UjiIndoorLoc: An indoor localization dataset

# Context

This data set is focused on WLAN fingerprint positioning technologies and methodologies (also know as WiFi Fingerprinting). It was the official database used in the IPIN2015 competition.

Many real world applications need to know the localization of a user in the world to provide their services. Therefore, automatic user localization has been a hot research topic in the last years. Automatic user localization consists of estimating the position of the user (latitude, longitude and altitude) by using an electronic device, usually a mobile phone. Outdoor localization problem can be solved very accurately thanks to the inclusion of GPS sensors into the mobile devices. However, indoor localization is still an open problem mainly due to the loss of GPS signal in indoor environments. Although, there are some indoor positioning technologies and methodologies, this database is focused on WLAN fingerprint-based ones (also know as WiFi Fingerprinting).

Although there are many papers in the literature trying to solve the indoor localization problem using a WLAN fingerprint-based method, there still exists one important drawback in this field which is the lack of a common database for comparison purposes. So, UJIIndoorLoc database is presented to overcome this gap.

The UJIIndoorLoc database covers three buildings of Universitat Jaume I ([http://www.uji.es][1]) with 4 or more floors and almost 110.000m2. It can be used for classification, e.g. actual building and floor identification, or regression, e.g. actual longitude and latitude estimation. It was created in 2013 by means of more than 20 different users and 25 Android devices. The database consists of 19937 training/reference records (trainingData.csv file) and 1111 validation/test records (validationData.csv file)

The 529 attributes contain the WiFi fingerprint, the coordinates where it was taken, and other useful information.

Each WiFi fingerprint can be characterized by the detected Wireless Access Points (WAPs) and the corresponding Received Signal Strength Intensity (RSSI). The intensity values are represented as negative integer values ranging -104dBm (extremely poor signal) to 0dbM. The positive value 100 is used to denote when a WAP was not detected. During the database creation, 520 different WAPs were detected. Thus, the WiFi fingerprint is composed by 520 intensity values.

Then the coordinates (latitude, longitude, floor) and Building ID are provided as the attributes to be predicted.

The particular space (offices, labs, etc.) and the relative position (inside/outside the space) where the capture was taken have been recorded. Outside means that the capture was taken in front of the door of the space.

Information about who (user), how (android device & version) and when (timestamp) WiFi capture was taken is also recorded.

# Content

- Attributes 001 to 520 (WAP001-WAP520): Intensity value for WAP001.
Negative integer values from -104 to 0 and +100. Positive value 100
used if WAP001 was not detected.

- Attribute 521 (Longitude): Longitude. Negative real values from
-7695.9387549299299000 to -7299.786516730871000

- Attribute 522 (Latitude): Latitude. Positive real values from
4864745.7450159714 to 4865017.3646842018.

- Attribute 523 (Floor): Altitude in floors inside the building.
Integer values from 0 to 4.

- Attribute 524 (BuildingID): ID to identify the building. Measures
were taken in three different buildings. Categorical integer values
from 0 to 2.

- Attribute 525 (SpaceID): Internal ID number to identify the Space (office, corridor, classroom) where the capture was taken. Categorical integer values.

- Attribute 526 (RelativePosition): Relative position with respect to the Space (1 - Inside, 2 - Outside in Front of the door). Categorical integer values.

- Attribute 527 (UserID): User identifier (see below). Categorical integer values.

- Attribute 528 (PhoneID): Android device identifier (see below). Categorical integer values.

- Attribute 529 (Timestamp): UNIX Time when the capture was taken. Integer value.

# Relevent Paper

More information can be found in this paper:

Joaquín Torres-Sospedra, Raúl Montoliu, Adolfo Martínez-Usó, Tomar J. Arnau, Joan P. Avariento, Mauri Benedito-Bordonau, Joaquín Huerta. UJIIndoorLoc: A New Multi-building and Multi-floor Database for WLAN Fingerprint-based Indoor Localization Problems. In Proceedings of the Fifth International Conference on Indoor Positioning and Indoor Navigation, 2014.

Available at: [http://www.ipin2014.org/wp/pdf/4A-3.pdf][2]

If your are going to use this dataset in your research, please cite this paper

# Acknowledgements

The dataset was created by:

Joaquín Torres-Sospedra, Raul Montoliu, Adolfo Martínez-Usó, Tomar J. Arnau, Joan P. Avariento, Mauri Benedito-Bordonau, Joaquín Huerta, Yasmina Andreu, óscar Belmonte, Vicent Castelló, Irene Garcia-Martí, Diego Gargallo, Carlos Gonzalez, Nadal Francisco, Josep López, Ruben Martínez, Roberto Mediero, Javier Ortells, Nacho Piqueras, Ianisse Quizán, David Rambla, Luis E. Rodríguez, Eva Salvador Balaguer, Ana Sanchís, Carlos Serra, and Sergi Trilles.

# Inspiration

The objective is to estimate the building, floor and coordinates (latitude and longitude) of the 1111 samples included in the validation set. Since the real values of the building, floor and coordinates are also included, it is posible to determine the localization error.

The formula used in the IPIN2015 competition was the mean of the localization error of each sample. The localization error of each sample can be estimated as follows:

Error = building_penality * building_error + floor_penality * floor_error + coordinates_error

where:

- building_error is 1 if the estimated building is not equal to the
real one. 0 otherwise
- floor_error is 1 if the estimated floor is not
equal to the real one. 0 otherwise
- coordinates_error is sqrt( (estimated_latitude - real_latitude)^2 + (estimated_longitude-real_longitude)^2)

In the IPIN2015 competition building_penalty and floor_penalty where set to 50 and 4 meters, respectively.

[1]: http://www.uji.es
[2]: http://www.ipin2014.org/wp/pdf/4A-3.pdf

本数据集专注于无线局域网指纹定位技术与方法（亦称WiFi指纹定位）。该数据库为2015年IPIN竞赛的官方数据库。众多现实世界应用需确定用户在世界中的位置以提供相应服务，因此，自动用户定位在近年来成为热门研究课题。借助移动设备中集成GPS传感器，室外定位问题可得到非常精确的解决。然而，室内定位问题依然是一个未解之谜，主要由于室内环境中GPS信号的缺失。尽管存在一些室内定位技术与方法，但本数据库专注于基于WLAN指纹的方法（亦称WiFi指纹定位）。尽管文献中存在许多尝试利用WLAN指纹定位方法解决室内定位问题的论文，但该领域仍存在一个重要缺陷，即缺乏用于比较的通用数据库。因此，UJIIndoorLoc数据库应运而生，以填补这一空白。UJIIndoorLoc数据库覆盖了Universitat Jaume I（[http://www.uji.es][1]）的三栋建筑，每栋建筑拥有4层以上，总面积近110,000平方米。该数据库可用于分类，例如实际建筑和楼层识别，或回归，例如实际经纬度估计。UJIIndoorLoc数据库由2013年超过20名用户和25部Android设备创建，包含19937条训练/参考记录（trainingData.csv文件）和1111条验证/测试记录（validationData.csv文件）。529个属性包含WiFi指纹、采集坐标以及其他有用信息。每个WiFi指纹可由检测到的无线接入点（WAPs）和相应的接收信号强度指示（RSSI）来表征。强度值以负整数形式表示，范围从-104dBm（信号极弱）到0dBm。正值100用于表示WAP未被检测到。在数据库创建过程中，共检测到520个不同的WAPs。因此，WiFi指纹由520个强度值组成。同时提供了坐标（纬度、经度、楼层）和建筑ID作为预测属性。记录了采集的特定空间（办公室、实验室等）以及相对位置（空间内/外）。空间外表示采集是在空间门口前方进行的。同时记录了有关谁（用户）、如何（Android设备及版本）以及何时（时间戳）进行WiFi捕获的信息。

- 属性001至520（WAP001-WAP520）：WAP001的强度值。负整数，范围从-104到0，正值100用于表示WAP001未被检测到。

- 属性521（经度）：经度。负实数值，范围从-7695.9387549299299000到-7299.786516730871000。

- 属性522（纬度）：纬度。正实数值，范围从4864745.7450159714到4865017.3646842018。

- 属性523（楼层）：建筑内部楼层高度。整数值，范围从0到4。

- 属性524（BuildingID）：用于识别建筑的ID。在三个不同的建筑中进行测量。分类整数值，范围从0到2。

- 属性525（SpaceID）：用于识别采集空间的内部ID号码（办公室、走廊、教室等）。分类整数值。

- 属性526（RelativePosition）：相对于空间的相对位置（1 - 内部，2 - 在空间门口前方外部）。分类整数值。

- 属性527（UserID）：用户标识符（见下文）。分类整数值。

- 属性528（PhoneID）：Android设备标识符（见下文）。分类整数值。

- 属性529（Timestamp）：采集时的UNIX时间戳。整数值。

# 相关论文

更多信息可参见以下论文：
Joaquín Torres-Sospedra, Raúl Montoliu, Adolfo Martínez-Usó, Tomar J. Arnau, Joan P. Avariento, Mauri Benedito-Bordonau, Joaquín Huerta. UJIIndoorLoc: A New Multi-building and Multi-floor Database for WLAN Fingerprint-based Indoor Localization Problems. In Proceedings of the Fifth International Conference on Indoor Positioning and Indoor Navigation, 2014.

可在以下链接获取：[http://www.ipin2014.org/wp/pdf/4A-3.pdf][2]

如需在研究中使用此数据集，请引用该论文。

# 致谢

数据集由以下人员创建：
Joaquín Torres-Sospedra, Raul Montoliu, Adolfo Martínez-Usó, Tomar J. Arnau, Joan P. Avariento, Mauri Benedito-Bordonau, Joaquín Huerta, Yasmina Andreu, óscar Belmonte, Vicent Castelló, Irene Garcia-Martí, Diego Gargallo, Carlos Gonzalez, Nadal Francisco, Josep López, Ruben Martínez, Roberto Mediero, Javier Ortells, Nacho Piqueras, Ianisse Quizán, David Rambla, Luis E. Rodríguez, Eva Salvador Balaguer, Ana Sanchís, Carlos Serra, 和 Sergi Trilles。

# 激发灵感

目标是估计验证集中包含的1111个样本的建筑、楼层和坐标（纬度和经度）。由于建筑、楼层和坐标的真实值也包含在内，因此可以确定定位误差。在IPIN2015竞赛中使用的公式是每个样本定位误差的平均值。每个样本的定位误差可以估计如下：

误差 = 建筑惩罚 * 建筑误差 + 楼层惩罚 * 楼层误差 + 坐标误差

其中：

- 建筑误差为1，如果估计的建筑不等于真实建筑。否则为0。

- 楼层误差为1，如果估计的楼层不等于真实楼层。否则为0。

- 坐标误差 = sqrt((估计纬度 - 真实纬度)^2 + (估计经度 - 真实经度)^2)

在IPIN2015竞赛中，建筑惩罚和楼层惩罚分别设置为50和4米。

[1]: http://www.uji.es
[2]: http://www.ipin2014.org/wp/pdf/4A-3.pdf