WiFi指纹定位的一种新组合算法

探讨AP选取策略和贝叶斯位置估计算法对基于RSS的WiFi室内定位技术位置估计精度的影响。国内外学者分别对AP选取算法和贝叶斯位置估计算法进行了大量的研究。为了进一步深入研究不同算法的优劣性,利用组合优化的思想对不同算法进行组合,通过找出最优算法组合从而提升WiFi室内定位系统的性能。基于互信息最小化的AP选取策略和考虑AP相关性的贝叶斯位置估计算法,提出一种新的WiFi指纹定位组合算法。实验分析表明:新算法具有良好的实用性和定位性能。     

多空间数据库位置匹配方法及其应用

同一地区由于不同的要求和目的而建立了多种数据库 ,包括不同比例尺的基础地理信息数据库或空间专题属性数据库。在实际应用中经常需要将这些数据库中的数据联系起来 ,相互利用。在实践中通常可以利用存在位置属性这一特点将这些数据库关联 ,达到充分利用数据、挖掘数据潜在信息的目的。本文以国家基础地理信息数据库中全国 1∶2 5万和全国 1∶1 0 0万数据的公路和河流要素为实例 ,阐述利用空间位置这一特性来建立数据库联系的思路方法     

GIS中3维空间圆曲线的不确定性εσ模型

随着GIS向3维领域的不断发展,对3维空间元素的不确定性研究日趋重要.基于不确定性理论及概率统计理论,研究GIS中3维空间圆曲线的不确定性模型.首先建立3维空间圆曲线上任意点坐标的微分关系式,然后得到圆曲线上任意点坐标的方差, 进而获得3维空间圆曲线的εσ模型.实际算例表明,3维空间圆曲线误差分布呈现出"两端大,中间小"的趋势.     

伪卫星定位系统中的频率与时间统一

在电磁波定位系统中,时间和频率的统一与稳定性非常重要,会直接影响系统的可用性和精度．通过对一个伪卫星定位系统中时间和频率问题的研究,设计出了一套完整的时频统一方案,并编写了程序实现。该方案可以有效提高系统初始化速度,并提高观测质量。     

Nontectonic signals in reprocessed continuous GPS position time series of CMONOC

The nearly nine-year continuous GPS data collected since 1 March 1999 from the Crustal Motion Observation Network of China(CMONOC) were consistently analyzed.Most of the nonlinear movements in the cumulative position time series pro-duced by CMONOC data center disappeared;and more accurate vertical terms and tectonic signals were extracted.Displacements caused by atmospheric pressure loading,nontidal ocean loading,soil moisture mass loading,and snow cover mass loading using the National Centers for Environmental Prediction(NCEP) Reanalysis I/II models and Estimation of the Circulation and Climate of the Ocean(ECCO) data can explain most of the vertical annual terms at many stations,while only parts can be explained at Lhasa and southern coastal sites,indicating that there are some deformation mechanisms that are still unknown or not modeled accurately.The remarkable differences in vertical position time series for short-baseline sites reveal that GPS stations can be greatly affected by lo-cal factors;and attention should be paid when explaining observed GPS velocity vectors.     

全站仪坐标导线的施测与平差探讨

提出了全站仪坐标导线外业施测时的检核方法和内业几种严密平差方法。     

Separation problem in satellite gravimetry

The problem of separating gravitation from inertia is discussed in very general sense, and the conclusion is positive: man can separate gravitation from inertia, if various observation techniques are applied for. The accelerometer’s position problem in satellite gravimetry is investigated, and the additional acceleration effect due to the position error of an instrument as well as the difference between the mass center and the gravity center is explored.     

Nontectonic signals in reprocessed continuous GPS position time series of CMONOC

The nearly nine-year continuous GPS data collected since 1 March 1999 from the Crustal Motion Observation Network of China (CMONOC) were consistently analyzed. Most of the nonlinear movements in the cumulative position time series produced by CMONOC data center disappeared; and more accurate vertical terms and tectonic signals were extracted. Displacements caused by atmospheric pressure loading, nontidal ocean loading, soil moisture mass loading, and snow cover mass loading using the National Centers for Environmental Prediction (NCEP) Reanalysis I/II models and Estimation of the Circulation and Climate of the Ocean (ECCO) data can explain most of the vertical annual terms at many stations, while only parts can be explained at Lhasa and southern coastal sites, indicating that there are some deformation mechanisms that are still unknown or not modeled accurately. The remarkable differences in vertical position time series for short-baseline sites reveal that GPS stations can be greatly affected by local factors; and attention should be paid when explaining observed GPS velocity vectors.     

最佳小波包基的矿区植被及红边位置提取

小波包变换能同时对植被光谱信息的低频和高频分量进行分解,并能克服小波变换时间分辨率高而频率分辨率低的缺陷从而具有能够探测植被细微变化的优势。实验利用Hyperion高光谱影像对云南省普朗铜矿区植被像元的光谱进行最佳小波包基参量获取与植被信息识别,并在此基础上提出一种提取重金属污染下植被红边位置的最佳小波包基系数应用模型。研究结果表明基于最佳小波包基参量的植被信息识别及基于最佳小波包基系数的重金属污染探测具有可行性与一定优越性。     

签到位置数据的密度峰值快速搜索与聚类方法

位置签到数据蕴含了城市居民活动变化。由于客户端位置候选问题,不同的签到行为以同一候选位置签到时会产生位置重复现象。针对现有密度聚类方法在签到数据聚类上存在的问题,以快速搜索和查找密度峰值聚类算法(CFSFDP)为基础,提出了签到位置数据的密度峰值快速搜索与聚类方法。首先,引入位置重复频率来表达签到位置重复,然后,对原始签到位置数据点统计位置重复频率并重新设计数据结构,以新的空间点要素为研究对象寻找密度峰值点;最后,构建了峰值点密度簇聚类算法,在点要素集聚类过程中考虑密度连通性来保证峰值密度簇的连续与完整。试验表明,所提出的聚类方法有效避免了重复度较高的离群位置对象选为峰值并聚类的情况,并具有良好的空间适应性。所提取的密度峰值点不仅可以用来表示热区的中心,还能够反映热区的集中趋势,进而可以帮助探索热区的动态变化情况。