What type of data could sensors pull from your fitness wearable?

The fitness tracker can track heart rate, heart rate variability, sleep time and quality, fitness progression, elevation, temperature and location with the help of several device sensors and the inclusion of Global Positioning System (GPS) technology. To track movement in all directions. An accelerometer sensor takes inertial measurements of speed and position. Usually, on three axes, it can also detect the inclination, inclination and orientation of the body.

Naturally, this is very important for any fitness tracker, since this sensor will record most of the steps taken by the person. Fitness trackers are bracelets that measure heart rate, steps and calories. They can also monitor sleep patterns and suggest ways to improve your activity levels. The device includes several sensors that work together to record your activity levels.

It uses an accelerometer to count steps, a gyroscope to detect movement and a magnetometer to measure the heart rate (when used close to the skin). The data collected by these sensors is then sent wirelessly to an application on your smartphone or computer so you can see how active you've been over time. An official website of the United States government The. gov means it's official.

Federal government websites usually end up in. government or. thousand. Before sharing sensitive information, make sure you're on a federal government site.

Wearable wrist devices are being developed in a variety of ways, with differences in form, purpose, hardware features, and so on. It is possible to classify them into two categories based on their functionalities. On the one hand, there are simple devices with limited capacities, called smart bands, whose purpose is to quantify very specific characteristics such as the user's physical activity, sleep, etc. On the other hand, there are more advanced and general-purpose devices, known as smartwatches.

They include an integrated operating system that allows the installation of third-party applications and functionalities similar to those available on smartphones. This classification is key to carrying out any development or research in this area, since the possibilities of collecting data are very different. In general, smart bands do not provide solutions for transferring data to third-party systems. In addition to the main platforms, there are others that support specific devices such as Microsoft, Fitbit or Jawbone.

However, these platforms are aimed at storing and providing data collected from the specific device. As far as we know, to date, they haven't focused on supporting a homogenous solution for integrating data from portable devices from other vendors. To transfer data to the third-party server, several options may be available depending on the portable device. In some cases, handheld device vendors (p.

ex.,. In the case of smartphones, tablets and PCs, SDKs are available for developing applications that collect data directly from linked smartwatches, from the smartphone itself or from its own warehouse. With these SDKs, third parties can develop specific software to collect data on portable devices and send it to other systems. Patented warehouses and cloud services often provide a REST API.

This REST API allows third parties to access user data. Warehouse data transfer has some disadvantages compared to portable data transfer. The main one is related to the collection of data in real time. In the case of the transfer of data from the warehouse, the transfer of data from the portable device to the proprietary warehouse takes place sporadically at random times.

Sometimes the time between transfers can be on the order of several days. Conversely, in the case of portable data transfers, this can occur at specific time intervals. A second drawback of transferring warehouse data relates to the nature of the data. Usually, when data is transferred to the owner warehouse, some type of processing takes place (p.

ex. In the case of portable data transfer, this is generally raw data. For example, in general, the nominal values of the accelerometer cannot be collected from a warehouse. Taking into account these two transfers and the access options, four different configurations can be distinguished, as explained in the following subsections.

This option requires a lot of effort in software development. In particular, it is necessary to develop a specific application that runs on the smartphone, usually taking advantage of the available SDK. This application has to manage the subscription to events on the portable device and the problems related to possible interruptions in the Internet connection. In addition, for some portable devices, we also need to develop an application for the portable device to record sensor data and transfer it to the smartphone application.

This is the case with Android Wear wearables. Data transfer from a portable device and direct access between the portable device and the third-party server are possible in theory, but not in practice today. Figure 8 shows the configuration of link 7. The intermediate application is removed from the smartphone and a direct link is established between the portable device and the third-party server.

This could be compatible with some systems, for example, the new Android Wear portable devices allow the inclusion of a SIM card or the use of Wi-Fi access points (p. However, today, this option is only available to connect the portable device to a mobile device (p. In addition, the energy consumption is too high. Therefore, this configuration is not yet a valid option.

Fitbit has some special features that make it an interesting device. You can store data for several days without loss because your energy needs are very low. The problem is that, once a certain level of storage has been reached, a summary is made and the raw data is discarded. This can be considered a disadvantage, but also a positive feature, since the device stores the most relevant information on its own.

Sensors available on portable devices (note 1: this data cannot be accessed). Sleep analysis performed by Fitbit based on Fitbit Surge data. The LG Watch R is based on the Android Wear operating system. As a result, it's compatible with any Android device, especially smartphones.

To enable the connection between both devices, the Android Wear app must be installed on the Android device. Sleep analysis performed by Microsoft through Microsoft Band. This wearable is very basic and simple. Its framework focuses on reducing energy consumption.

As a result, your non-rechargeable battery lasts approximately 6 months. There are some issues to consider related to sleep indicators (cf. The portable device does not capture the onset of sleep automatically. The user must press a button on the device to indicate the start.

It is used to switch between phases of rest and physical activity. Alternatively, using a smartphone application, the user can add the sleep period to the warehouse. Analysis of the dream performed by Jawbone using the Jawbone Up movement. As explained in section 3.2, there are different ways in which data from portable sensors can be transferred to our server.

In this section, we explain the options adopted on the selected devices. A software architecture has been developed to perform this transfer and process the data to obtain the desired indicators (cf. This architecture includes several modules, each with a specific function that is described in the following sections. Using the available transfer modes, the selected portable devices provide us with the data to calculate sleep and stress indicators.

However, before data analysis can be performed, it is necessary to address the differences between the data that comes from devices from different vendors. The architecture of our analytical engine shown in Figure 12 is responsible for the homogenization, storage and analysis of data. The architecture is comprised of several modules. The first is the task scheduler, which is responsible for scheduled tasks, such as collecting data in its own warehouses and periodically calculating sleep and stress indicators.

Another module is the access layer, which is used as an entry point for data sent by smartphones and PCs or collected in own warehouses. Once the data is in the analytical engine, the Data Homogenizer module prepares and stores the data to allow it to be processed in a homogeneous way. The latest modules of the architecture, located on the right side of the figure, involve data analysis. A basic calculation module is included to perform basic operations, such as correlation calculations and statistical calculations.

The business calculation module includes machine learning techniques to detect routines and outliers. The results obtained from the analyses are stored in the database. This data is provided to end users through a control panel, as part of recommendations, etc. To support this communication, a REST API is included that allows you to receive and send information.

In the case of real-time services, data is provided directly from the calculation modules. We can only detect two sleep states on Fitbit wearables. We decided to adopt the Microsoft identification, but marking the longest sleep period as the main period and the other periods as naps. The sleep period if a user sleeps from day 6 to day 7 is assigned to day 6, even if the user starts sleeping after midnight.

Many systems typically provide daily summaries of different variables. To do this, in the case of Microsoft, we calculate the differences between the values at the beginning and at the end of the day. In the case of Android Wear, we developed a strategy based on watch data. All changes detected in the variables are recorded.

When the clock is reset, the last available value for the variables is retrieved and taken into account when calculating the daily summaries. The IoT vision involves connecting all types of physical devices to the virtual world, supporting their communication with existing Internet entities. There would no longer be only people or software systems in the virtual world, but also a myriad of devices that could be addressed, identified, located, detected, activated and, in general, interact through information and communication technologies. Francisco de Arriba, Manuel Caeiro and Juan Manuel Santos conceived the idea, designed the use cases, analyzed the results and proposed the conclusions.

Francisco de Arriba and Juan Manuel Santos conducted the studies. Francisco de Arriba and Manuel Caeiro Rodríguez reviewed the state of the art and wrote the article. This cardiovascular fitness score is a measure of your cardiovascular status based on your VO2 max (the maximum amount of oxygen your body can use when you exercise at maximum intensity) and can be found in the heart rate section of the Fitbit app. Fortunately, this metric is another basic statistic for fitness trackers, so you should find it in just about every option that appears on your shopping comparison list.


Amie Atanacio
Amie Atanacio

Evil coffee enthusiast. Professional beer ninja. Devoted social media aficionado. Proud pizza advocate. Typical internet nerd.