高精度室內定位問題一直是定位領域的最大難題。論文以信息柔性融合技術為基礎,主要研究如何在復雜室內環(huán)境下得到良好的室內定位結果。論文所做的主要工作如下:(1)論文對室內定位技術的研究現狀進行了總結,剖析了各種室內定位技術的優(yōu)缺點和技術難點�；诳死懒_界,分析了TOA定位算法的理論極限,并推導了基站噪聲特性不一致時TOA算法的CRLB。針對制約室內定位精度提高的瓶頸問題,設計了基于信息柔性融合的室內定位系統框架與結構。(2)針對信息柔性融合的數據層融合算法,分別研究了單/多傳感器條件下的定位參數估計、多定位參數融合定位等問題。在此基礎上提出了一種基于RSSI/AOA的新型室內精確定位方法,仿真和測試結果顯示本文所給算法可以有效濾除參數測量波動造成定位結果誤差,有效提高了室內復雜環(huán)境下的目標定位精度。(3)建立了基于強跟蹤Kalman濾波器的決策層融合算法。解決Kalman濾波器在運動目標定位跟蹤方面的缺陷:對突變狀態(tài)的跟蹤能力差,對于大噪聲的濾除能力弱。改進強跟蹤算法中的次優(yōu)漸消因子,根據前次結果對強跟蹤結果做反饋,形成了基于指數漸消因子的融合跟蹤算法(EFF-STF)。詳細說明了算法的實現步驟,建立系統實際采集數據驗證了算法的有效性,并和相同條件下和卡爾曼濾波性能對比驗證了算法在大噪聲濾除方面的優(yōu)越性,通過跟蹤前后的定位誤差CDF圖驗證了跟蹤性能對定位誤差的良好修正作用。(4)論文利用MATLAB和VS2010聯合編程,實現了基于低功耗藍牙的室內定位系統,將定位結果可視化,驗證了算法的性能,實驗測試表明經過信息柔性融合和跟蹤處理之后,可以將85%的定位誤差約束在1m以內,實現了室內實時精確定位的需求。
[Abstract]:With the rapid development of mobile Internet industry, the demand for indoor positioning becomes more and more urgent. As the most important part of location information service, indoor positioning technology has attracted more and more commercial attention. Due to the progress of technology, the accuracy of indoor positioning is more and more demanding. However, due to the reflection and multipath propagation of indoor environment, the problem of high-precision indoor positioning has always been the biggest problem in the field of positioning. Based on information flexible fusion technology, this paper mainly studies how to obtain good indoor positioning results in complex indoor environment. The main work of this paper is as follows: (1) the research status of indoor positioning technology is summarized, and the advantages and disadvantages and technical difficulties of various indoor positioning techniques are analyzed. Based on the Caramello bound, the theoretical limit of the TOA localization algorithm is analyzed, and the CRLB. of the TOA algorithm is derived when the noise characteristics of the base station are inconsistent. In order to solve the bottleneck problem which restricts the improvement of indoor positioning accuracy, the frame and structure of indoor positioning system based on information flexible fusion are designed. (2) the data layer fusion algorithm for information flexible fusion is proposed. The estimation of location parameters and fusion localization of multi-location parameters under single / multi-sensor conditions are studied respectively. On this basis, a new precise indoor location method based on RSSI/AOA is proposed. The simulation and test results show that the proposed algorithm can effectively filter the error caused by parameter measurement fluctuation. The accuracy of target location in complex indoor environment is improved effectively. (3) A decision-level fusion algorithm based on strong tracking Kalman filter is proposed. The defects of Kalman filter in moving target location and tracking are solved: poor tracking ability for abrupt state and weak filtering ability for large noise. A fusion tracking algorithm based on exponential fading factor (EFF-STF) is developed by improving the suboptimal fading factor in the strong tracking algorithm and feedback the strong tracking results according to the previous results. The implementation steps of the algorithm are described in detail, and the validity of the algorithm is verified by the establishment of the system's actual data collection, and the superiority of the algorithm in large noise filtering is verified by comparing the performance of the algorithm with the Kalman filter under the same conditions. The positioning error CDF diagram before and after tracking verifies the good correction effect of tracking performance on the positioning error. (4) in this paper, the indoor positioning system based on low power Bluetooth is realized by using MATLAB and VS2010 joint programming, and the positioning results are visualized. The performance of the algorithm is verified. The experimental results show that after information flexible fusion and tracking processing, 85% of the positioning error can be limited to less than 1m, and the requirement of real-time and accurate indoor positioning can be realized.
【學位授予單位】：河南師范大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TN713;TN925

【參考文獻】

相關期刊論文 前8條

1 李協;張效義;于旭;胡峗鵬;;相對定位估計的修正克拉美羅界[J];數據采集與處理;2013年02期

2 范馨月;陳庭盈;周非;;基于參數重構的非直達波混合定位算法[J];信號處理;2011年11期

3 孫章國;錢峰;;一種基于指數漸消因子的自適應卡爾曼濾波算法[J];電子測量技術;2010年01期

4 王鑫;趙春暉;戎建剛;;多路延遲結構的修正MUSIC算法頻率估計[J];系統工程與電子技術;2009年04期

5 劉海燕;趙宗貴;劉熹;;D-S證據理論中沖突證據的合成方法[J];電子科技大學學報;2008年05期

6 焦磊;邢建平;張軍;張璇;趙朝麗;;一種非視距環(huán)境下具有魯棒特性TOA無線傳感網絡定位算法[J];傳感技術學報;2007年07期

7 凌林本,李滋剛,陳超英,谷云彪,李傳學;多傳感器數據融合時權的最優(yōu)分配原則[J];中國慣性技術學報;2000年02期

8 翟翌立,戴逸松;多傳感器數據自適應加權融合估計算法的研究[J];計量學報;1998年01期

相關博士學位論文 前1條

1 王建輝;基于信息融合的蜂窩網定位技術研究[D];解放軍信息工程大學;2011年

相關碩士學位論文 前1條

1 竹博;蜂窩網目標的定位跟蹤技術研究[D];解放軍信息工程大學;2013年

，

本文編號：2438052