Knowing the location of an item or person is paramount in many applications. Tracking moving patients at a hospital, finding products at a manufacturing plant, analyzing consumer behavior – these are all applications for real time location services, or RTLS for short. RTLS allow us to see where an asset or person is located, track this information, and manage it – all in real-time.
RTLS itself is a very wide term, under which there’s a plethora of technologies, capabilities, and systems. When we look at RTLS, we can look at the level of location accuracy, the technology that it’s based on, the locating technique underlying it. At the heart of every real time location services system, though, five main components can be identified: the tag, the location sensors, the location engine, the middleware, and the application.
The tag is usually attached to the person or asset that is being located. It can be in the form of a wristband, a sticker, or a small computer. Location sensors are readers that identify the tags. The location engine is the software that determines the location of the tags, based on various wireless estimation techniques.
While the location engine is “low-level” software, the middleware is a step higher and presents the “glue” logic between the application and the RTLS. The application is the end user’s software, which can be, for example, an inventory system, a security software, or an EMR at a hospital.
How fine-grained the locating capabilities of a system are, is another question to ask with RTLS. Do you need to determine if a person is in a room? In this case, presence-based systems, which do not have complicated location estimation techniques, can be used. Do assets need to be tracked when entering or exiting a warehouse? In this case a “choke-point” system can be used. Does the exact location of expensive medical equipment need to be found on the shelves? Here, a more complex and advanced locationing system would be needed.
Another way of classifying RTLS is with the technology used. Most of the implementations rely on wireless technology, but some are based on computer vision and acoustics. These two are special cases and don’t have the regular system components described earlier – they don’t even need tags, but most solutions found are based on wireless technology.
WiFi, Zigbee, RFID, Ultra-wideband – there’s a lot of selection of technology, too much to list. Each technology comes with its list of pros and cons. The selection relies heavily on the accuracy of locating needed, on the environment, the cost, existing infrastructure, and application in question.
For example, looking at passive RFID, the tags are really cheap, and the distance they can be read is in the range of 10 meters. The downside of the technology is that it performs really poorly around liquids and metal and the readers output a large amount of energy that can affect surrounding equipment, say medical equipment at a hospital. Wifi, on the other hand, already has infrastructure implemented in most places, but requires expensive tags that would need to be powered with batteries. There are lots of considerations to take into account when selecting a real time location services technology.
Now as for the locationing engine – determining the estimated location of a tag relies on ranging and positioning. Ranging tries to determine the distance to the tag from the location sensor. It can be estimated from the signal strength, the time it takes for the signal to travel to the tag, the angle of the signal. More simple ranging relies on proximity, a reader reads a tag, that means it is in its proximity.
Positioning derives the position of the tag with the ranging information as input. If you have multiple location sensors, triangulation or trilateration can be used to estimate the location of a tag, with the ranging information taken from each sensor. If there’s prior knowledge about the environment, a technique called scene analysis can be used. Comparing captured information, to previously collected information in an environment can give an estimate of the location.
Once the location of a tag has been determined, the middleware comes into play. In locationing systems, there is a lot of data that needs to be processed and analyzed. If a tag is in the range of a location sensor, it might be constantly reporting this information. The information needs to be consolidated to just report one event, instead of a multitude. What if we’re only interested in tags that enter a certain room? Then we would like to filter tags to show only events from that room. Filtering and consolidating is something the middleware would take care of.
Giving commands to location sensors, integration with the end user application, and setting up business rules are other tasks of the middleware. The middleware can check the status of location sensors at certain places, command them to start scanning at a defined time. Some examples of the business rules that can be set up include reporting patients that move from hospital beds to outside the hospital, reporting a new shipment of products at a warehouse.
Real time location services are a powerful technology that can provide information on assets or personnel. Knowing the real-time location can reduce operational costs, improve efficiency, improve customer and staff satisfaction. There are a lot of considerations to take into account when selecting an RTLS, much of which relies on the application in question, the budget, and the environment, but the return on investment can be substantial.