近年来，无线定位技术发展得如火如荼，被广泛应用于人员和物体的定位和导航，成为物联网的基础性服务。由于高带宽的物理特性，超宽带（ultra-bandwidth, UWB）技术具有抗多径能力强、辐射功耗低、覆盖范围大等优点，基于超宽带的定位系统受到国内外研究人员的广泛关注。现有的超宽带定位系统通常采用基于到达时间的定位方法或基于到达时间差的定位方法，这两种方法各有优劣。其中基于到达时间的定位方法首先计算出定位设备到基站的距离，再根据设备与多个基站的几何关系求解出设备的位置。为了保证定位精度，该方法需要定位设备与基站进行多次通信，从而准确推断出两者间的距离。因此大量设备同时定位难免会产生信号冲突，极大地影响定位效率。基于到达时间差的定位方法计算的是定位设备到不同基站的距离差，再利用几何关系求解出定位设备的位置，具有支持大量设备同时定位的特点，但其定位精度受制于基站时间同步的效果。在实际应用中，时间同步实现复杂且效果不理想，导致该方法定位精度不高。本文提出一种基于超宽带UWB的高精度可扩展定位系统，它同时具备精度高和扩展性好的优点。系统的核心思路是利用基于侦听模式的定位算法进行定位，定位设备在定位过程中只需处在侦听模式，不需要发送任何消息，因此定位设备彼此间并不会产生影响，从而实现了支持无限数量设备同时定位的效果。在系统实现方面，基站间发送测距消息，定位设备利用侦听到的基站消息，转换并计算出其与不同基站的距离差，再利用几何关系以及改进的最小二乘法算法求解出定位设备的位置。本文在终端UWB设备、边缘端树莓派和云端服务器实现并部署了高精度可扩展定位系统，能够应用于实际生产生活。为了验证高精度可扩展定位系统的可靠性和稳定性，本文在停车场、空教室、实验室等多种室内外场景对定位系统做了全面的性能评估。实验结果显示，本定位系统的平均误差是10.5cm，且目标跟踪效果十分精确。由此可见，基于超宽带UWB的高精度可扩展定位系统能够被广泛应用于室内外定位场景。
In recent years, wireless positioning has developed rapidly and has been widely used in the positioning and navigation of people and objects. It becomes a basic service for the Internet of Things. Due to the high bandwidth feature, ultra-bandwidth (UWB) technology has the advantages of strong anti-multipath capability, low power consumption, and large coverage. Therefore, localization system based on UWB is getting more and more popular.The existing UWB localization systems usually use two methods, namely Time of Arrival (TOA) and Time Difference of Arrival (TDOA). The positioning method based on TOA first calculates the distance between the device and anchors. It calculates the device's location according to the geometric relationship between the device and anchors. This method requires multiple exchange messages between the device and the anchors to infer the distance to ensure accuracy. The simultaneous positioning of many devices results in signal conflicts, which greatly affects the positioning efficiency. The positioning method based on TDOA calculates the distance difference between the device and the anchors, it uses the geometric relationship to find the device's position. It has the characteristics of supporting the simultaneous positioning of many devices, but its positioning accuracy is limited by time synchronization. Moreover, the implementation of time synchronization is complicated and the effect is not ideal, which leads to low positioning accuracy.This paper presents a high precision and scalable localization system based on UWB. The idea of the localization system is to use the positioning algorithm based on overhearing. The device only needs to receive messages during the positioning, and it does not need to send any messages. Thus, the devices will not affect each other, and an unlimited number of devices can be positioned simultaneously. In terms of system implementation, the anchors send exchange messages to each other. The device receives messages to convert and calculate the distance difference between the device and the anchors. It then uses the geometric relationship and the improved least square algorithm to find the device's location. In this paper, the high precision and scalable localization system has been implemented and deployed on UWB devices, Raspberry Pi, and cloud servers. It can be applied in actual production and daily life.To verify the reliability and stability of the system, this paper evaluates the performance of the localization system in various indoor and outdoor scenes, such as parking lot, classroom, laboratory, etc. The experiment shows that the localization error is 10.5cm, and it has an accurate object tracking effect. Therefore, the high precision and scalable localization system based on UWB can be widely used in indoor and outdoor positioning scenes.
