WLAN下移动定位的研究与实现-电子与通信工程专业论文.docxVIP

万方数据 万方数据 摘 要 随着无线局域网技术的日趋成熟，人们对无线局域网下基于位置服务的需求 增长迅速，但是常用的 GPS 定位系统无法在室内环境下很好的工作，所以室内定 位技术成为了目前比较热门的研究方向。本文主要研究了无线局域网下基于 Wi-Fi 技术的移动定位，并在实际的室内场景下进行了定位与预测的简单实现。文中首 先介绍了两种位置指纹匹配算法和两种轨迹预测算法，紧接着介绍了基于 Wi-Fi 定位的轨迹预测流程，重点讲解了基于接收信号强度位置指纹匹配的室内定位技 术，并且对流程进行了深入分析，从离线阶段接收信号强度中存在的奇异值问题 和在线阶段用于位置评估的加权项选择问题出发，分别利用奇异值检测算法和相 似性传播聚类算法进行解决，紧接着把在线定位阶段分为粗定位和细定位两个部 分，然后在此基础上对现有的定位算法做了一些改进，并对改进后的定位流程进 行了仿真，验证了改进后流程的有效性。最后在实际的无线局域网环境下简单实 现了移动定位与轨迹预测，并用轨迹预测算法来限制位置评估的范围，增强移动 定位的性能，最终的测试结果很好的验证了无线局域网环境下基于位置服务的可 行性。 关键词：Wi-Fi 移动定位 位置指纹 奇异值检测 相似性传播聚类 轨迹预测 Abstract Along with the increasing maturity of the WLAN technology, peoples demand for WLAN location services is growing fast, but the commonly used GPS positioning system cannot work well in the indoor environment, so the indoor positioning technology has become more popular in the current research directions. This paper mainly studied the mobile location based on Wi-Fi technology in the WLAN environment, and implemented a simple indoor positioning and trajectory prediction in the actual indoor scenarios. This paper first introduces two fingerprinting matching algorithms and two trajectory prediction algorithms, followed by introduction of the Wi-Fi-based positioning trajectory prediction process, then focuses on the positioning technology based on received signal strength fingerprinting matching, and carries on the thorough analysis to the process. From the singular values existence problem of received signal strength in offline phase and the weighted item selection problem of position evaluation in online phase, respectively, this paper uses the singular value detection algorithm and affinity propagation clustering algorithm to solve, followed by the online phase of coarse positioning and fine positioning two parts, and then made some improvement on the basis of the existing localization algorithm. The simulation verifies the effectiveness of the improved localization process. Finally, we achieve a simple mobile location and trajectory pred