无线互联科技
無線互聯科技
무선호련과기
WUXIAN HULIAN KEJI
2014年
4期
121-123,155
,共4页
车联网WAVE%高斯混合模型%EM算法%定位系统
車聯網WAVE%高斯混閤模型%EM算法%定位繫統
차련망WAVE%고사혼합모형%EM산법%정위계통
Internet of Vehicles%WAVE%GMM%EM%Positioning Systems
无线定位是车联网中各应用的基础,对实时性和精度要求更高。基于此,本文提出了一个在WAVE系统下基于混合高斯模型的位置指纹定位算法。该算法由离线训练和在线定位两部分组成,采用GMM对离线训练阶段的指纹数据库进行聚类建模处理,并在定位阶段对结果采用多点均值方法,不仅降到了系统定位误差,也减少了定位阶段的运算量,提高了定位的实时性。实验结果表明,该算法有较好的定位精度和实时性,为WAVE环境下的快速定位进一步研究提供有效参考价值。
無線定位是車聯網中各應用的基礎,對實時性和精度要求更高。基于此,本文提齣瞭一箇在WAVE繫統下基于混閤高斯模型的位置指紋定位算法。該算法由離線訓練和在線定位兩部分組成,採用GMM對離線訓練階段的指紋數據庫進行聚類建模處理,併在定位階段對結果採用多點均值方法,不僅降到瞭繫統定位誤差,也減少瞭定位階段的運算量,提高瞭定位的實時性。實驗結果錶明,該算法有較好的定位精度和實時性,為WAVE環境下的快速定位進一步研究提供有效參攷價值。
무선정위시차련망중각응용적기출,대실시성화정도요구경고。기우차,본문제출료일개재WAVE계통하기우혼합고사모형적위치지문정위산법。해산법유리선훈련화재선정위량부분조성,채용GMM대리선훈련계단적지문수거고진행취류건모처리,병재정위계단대결과채용다점균치방법,불부강도료계통정위오차,야감소료정위계단적운산량,제고료정위적실시성。실험결과표명,해산법유교호적정위정도화실시성,위WAVE배경하적쾌속정위진일보연구제공유효삼고개치。
Wireless Location is the basis of each vehicle networking applications, needing real-time and high precision. So as,we propose an algorithm for fingerprint positioning based on a Gaussian mixture model(GMM) in the WAVE system,which consists of an offline training phase and a real time localization phase. On offline training phase, we use GMM clustering process the whole fingerprint database,and use the multi-point average method processing results of the positioning on a real time localization phase, which not only reduces the system positioning errors, but also reduces the amount of computation, and improves real-time positioning. Experimental results show that the algorithm has better performance in both accuracy and real-time, and also provides effective reference value for further researching the fast positioning in WAVE environment.