Distance and Angle Measurements for Hybrid Localization in IEEE 802.15.4 TSCH Networks

Overview The dataset includes distance and angle measurements obtained from a hybrid localization system integrated with the IEEE 802.15.4 Time Slotted Channel Hopping (TSCH) protocol [1]. The system combines multi-carrier phase difference (MCPD) or phase-based ranging (PBR) for distance estimation [2] with phase-based direction finding technique for angle-of-arrival (AoA) estimation [3], enabling hybrid localization using a single anchor node in an IoT network. Setup The measurements were collected in an office environment with a IEEE 802.15.4 TSCH network formed by six low-cost devices. One node, positioned in the center of a 5.1 m by 6.5 m room, acted as the root and performed localization of five surrounding nodes. Each device was built upon the AT86RF215 transceiver, which provides access to the phase of the received signal. To enable angle estimation, the root node was equipped with an RF switch and a uniform circular antenna (UCA) array composed of eight dipole elements spaced equally across 360 degrees. The pictures of the setup, used equipment and configuration are attached in the dataset files along with detailed description. Measurement Procedure The root of the node first formed a TSCH network. When all devices joined the network, the root started acting as an initiator, pinging its child devices. Each packet transmission initiated a Phase Measurement Process (PMP) that acquired phase measurements for angle and distance estimation [1]. During each PMP, the initiator captured eight reference phase samples from the first antenna to estimate and correct the carrier frequency offset (CFO), followed by ninety phase samples across the antenna array for AoA estimation. The switching between antenna elements was performed in a Round Robin (RR) pattern. The sampling rate of the system was 8 us per phase sample [3]. For distance estimation, fifteen frequency samples were measured sequentially using the Golomb frequency set to minimize measurement time and spectral overlap [2]. Software Implementation The software implementation of the modified Contiki-NG stack that supports hybrid localization alongside TSCH communication is available at: https://github.com/9morano/contiki-ng/tree/vesna-atasw Dataset Structure The dataset is stored in a single JSON file named `data.json`. It is organized by device, where each entry includes the device address and an array of records. Each record contains the absolute slot number (ASN) corresponding to the TSCH schedule, two arrays of phase samples representing measurements from the initiator and the reflector, an array of eight reference phase values used for frequency offset calibration, and a second array of ninety phase values used for angle estimation. The JSON file can be parsed using Python built-in json module.### 概述 本数据集包含融合了IEEE 802.15.4时隙信道跳变（Time Slotted Channel Hopping, TSCH）协议[1]的混合定位系统所获取的距离与角度测量数据。该系统将用于距离估计的多载波相位差（Multi-Carrier Phase Difference, MCPD）或基于相位的测距（Phase-Based Ranging, PBR）技术[2]，与用于到达角（Angle-of-Arrival, AoA）估计的相位测向技术[3]相结合，可在物联网（Internet of Things, IoT）网络中依托单个锚节点实现混合定位。 ### 实验配置 本次测量在办公环境中开展，所用的IEEE 802.15.4 TSCH网络由6台低成本设备搭建而成。其中一台节点部署于5.1m×6.5m房间的中心，作为根节点并对周围5个节点执行定位任务。所有设备均基于AT86RF215收发器构建，该器件可获取接收信号的相位信息。为实现角度估计，根节点配备了射频（Radio Frequency, RF）开关与均匀圆形天线（Uniform Circular Antenna, UCA）阵列，该阵列由8个偶极子天线单元组成，在360度范围内均匀分布。实验部署场景、所用设备与配置参数的细节说明及相关图片均随数据集文件一并提供。 ### 测量流程 根节点首先构建TSCH网络。待所有设备接入网络后，根节点将作为发起端，向其子节点发送ping报文。每一次数据包传输都会触发相位测量过程（Phase Measurement Process, PMP），该过程将采集相位数据以用于角度与距离估计[1]。在每一次PMP中，发起端首先从第一天线采集8个参考相位样本，用于估计并校正载波频率偏移（Carrier Frequency Offset, CFO）；随后在天线阵列中采集90个相位样本，用于到达角（AoA）估计。天线单元间的切换采用轮询（Round Robin, RR）模式。系统的采样率为每相位样本8微秒[3]。针对距离估计，本研究采用戈隆布（Golomb）频率集依次采集15个频率样本，以尽可能缩短测量时长并降低频谱重叠[2]。 ### 软件实现 支持混合定位与TSCH通信的改进型Contiki-NG协议栈软件实现可通过以下链接获取：https://github.com/9morano/contiki-ng/tree/vesna-atasw ### 数据集结构 本数据集存储于名为`data.json`的单个JSON文件中。数据集按设备进行组织，每个条目包含设备地址与一组记录数组。每条记录均包含对应TSCH调度的绝对时隙号（Absolute Slot Number, ASN）、分别代表发起端与反射端测量结果的两组相位样本数组、用于频率偏移校准的8个参考相位值数组，以及用于角度估计的90个相位值数组。该JSON文件可通过Python内置的json模块进行解析。