This article goes under-the-hood to explain how GPS technology works and the settings and situations that affect its precision.
The data collected from GPS devices is used by sports science practitioners to quantify the movement demands of their athletes. This information enables the quantification of external training loads across training and competition, as well as different contextual factors (e.g., playing position, individual athlete, technical-tactical dynamics).
Such information is used by key stakeholders (i.e. technical coaches, fitness/strength and conditioning coaches, medical practitioners) to plan and evaluate their sessions. Yet, it is critical to understand how the technology collects this data, in order to best understand its accuracy and limitations.
How Does GPS Technology Work?
Global Positioning Systems (GPS) utilise satellite-based navigation systems to track the position of individuals or items. Initially developed by the United States Department of Defense for military use, GPS has since been validated for use in sport.
Athletes wear GPS devices, often in a custom-fitted vest or pouch between the shoulder blades, to capture and transmit data regarding displacement, velocity, and acceleration.
When outside, GPS devices receive accurate time data from atomic clocks via radio signals transmitted by satellites and travelling at the speed of light. By calculating the time differential between the satellite and GPS receiver, the system determines the signal travel time, enabling the determination of the distance from each satellite to the receiver.
Once the device's position is established, distance travelled is calculated by positional differentiation: the change in device location with each received satellite signal (Varley et al., 2017).
Although this distance could then be used to derive velocity, a more accurate velocity with lower error can be calculated via the Doppler-shift method, which uses the change in frequency of satellite signal (Townshend et al., 2008). Acceleration is then subsequently derived from velocity.
How Does GPS Technology Differ from other Tracking Technologies?
Sometimes the term “GPS” is used ubiquitously for tracking technologies. However, not all tracking technologies utilise GPS and, as I recently discussed in my essential guide to GPS monitoring for Sportsmith, it is important to understand the differences to appreciate the relative advantages and disadvantages of each.
As well as GPS, these devices, sometimes referred to collectively as micro-electromechanical systems (MEMS), also contain inertial measurement units (IMU). The IMU typically consists of an accelerometer, gyroscope, and magnetometer:
Accelerometer: most commonly a piezo-electrical tri-axial accelerometer. This sensor quantifies acceleration in G-force (where 1 G = 9.81 m/s/s) across all three planes (x, y, and z) hence “tri-axial”.
Gyroscope: uses gravity to determine angular velocity, detecting tilt and rotation of the device.
Magnetometer: measures the orientation with the Earth using the magnetic pull of the ‘true north’.
Taken together, data from these sensors is combined to provide the IMU-related outputs. Remember, these are different to GPS that utilises satellites and therefore, can work indoors.
Beyond these MEMS devices, optical tracking (OT), local positioning systems (LPS) that can include radio frequency information (RFID) or Ultra-wideband (UWB), and light detection and ranging (LiDAR) technologies also exist. A brief overview of these technologies is shown below in Table 1.
Table 1. Comparison of Characteristics Across Different Tracking Systems (adapted from Torres-Ronda et al., 2022a)
| GPS | IMU | OT | LPS | LIDAR |
Hardware | The travel time of radio signals transmitted by satellites to GPS devices.
| Multiple microsensors detect movement on the body upon which they are placed. | Camera system around a playing environment tracks coordinates. | Anchors placed around the environment triangulates radio signals with tags. | A device placed by the paying environment discharges short laser pulses. |
Measures | Displacement Velocity via Doppler-shift Distance and acceleration derived from velocity. | Accelerometer = acceleration
Gyroscope = angular velocity
Magnetometer = orientation | Trajectories of moving objects from which distance, speed, and acceleration are derived. | Reception time of signal used to determine location from which distance, speed, and acceleration are derived. | Records time taken for laser pulses to bounce off objects and return, from which speed and acceleration calculated. |
Dimension | 2D | 3D Triaxial | 3D | 3D | 3D |
Setting | Outdoor | Indoor and outdoor | Indoor and outdoor | Indoor and outdoor | Indoor and outdoor |
Factors that Influence GPS Quality
It is vital to appreciate the factors that influence the precision of GPS data collected. Firstly, data quality in GPS is largely influenced by signal quality, which is determined by the quantity and geometrical organisation of the satellites interacting with the device.
Structural interferences, such as stadium obstructions, can affect the line of sight and signal transmission, as GPS signal will not travel through metal or concrete. While 4 satellites are required as a minimum to triangulate the data, 6 satellites and above are recommended (Malone et al., 2017).
Manufacturers have attempted to improve data quality through the integration of both the United States GPS and the Russian GLObal NAvigation Satellite System (GLONASS), which together doubles the number of available satellites.
In addition, the horizontal dilution of precision (HDOP) refers to the geometrical arrangement of satellites, whereby one satellite is positioned directly overhead and the others are evenly spaced, with less than 1.0 representing ideal positioning.
As described by James Malone and colleagues (2017): “When satellites are bunched together HDOP is high and precision is poor whereas when satellites are spread out HDOP is low and precision is good”. Subsequently, practitioners working with GPS are encouraged to verify and track satellite and HDOP information for each session.
Practitioners working with GPS are encouraged to verify and track satellite and HDOP information for each session.
Validity and Reliability of GPS Data
Research into the early GPS technology, with low frequency (i.e. 1 and 5 Hz) expressed concerns in the validity and reliability of the data. Even with technological developments since, research has still expressed caution regarding the data collected, such as Martin Buchheit and colleagues’ 2014 study: ‘Monitoring Accelerations With GPS in Football: Time to Slow Down?’. We've previously explored this research on the blog here: GPS Validity & Reliability – Exploring the Research.
It remains that GPS validity and reliability can be compromised by movements that require shorter durations, higher velocities and/or more complexity (Aughey, 2011). Of course, such movement demands may be the most vital in sport, both from a performance and injury perspective, but the least accurately captured!
Advancements in GPS technology include higher sampling frequencies, updated GPS motherboards and antennae, and more powerful microprocessors (Malone et al., 2017), which seek to enhance the data quality.
Interestingly, although higher sampling frequencies (the number of samples per second, measured in Hertz [Hz]) in theory should improve data quality, this is not necessarily the case. The 15 Hz units sampling rate is calculated by supplementing a 10 Hz GPS sampling rate with accelerometer data, but was shown to have lower validity and inter-unit reliability than the 10 Hz GPS device in one study (Johnston et al., 2014).
Additionally, manufacturers may release firmware updates, and although these are designed to enhance operations they can affect the data output (Varley et al., 2017). So, practitioners should be critical of when and how they make updates and keep a note of these versions.
Data Settings When Using GPS
There are a number of important software settings that warrant review when using GPS technology. Among these are filtering settings, the Minimum Effort Duration (MED), and threshold selection.
The MED, also referred to as the ‘dwell time’, specifies the minimum duration an athlete must exceed a certain threshold (velocity, acceleration, or deceleration) to classify as a distinct effort. This parameter plays a pivotal role in quantifying high-intensity activities such as sprints, accelerations, and decelerations.
A shorter MED may lead to the capture of erroneous spikes and an overestimation of efforts, while a longer MED results in an underestimation. Like many decisions, the selection of an appropriate MED often involves a nuanced blend of practical experience, knowledge of the sport and athlete population, and insights gleaned from research findings.
Similarly, practitioners must determine threshold values that demarcate different bands for velocity, acceleration, and deceleration. These thresholds can be absolute values (e.g., velocity thresholds in metres per second or miles per hour) or relative to individual characteristics (e.g., a percentage of maximum velocity).
The process of binning continuous data into discrete groups presents inherent challenges, which is only exacerbated by the lack of consensus in the literature regarding appropriate sport- and population-specific thresholds, so this is a challenging decision. For a deeper dive into MED and threshold selection, have a read of the Global Performance Insights post on 'Defining high-speed, acceleration, and deceleration efforts in sport'.
Filtering settings refer to the methods used to clean and process raw GPS data, in an attempt to reduce noise and improve data accuracy. Filtering techniques are typically chosen at the discretion of the manufacturer and may include moving average, median, and exponential filters (Malone et al., 2017).
While raw data export permits practitioners to apply custom filters outside the software, it's worth noting that data may undergo prefiltering by the manufacturer. Any changes to filtering settings, whether through manufacturer updates or internal customisation, should be approached intentionally due to the potentially significant impact on GPS outputs.
Day-to-day GPS Data Collection Workflow
Incorporating GPS tracking into daily operations entails a number of important considerations particularly before, during, and after a session to uphold precise and consistent data collection. As ‘Data Stewards’ entrusted with managing this system, we have a responsibility to prioritise the maximisation of data quality.
A workflow that promotes clean and consistent data collection ensures the best decisions can be made on the data. Although designing this workflow may initially feel overwhelming, it will soon become an automated routine.
The checklist below was originally published in our Tracking Systems in Team Sports: Back to Basics editorial on Sports Performance and Science Reports.
Getting Started with GPS Technology
At the onset of integrating GPS technology, it can be tempting to jump straight into the metrics, analysis, and implementation of the data. However, as scientists, we have a responsibility to “unpack the black box” (Malone et al., 2017) and understand their inner workings. This understanding empowers practitioners to make informed decisions regarding their settings and workflows. Only by comprehending the intricacies of the technology can we determine the most appropriate approach to GPS monitoring tailored to our specific sport and population.
Now with this understanding of the technology, we can turn our attention to the data itself. I recently discussed for Sportsmith how to get started with GPS technology analysis. Read the Essential Guide to GPS Monitoring here.
Frequently Asked Questions (FAQs)
What is GPS technology and how is it used in sports?
GPS (Global Positioning System) technology uses satellite signals to track the position and movement of objects or individuals. GPS devices receive time data from satellites and calculate the distance based on signal travel time. This information is used to determine the device's position and movement metrics such as distance, velocity, and acceleration. In sports, athletes wear GPS devices to capture data on their displacement, velocity, and acceleration, which is then used to optimize training and performance.
What factors can affect the accuracy of GPS data?
Factors include satellite signal quality, structural interferences (like stadium obstructions), and the number and arrangement of satellites in view. Most technology now uses both GPS and GLONASS systems to improve accuracy by increasing the number of available satellites.
How reliable is the data collected by GPS devices?
The reliability of GPS data can be influenced by device sampling rates, the complexity of movements, and technological updates. Higher sampling frequencies generally improve data quality, but it's crucial to be aware of firmware updates that might affect data output.
What settings should be considered when using GPS devices?
Important settings include filtering techniques, Minimum Effort Duration (MED; the minimum time an athlete must exceed a certain threshold to count as an effort), and threshold values for velocity, acceleration, and deceleration. These settings help ensure accurate and relevant data collection.
References
Aughey RJ. Applications of GPS technologies to field sports. International journal of sports physiology and performance. 2011 Sep 1;6(3):295-310.
Buchheit M, Al Haddad H, Simpson BM, Palazzi D, Bourdon PC, Di Salvo V, Mendez-Villanueva A. Monitoring accelerations with GPS in football: time to slow down?. International journal of sports physiology and performance. 2014 May 1;9(3):442-5.
Buchheit M, Simpson BM. Player-tracking technology: half-full or half-empty glass?. International journal of sports physiology and performance. 2017 Apr 1;12(s2):S2-35.
Johnston RJ, Watsford ML, Kelly SJ, Pine MJ, Spurrs RW. Validity and interunit reliability of 10 Hz and 15 Hz GPS units for assessing athlete movement demands. The Journal of Strength & Conditioning Research. 2014 Jun 1;28(6):1649-55.
Malone JJ, Lovell R, Varley MC, Coutts AJ. Unpacking the black box: applications and considerations for using GPS devices in sport. International journal of sports physiology and performance. 2017 Apr 1;12(s2):S2-18.
Nosek P, Brownlee TE, Drust B, Andrew M. Feedback of GPS training data within professional English soccer: a comparison of decision making and perceptions between coaches, players and performance staff. Science and medicine in football. 2021 Jan 2;5(1):35-47.
Osgnach C, Poser S, Bernardini R, Rinaldo R, Di Prampero PE. Energy cost and metabolic power in elite soccer: a new match analysis approach. Med Sci Sports Exerc. 2010 Jan 1;42(1):170-8.
Torres-Ronda L, Clubb J, Beanland E. Tracking systems in team sports: Back to basics. Sport Perform Sci Rep. 2022a;159:1-9.
Torres-Ronda L, Beanland E, Whitehead S, Sweeting A, Clubb J. Tracking systems in team sports: a narrative review of applications of the data and sport specific analysis. Sports Medicine-Open. 2022b Dec;8(1):15.
Townshend AD, Worringham CJ, Stewart IB. Assessment of speed and position during human locomotion using nondifferential GPS. Medicine & Science in Sports & Exercise. 2008 Jan 1;40(1):124-32.
Varley MC, Jaspers A, Helsen WF, Malone JJ. Methodological considerations when quantifying high-intensity efforts in team sport using global positioning system technology. International journal of sports physiology and performance. 2017 Sep 1;12(8):1059-68.
West SW, Clubb J, Torres-Ronda L, Howells D, Leng E, Vescovi JD, Carmody S, Posthumus M, Dalen-Lorentsen T, Windt J. More than a metric: how training load is used in elite sport for athlete management. International journal of sports medicine. 2021 Apr;42(04):300-6.