Abstract: Indoor positioning is one of the key problems of the Location Based Service（LBS） research. WIFI is widely used in indoor positioning systems because of its cheap price and broad coverage. However, the Received Signal Strength Index（RSSI） of WIFI is heavily affected by the complex indoor environment which leads to a large range of spatial and temporal variance. In order to solve this problem, Channel State Information（CSI） is applied in our novel indoor positioning system. The CSI information can be used to regress the distance between the transmitter and the receiver. The distance of the receiver to multiple different transmitters can be utilized to get the receiver’s coordinate, which is the positioning result. The experiment results show that this new method has a higher positioning accuracy compared with the RSSI method.

Key words: indoor positioning, Channel State Information（CSI）, distance estimation, least square method

摘要： 室内定位是基于位置服务研究的关键问题，WIFI作为一种廉价易得的信号源被广泛应用在室内定位系统中。但是室内复杂环境造成同一点处的WIFI接收信号强度随时间变化范围较大，影响了基于WIFI接收信号强度的室内定位系统的性能。为了克服这个问题，提出了基于信道状态信息的室内定位方法。利用信道状态信息有效值非线性回归得出测量点与信号接入点之间的距离。利用测量点与多个不同信号接入点之间的距离回归出最后的定位结果。实验结果表明，相比典型的基于WIFI信号接收强度对数衰减模型的定位方法，方法有效提高了定位结果的准确度。

关键词: 室内定位, 信道状态信息, 距离估计, 最小二乘法