基于滤波方法的无需测距室内定位研究

发布时间：2018-01-26 09:14

本文关键词： 室内定位无需测距指纹匹配滤波算法　出处：《北京工业大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着物联网、移动互联等新概念新技术的普及与发展,室内定位技术的特殊地位和重要性日益彰显。在商场、社区、医院等城市建筑的室内环境中,目标位置是不可或缺的语境信息。对于移动互联领域的实际应用而言,室内定位技术的精度性能决定着上层服务的质量。关于室内定位技术的研究对物联网、移动互联的发展起到了直接的推动作用。综上所述,研究室内定位技术具有重要的理论意义和应用价值。室内定位技术可以分为两类,基于测距的与无需测距的。基于测距的室内定位技术精度较高,但由于需要满足视距传播和特定设备等要求,因而不易于推广应用。无需测距的室内定位技术大多使用信号强度作为定位依据,但由于无线信号容易受到干扰,其精度性能不稳定。若能充分利用环境信息和移动目标的轨迹信息,则可以降低无需测距技术的室内定位误差。针对无需测距的室内定位问题,论文的主要工作和研究成果如下:1.基于多源信息感知的思想,论文定义了室内环境的多源信息模型,并使用典型的指纹匹配定位算法进行了验证。实验证明基于上述多源信息模型进行匹配定位可以较好地降低定位误差。2.鉴于室内定位目标的移动轨迹所具有的连续性特征,论文研究了滤波方法应用在室内移动追踪中的关键问题,系统归纳并提出了状态转移模型的设计方法。基于仿真实验针对移动追踪问题优化了噪声模型的设计方案。3.在上述研究的基础上,论文设计并构建了适用于室内移动目标追踪的原型系统。该系统采用多源信息模型并以轨迹信息作为贝叶斯条件进行滤波,实现了室内移动目标的追踪。通过真实场景中的移动目标追踪实验证明,相比无滤波的原型系统,该系统进一步降低了定位误差。
[Abstract]:With the popularization and development of Internet of things, mobile interconnection and other new concepts and new technologies, the special status and importance of indoor positioning technology are increasingly evident in the indoor environment of city buildings such as shopping malls, communities, hospitals and so on. Target location is an indispensable contextual information. For the practical application of mobile interconnection, the precision performance of indoor positioning technology determines the quality of upper layer service. The development of mobile interconnection has played a direct role in promoting. To sum up, the research of indoor positioning technology has important theoretical significance and application value. Indoor positioning technology can be divided into two categories. The precision of indoor location technology based on ranging is higher than that without ranging, but it needs to meet the requirements of line-of-sight propagation and special equipment. Therefore, it is difficult to popularize the application. The indoor positioning technology without ranging mostly uses the signal strength as the basis of location, but the wireless signal is vulnerable to interference. If we can make full use of the environment information and the track information of moving targets, we can reduce the indoor positioning error without ranging technology. The main work and research results are as follows: 1. Based on the idea of multi-source information perception, the paper defines a multi-source information model for indoor environment. The experimental results show that the matching location based on the above multi-source information model can reduce the location error. 2. In view of the mobile trajectory of the indoor positioning target, the algorithm is verified by the typical fingerprint matching algorithm. The continuity of the. In this paper, the key problems of the application of filtering method in indoor mobile tracking are studied. The design method of the state transition model is summarized and put forward systematically. Based on the simulation experiment, the design scheme of the noise model is optimized. 3. On the basis of the above research. A prototype system for indoor moving target tracking is designed and constructed in this paper. The system adopts multi-source information model and takes trajectory information as Bayesian condition for filtering. The tracking experiment of moving target in real scene shows that compared with the prototype system without filtering, the system can further reduce the positioning error.
【学位授予单位】：北京工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN713

【相似文献】