基于弱多径效应识别的室内定位方法研究

发布时间：2018-05-13 10:10

本文选题：室内定位 + 信道状态信息　；参考：《兰州交通大学》2017年硕士论文

【摘要】：无线局域网,尤其是Wi-Fi已经普遍部署,这也使得大量无线交互服务和普适的手机应用程序如雨后春笋般层出不穷。而室内定位就是这些关键应用中的一个。人们主要关心的是如何设计出一个适应环境的定位应用。这要求在各种复杂室内环境下构建的信号传播模型能够抵抗各种干扰,尤其是多径效应的影响,从而获得精确的距离估计。传统的基于信号接收强度的信号传播模型非常容易受到由于多径效应带来的时域和空间的信号波动。多径效应带来的影响是导致大多数室内定位系统出现距离估计误差的原因。凭直觉我们认为那些处于弱多径效应的点,也就是受多径效应干扰较小的点,在路径损耗传播模型下将会获得更精确的距离估计。因此,如何区分多径效应的强弱成为获取精确距离估计的关键。然而,目前商用Wi-Fi设备采集的MAC层的RSSI信号对多径效应的区分度不高,很难区分出多径效应的强弱。传统的室内定位解决方案总是要在定位性能和定位通用性上做出取舍。这两者是鱼与熊掌的关系,不可兼得。如果我们想要提高定位性能,那么定位通用性就会下降。大部分高精度的室内定位解决方案的通用性都很低。这是因为他们都是针对特定的环境,所以当他们应用于其他环境时,定位性能就会急剧下降。因此,对于室内定位而言,我们急需的是一个通用的精确的室内定位方案。本文提出了一个基于弱多径效应判别的通用的精确的室内定位方案。本文采用物理层的信道状态信息(CSI)来判断多径效应的强弱。我们分别从CSI信号的时域和频域中提取三个特征。然后我们用这三个特征的阈值来区分弱多径效应环境和强多径效应环境。基于弱多径效应识别,我们提出了一个通用的精确的室内定位方案。这一解决方案融合RSSI信号和CSI信号来进行定位。然后,我们用这一解决方案设计实现了一个室内定位系统。在我们的室内定位系统中,我们首先通过提取的CSI特征筛选那些出处于弱多径效应环境的点。然后,我们把这些点采集的RSSI信号代入预先构建的路径损耗传播模型,就算发射端和接收端之间的距离。最后,我们用定位算法计算接收端的位置。实验数据证明处于弱多径效应环境的点能够获得更精确的距离估计。因此,系统的定位性能,包括定位精度和定位准确度都很高。这样,我们获得了一个通用的精确的室内定位系统。
[Abstract]:Wireless LAN, especially Wi-Fi, has been widely deployed, which makes a large number of wireless interactive services and ubiquitous mobile applications springing up. And indoor positioning is one of these key applications. The main concern is how to design a positioning application that adapts to the environment. This requires a variety of complex rooms. The signal propagation model built in the internal environment can resist all kinds of interference, especially the influence of multipath effect, so as to obtain accurate distance estimation. The traditional signal propagation model based on the signal receiving intensity is very easy to be subjected to the time-domain and space signal fluctuation caused by the multipath effect. The influence of the multipath effect is big. Most of the indoor positioning systems have the cause of distance estimation error. By intuition, we think that those points in the weak multipath effect, that is, the point that is less interfered by the multipath effect, will get more accurate distance estimation under the path loss propagation model. Therefore, how to distinguish the strength of the multipath effect is the key to obtaining the precise distance estimation. However, at present, the RSSI signal of the MAC layer collected by commercial Wi-Fi equipment is not very distinguished from the multipath effect, and it is difficult to distinguish the strength of the multipath effect. The traditional indoor positioning solutions always make a trade-off between the positioning performance and the positioning universality. Bit performance, then positioning generality will decline. Most high-precision indoor positioning solutions are very low in generality. This is because they are specific to a specific environment, so when they are applied to other environments, the positioning performance will drop dramatically. So, for indoor positioning, we are in urgent need of a general precision. This paper proposes a general and accurate indoor location scheme based on the weak multipath effect discrimination. This paper uses the channel state information of the physical layer (CSI) to determine the strength of the multipath effect. We extract three features from the time domain and frequency domain of the CSI signal respectively. Then we use the threshold of the three features to come to the region. On the basis of weak multipath effect environment and strong multipath effect environment. Based on the weak multipath effect recognition, we propose a general and accurate indoor location scheme. This solution combines RSSI signal and CSI signal to locate. Then, we use this solution to design an indoor positioning system. In the system, we first select the points that are located in the weak multipath environment through the extracted CSI features. Then, we replace the RSSI signals collected by these points into a pre built path loss propagation model, even if the distance between the transmitter and the receiver. Finally, we use the location algorithm to calculate the location of the receiver. The points in the weak multipath environment can obtain more accurate distance estimation. Therefore, the positioning performance of the system, including the positioning accuracy and location accuracy, is very high. In this way, we have obtained a general and accurate indoor positioning system.

【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN92

【相似文献】