用于室内定位的地磁场和WLAN数据集，来自腕带和智能手机的数据集

室内定位是环境智能（AmI）研究界的一个关键课题。在这种情况下，最近可穿戴技术的进步，特别是内置传感器的智能手表和个人设备，如智能手机，被认为是使设想的智能环境（SE）范式具体化的突破口。 特别是，致力于室内定位的场景代表了一个需要解决的关键挑战。许多工作试图解决室内定位问题，但缺乏一个共同的数据集或框架来比较和评估解决方案，这在该领域是一个需要克服的巨大障碍。公共数据集的不可用性和不确定性阻碍了比较不同室内定位算法的可能性。这就构成了本文所述的拟议数据集的主要动机。 我们收集了Wi-Fi和地磁的ï¬ngerprints，以及在同一环境中进行的两次活动的惯性传感器数据。从用户佩戴的智能手表和智能手机中检索同步的数据，以创建和展示一个公开可用的数据集，是这项工作的目标。 Attribute Information: Pointsmapping.ods。 一个三列电子表格（ID,X,Y），以当地坐标进行点映射。 每个ID代表地图上一个独特的地方。X-Y坐标代表当地的坐标。对于每一个措施。 measure1(2)_timestamp_id.csv。 到达placeID的时间戳（Unixtime），离开placeID的时间戳（Unixtime），place ID标识符（0-324）。 measure1(2)_smartphone_sens.csv。根据measure1(2)_timestamp_id.csv，这个csv包含由智能手机检索的数据传感器。时间戳、加速度X、加速度Y、加速度Z、磁场X、磁场Y、磁场Z、Z-轴角度（方位角）、X-轴角度（俯仰）、Y-轴角度（滚动）、陀螺仪X、陀螺仪Y、陀螺仪Zmeasure1(2)_smartwatch_sens.csv。根据measure1(2)_timestamp_id.csv，这个csv包含智能手表检索到的数据传感器。时间戳、加速度X、加速度Y、加速度Z、磁场X、磁场Y、磁场Z、Z-轴角（方位角）、X-轴角（俯仰角）、Y-轴角（滚动角）、陀螺X、陀螺Y、陀螺Zmeasure1(2)_smartphone_wifi.csv。每行包含PlaceId（升序）和127列，其中有每个不同的RSSI级别。活动期间检索到的WAPs。并非所有的WAP都在每次扫描中被检测到。对于这些WAPs来说，实际的RSSI值是-100（dbm）。 Citation Request: Barsocchi, P., Crivello, A., La Rosa, D., & Palumbo, F. (2016, October). A multisource and multivariate dataset for indoor localization methods based on WLAN and geo-magnetic field fingerprinting. In Indoor Positioning and Indoor Navigation (IPIN), 2016 International Conference on (pp. 1-8). IEEE.

Indoor positioning is a critical topic in the Ambient Intelligence (AmI) research community. In this context, recent advances in wearable technology, particularly smartwatches and personal devices like smartphones with built-in sensors, are regarded as a breakthrough to materialize the envisioned Smart Environment (SE) paradigm. Specifically, scenarios focused on indoor positioning represent a key challenge that needs to be addressed. Numerous studies have attempted to solve the indoor positioning problem, but the lack of a shared dataset or framework for comparing and evaluating solutions remains a major barrier to progress in this field. The unavailability and inconsistency of public datasets hinder the ability to compare different indoor positioning algorithms. This constitutes the core motivation behind the proposed dataset introduced in this paper.

We collected Wi-Fi and geomagnetic fingerprints, as well as inertial sensor data from two distinct activities conducted in the same environment. The objective of this work is to acquire synchronized sensor data from smartwatches and smartphones worn by users, in order to develop and release a publicly available dataset.

Attribute Information:
1. Pointsmapping.ods: A three-column spreadsheet (ID, X, Y) for point mapping using local coordinates. Each ID represents a unique location on the map, and the X-Y coordinates correspond to the local coordinates for each measurement.
2. measure1(2)_timestamp_id.csv: Contains the arrival timestamp (Unix time) at a placeID, departure timestamp (Unix time) from the placeID, and the place ID identifier (ranging from 0 to 324).
3. measure1(2)_smartphone_sens.csv: Corresponding to measure1(2)_timestamp_id.csv, this CSV file contains sensor data collected by smartphones, including timestamp, acceleration X, acceleration Y, acceleration Z, magnetic field X, magnetic field Y, magnetic field Z, Z-axis angle (azimuth), X-axis angle (pitch), Y-axis angle (roll), gyroscope X, gyroscope Y, and gyroscope Z.
4. measure1(2)_smartwatch_sens.csv: Corresponding to measure1(2)_timestamp_id.csv, this CSV file contains sensor data collected by smartwatches, including timestamp, acceleration X, acceleration Y, acceleration Z, magnetic field X, magnetic field Y, magnetic field Z, Z-axis angle (azimuth), X-axis angle (pitch), Y-axis angle (roll), gyroscope X, gyroscope Y, and gyroscope Z.
5. measure1(2)_smartphone_wifi.csv: Each row contains PlaceId (in ascending order) and 127 columns corresponding to the RSSI levels of distinct WAPs detected during the activity. Not all WAPs are detected in every scan; for undetected WAPs, the actual RSSI value is set to -100 dBm.

Citation Request:
Barsocchi, P., Crivello, A., La Rosa, D., & Palumbo, F. (2016, October). A multisource and multivariate dataset for indoor localization methods based on WLAN and geomagnetic field fingerprinting. In *Indoor Positioning and Indoor Navigation (IPIN)*, 2016 International Conference on (pp. 1-8). IEEE.

搜集汇总

背景与挑战

背景概述

该数据集是一个用于室内定位研究的公开数据集，包含来自腕带和智能手机的地磁场、WLAN数据以及惯性传感器数据。数据集提供了丰富的时间戳、加速度、磁场、角度和陀螺仪数据，旨在解决室内定位算法比较和评估的公共数据集缺乏问题。