How do I locate an asset in the Wi-Fi-enabled network?

By Arunkumar posted Jul 09, 2014 05:54 PM


Product and Software: This article applies to all Aruba controllers and APs and ArubaOS 2.x and 3.x. 

You can locate an asset in a Wi-Fi-enabled network in various ways. The key to understanding which technology most suits your needs is to compare and contrast the options. The other challenge is to understand the business need for which the application is being deployed and if it is achievable efficiently. 

Wi-Fi Based Location Techniques 

The three major components to this solution are: 
•     Active asset: an active RFID tag or an asset with a built-in WLAN client that emits 802.11 frames. These frames can be analyzed to get information to track the asset. 
•     WLAN infrastructure: consists of APs, air monitors, and a WLAN controller to provide a secure and manageable infrastructure. 
•     Location engine: may be a dedicated appliance that gathers the information from the tags and WLAN infrastructure and provides location and business logic. 

Many technologies work on the 802.11 spectrum. The algorithms that can be used to identify location of an asset are: 

•     Triangulation 
•     RF Fingerprinting 
•     Time difference of Arrival (TDOA) 

Triangulation is the process of finding coordinates and distance to a point by calculating the length of one side of a triangle. This method relies on the asset being able to see at least three access points at a given instance. This in turn can be used with the receive signal strength that the asset reports to calculate the location. Receive signal strength information is contained in every 802.11 frame that gets sent out. 

The RF RSSI information can be a dynamic value, so the algorithm will also need to make sure that the location coordinates are reported after taking into account the propagation delays of the RF signal. 

RF Fingerprinting 
This method of calculating location coordinates relies on the characteristics that are exhibited in that location. A given location is divided into zones that may be large or small. Depending on the size of the zone, a number of RF patterns are measured in that zone using a WLAN client. These patterns are called a fingerprint. 

A number of these fingerprints are stored in the location appliance or in a database. 

Site survey and taking fingerprints becomes a very critical piece of this method. If no exact match of the fingerprint is found, then the location appliance can choose to display the closest match. The basic assumption for this kind of location tracking is that the RF remains fairly static. Any significant changes to the RF characteristics in the deployment will call for redoing the site fingerprinting. Having to do site surveys and spend time and planning effort on fingerprinting may be considered overheads in some enterprises. 

Time Difference of Arrival 
Location is found by measuring the time difference of arrival (TDOA) of a Wi-Fi signal from the emitter at three or more receiver access points. TDOA has been known to be more accurate than methods like triangulation. This methodology has been found more suitable in outdoor environments versus indoor because the reflections in the signal path are minimal in an outdoor environment. 

•     RTLS is real time. RF-based location tracking is far from being real time. It takes time to locate the asset, and when an asset is mobile, the location engine can seldom display its exact current location without a ~60 sec delay. 
•     Location tracking will work on any Wi-Fi network. Location tracking assumes that the WLAN infrastructure is intelligent enough to be able to pass on the RSSI and other relevant information to the location engine. The location engine functionality is usually a dedicated appliance on the network. 
•     RTLS technology can pinpoint the asset. Usually Wi-Fi based technologies can measure assets with 10-30m accuracy. Do not expect to see the tag on the table when it may be on the table. 
•     Static Tag will show as stationary on the map. Although the asset that is being tracked is not in motion, the display on the location engine may show it as moving.  This behavior is inherent to the RSSI variation on the environment and computation that takes place in the location appliance. 
•     Site planning is optional. For users to achieve good accuracy and search times, it is important that proper planning be done. The shape and size of the building play a role in AP placement and RF coverage considerations. The chirp interval from the RFID tag will influence the battery life of the tag as well as search times. If the tags chirp with a very long chirp interval, it may take longer times for the location engine to accurately measure its x,y coordinates. 
A few examples of where location tracking capability adds value are: 
•     Tracking where the wheelchair is in a hospital 
•     Finding IV pumps to service in a hospital 
•     Being able to locate guests in a large hotel 
•     Being able to tell when an asset crosses a marked boundary (doors, etc.)