January 12, 2021
By Pavel Stankoulov
This is the first part of a three-part blog series that explores micro-location technologies. Over the past few years, Micro-location has gone through extensive development. There has been a recent surge of interest in micro-location due to its applicability in COVID-19 contact tracing. It also has applications for vehicles as well. From smartphone-based car keys to contactless identification and payments.
This blog provides an overview of what micro-location is. It also covers the state of the art of micro-location technology and its applicability to smartphones. Part 2 covers in detail one of the most popular technology currently used for micro-location: Bluetooth Low Energy (BLE) and signal strength measurements. Part 3 looks at improvements proposed in the newer BT 5.1 standard and their implications on micro-location.
Why Micro Location?
Micro-location is when a system needs to pinpoint a device location within 1-3m (9 – 3ft) accuracy or in some cases centimeter accuracy (less than 3ft). Such systems are often required to work indoors. Traditional Global Navigation Satellite Systems (GNSS) such as GPS are not accurate enough for such use cases. GPS-enabled smartphones are typically accurate within a 5m (16 ft.) radius under the open sky. Their accuracy worsens near buildings, bridges, and trees. GPS does not work indoors.
Some of the modern smartphones include support for multiple GNSS providers such as GPS, Galileo, GLONASS or Baidu. By using more satellites and different frequencies, the accuracy can be improved compared to a GPS-only positioning system. Still, due to availability, smartphone cost constraints, and the fact that such solutions do not work indoors, there is a need for improved micro-location technology for smartphones.
There are many use cases where micro-location is needed. One of the biggest ones is indoor positioning. Users want to navigate inside indoor locations such as malls, sports venues, theaters, office buildings, etc.
Another big driver for micro-location is for retail and proximity marketing. In some cases, the exact device location is not needed, but the fact that the user (device) is close to the target product. For retail, micro-location could be used as part of consumer identification, authentication, and contactless payments.
For industrial and commercial applications, micro-location is used for asset tracking on the factory floor or in the warehouse.
In the home, micro-location is used by automatic cleaning robots such as the Roomba.
There are different technologies and solutions for micro-location. Most of them use techniques for distance or angle estimation based on radio signal measurements. These methods have different trade-offs in terms of accuracy, cost, availability, etc.
The positioning methods could be grouped into the following categories:
Multilateration – this is a process to determine the position based on distance measurements from fixed points. The distance can be calculated based on the measured signal strength or time of flight. Typically for 2D positioning, distances to three known points are enough.
Multiangulation – this is a process to determine position based on direction (angle) measurements of signals from fixed points. Typically, 2D positioning angles from two known points are enough.
Fingerprinting – this is a process where positioning is performed via a previously-constructed map of radio signals. This requires off-line map creation and updating if anything that affects the signals in the environment changes.
The following are the most widely used micro-location technologies:
Bluetooth Low Energy (BLE)
Probably one of the most widely used technologies for micro-location is Bluetooth Low Energy (BLE). It was introduced as part of the Bluetooth 4.0 standard. The Low Energy extension of Bluetooth enables proximity applications and distance approximation using signal strength. All modern smartphones support BLE. There are also wide range of low-cost beacons available from multiple vendors that enable micro-location applications. For more details about how BLE could be used for micro-location, see Part 2 of the blog series.
With the introduction of Bluetooth 5.1 standard in January 2019, micro location can be further improved with direction finding capabilities. For details on the new direction-finding features and their effect on micro location see Part 3 of this blog series.
This technology uses sound to determine the distance (and possibly direction) of a transmitter. The audio signal could encrypt a special code to be recognized by the receiver. The receiver can measure the signal travel time and as the speed of sound is known (343 m/s) the system can estimate the distance to the transmitting point. If there are multiple transmitting locations, the position of a device can be estimating using multilateration. The sound can use ultrasonic frequencies, so people do not hear it.
There are existing smartphone solutions that use ultra-sound for micro location such as the library from Listnr. Also, Google’s Nearby API supports sound-based identification for nearby devices.
The advantages of the acoustic location technology are that it does not require expensive hardware. Only a microphone and/or speakers are needed. They are already part of every smartphone.
The disadvantages of such solutions are that they are very dependent on the environment, which can affect sound propagation. Furthermore, if they are used with a smartphone, smartphone placement plays an important role. The smartphone could be in the user’s pocket or bag and the microphone could be covered.
Wi-Fi has been used for location application instead of or together with GPS. It could also be used for micro-location as well. Like BLE beacons, smartphones could use the characteristics of nearby Wi-Fi hotspots and other wireless access points to approximate the device location. The smartphones could use signal strength to estimate distances to hotspots and then use mutlilateration to estimate the device position.
Wi-Fi Positioning System (WPS) have been around for a while. It is currently used as part of smartphone positioning for rough battery-saving initial positioning. Both Google and Apple collect Wi-Fi hotspot information to be able to localize phones. However, it offers lower accuracy (~ 30 m) than GPS, but WPS works indoors.
There are techniques proposed to increase the accuracy of WPS such as fingerprinting, angle of arrival, time of flight, etc. There are various proprietary solutions out there, but they all require special Wi-Fi hotspots (special hardware and/or firmware). There is currently no accepted standard for Wi-Fi based micro-location, and as a result Wi-Fi based micro-location has limited applicability.
Dead reckoning is a process of calculating position of some moving object by using a previously determined position, by using estimations of speed and heading over time.
This technique is widely used for robots and autonomous vehicles. Speed and heading can be obtained using different sensors – gyroscopes, accelerometry, compass, speed pulse from the wheels, distance sensors using ultrasound, radars, etc.
There are solutions for smartphone positioning that use dead-reckoning. The phone includes many of the needed sensors such as compass, accelerometers, gyroscopes, etc. However, this method is not reliable as there is no guarantee how the user is holding the phone. For example, they could be holding the smartphone in their hand while walking and causing additional acceleration vs. when the phone is in a pocket. As a result, dead-reckoning is not suitable as the only method for smartphone-based micro-location. If used, it augments other positioning techniques.
Ultra-wideband (UWB) is a radio technology that can use a very low energy level for short-range, high-bandwidth communications over a large portion of the radio spectrum. UWB uses time of flight (ToF) measurement to determine distance. The UWB radios send information using very short burst (called chip) of electromagnetic perturbation. As a result, it is easier to filter out reflections on walls or ceiling and increase the reliability of the measurements. UWB promises centimeter accuracy for micro-location applications. With three or more UWB beacons with known location and by using multilateration the smartphone position can be estimated.
From iPhone 11, Apple started adding UWB chip to their iPhone models. On the Android side there are only few smartphone models that include UWB support such as the Galaxy Note 20 (SM-N985F). Google has not announced any official support of UWB in Android.
UWB has the promise of high-accuracy micro-location technology. However, it is still supported by only small number of smartphones (iPhones). Currently there is no Android support for it and also no established standard yet.
5G is the fifth-generation technology standard for broadband cellular networks, which cellular phone companies began deploying worldwide in 2019. 5G promises improved positioning, navigation and timing (PNT).
5G will use dense networks of picocells, which are small cellular base stations typically covering a small indoor area. The picocells can be used for improved micro-location similar to BLE beacons.
5G uses cellular architectures will use adaptive array technology to achieve high data rates, spectrum reuse and communications robustness. It can be used for digital beam-forming. This helps with multipath and for jamming and spoofing mitigations. 5G uses direction-of-arrival positioning and ranging in order to maintain connectivity with user nodes. This information can be shared with the smartphone devices to provide micro-location using multiangulation.
However, 5G for PNT is still emerging technology. This has not been demonstrated in production yet. Also, the micro-location accuracy depends very much on the cell density.
Another drawback of 5G at the moment is that PNT is not supported by the smartphones yet.
Still, 5G has the potential to become the prevailing global micro-location technology.
Radio-frequency identification (RFID) uses electromagnetic fields to automatically identify and track tags attached to objects. When triggered by an electromagnetic interrogation pulse from a nearby RFID reader device, the tag transmits digital data. This is usually an identifying inventory number.
There are two types of RFID tags. Passive, which needs to be activated by an electromagnetic field. Their range is up to 3 m. There are also active tags that are powered (battery, solar, AC, etc.) that can transmit signals up to 100m. Active tags are like BLE beacons and can be used for multilateration positioning estimation.
RFID technology is primarily used for asset tracking due to the low cost of the tags (especially if passive) and the low-cost scanning technology. RFID are not used much for positioning especially with smartphones where RFID is not supported without additional hardware.
This blog post provided an overview of the main technologies used for micro-location and highlight their applicability for smartphone-based location applications.
The next part of the series delves deeper into BLE based micro-location using the signal strength indicator to calculate distance and perform position estimation using multilateration.
Let Abalta Help with Your Micro-Location Needs
To learn more about how Abalta can help solve your micro-location needs, please contact us at firstname.lastname@example.org.