Are you interested to meet with fellow RFID developers, source out RFID Hardware folks? I recently started a RFID Spotlight Facebook Group with the aim to bring together RFID programmer and hardware providers to share resources and experience. Its just a great way to network and meet people in this field. I made it easy for you to find us, so come and join us by following the link below.

Coming back to the subject of this post. Previously I shared When do we use Real Time Location Systems (RTLS), I will continue to share on the pros and cons of using WIFI based RTLS. We had previously used both Ekahau and Placelab for projects. Although I do not rule out some situations that suggest to us RFID based RTLS system is more superior, in general there are just more tangible benefits to choose WIFI based RTLS systems.

Some benefits of using a WIFI based RTLS are:

1. Depending on the requirements for location granularity, if done properly WIFI based RTLS systems like Ekahau produce high tracking accuracies. Ranging on average between 1 metre - 3 metres indoors.
2. Reuse and leverage on existing WIFI infrastructure. Why spend all that money to invest in a proprietary RFID infrastructure that cannot be reused?
3. Quick ROI for WIFI investments by enabling other location-based applications or wireless applications.
4. With growing categories of devices built-in with WIFI, organizations can also track other assets such as PDAs, mobile phones, laptops and any other WIFI enabled devices. Ekahau client application supports a number of PDA devices.

The cons associated with WIFI based RTLS are :

1. Lack of Scalability : RTLS like EPE requires pre-calibration.In the language of Placelab, it means we need to train the system to “learn” about places on the map in relation to actual real locations. It is fine if you are calibrating for a bunch of classrooms in a small area, however it becomes increasingly complex to calibrate for large spaces - *think* stadium, 30-floors building, airport. EPE has a feature to “merge” calibration data of different maps ( e.g. one map for each floor ) which allows us to distribute that load among a few good persons. However the results from the merged calibrated data often falls short of the required accuracy, in fact occasionally erratic. Packages like Placelab also do not have “merge” capabilities.
2. Location Model. WIFI based RTLS builds a location model ground-up from inputs such as RSSI of any given location. RSSI as you know can be affected by structures, especially solid steel structures and even moving bodies, partitions etc. Several environment issues may potentially affect these variables - access points ( *imagine* a segment of WIFI access points crash out ), structures that block signals, new partitions etc. Particularly in a museum environment, where new exhibitions are frequently replaced every other quarter and new walls/partitions being built, the original location model may be “damaged”. In situations like this, the rule of thumb is to update the calibration model every three to four months.
3. WIFI Power Intensive. This is as plain as vanilla gets. WIFI is a pure power sucker. Its pure wishful thinking that a PDA with WIFI turned on can last more than 3 hours. Bearing in mind the fact, PDA internal battery actually performs poorer over time. It means a new WIFI PDA may last 2 - 3 hour the first time it is used; after several months of continual usage, it may last no more than 1 hour with WIFI turned on.

Well I hope we provided some notes (above) to help make your final decisions on RTLS implementations. Look forward to any form of comments to add to my points…

Share This

This is part 1 of 2 on RTLS. The idea is to give an introductory to RTLS and follow up with the pros and cons of RTLS in relation to an actual implementation.

Since the rise of RFID, everyone has something they want to track. From the big boys in the ranks of Wal-Mart, Metro AG, to small-mid-size inventory based setups, are quickly harvesting every RFID feature to reap its benefits. We have received some enquiries from tagging jewellery (*obviously*) , to producing RFID labels for medical bottles, tracking WIP items on manufacturing floor, to keeping tabs on people/patients real time. So the question is, can “one size” fit all?

Certainly not. RFID comes in several types and each may likely be more suitable for specific applications - (1) Passive based RFID (2) Semi-Passive RFID and (3) Active based RFID. For passive based RFID, they are further sub divided into LF, HF, UHF.

What I hope to share is on the practicality of Active RFID, or the term coined “Real Time Location Systems”. First to explain what it is. Real Time Locations Systems, commonly abbreviated RTLS generally falls into two implementation groups. The first uses RFID technology while the second approach adopts Wireless LAN, specifically what is commonly known as WIFI. It is debatable which technology was first used in RTLS; some say RFID and the other camp claims WIFI.

It was 2002 when we started working on RTLS based projects. Since that time, I have consulted in several projects that make use of RTLS. Our team has also developed several mobile applications using engines such as Ekahau Positioning Engine and Placelab.

Now, there are several options available in the market for RTLS. This is a list of WIFI based RTLS products, and you will notice one on the list is based on open source :

• Ekahau
• AeroScouts
• Radianse
• Pango Networks
• Placelab (open source, active tags not available)
• NIST
• RFIND ( thanks Marilyn!)

At the time of this writing, RTLS systems have also been implemented using Ultrawide Band (UWB) technology. RTLS setup at its basic, consists of a battery operated tag/badge that communicates with readers/beacons strategically placed within a building, park etc. In WIFI based RTLS, the active tag wireless transmits RSSI data to the location engine server on the wireless network. The location engine acts on the Received Signal Strength Indicator (RSSI) information for each device and calculates the coordinates which represent the device’s location on the map.

Not to be confused with passive based RFID, RTLS serves specific usage scenarios which cannot handled well (or efficiently) by HF or UHF RFID systems.

RTLS Use Case : In general RTLS use case will have a specific requirement to track assets or people whose locations are arbitrary at any point of time. For example, in a hospital there is an increasing requirement to track hospital mobile assets such as infusion pumps, beds, wheelchairs. With these assets, there are no associated planned routes that can suggest strategic placements of HF/UHF RFID readers. It is almost impractical to plant RFID interrogators at every corner, on every floor of a facility. Because of the short reading distance of passived based RFID systems, you will need at least one RFID reader every 3 - 5 metres (for HF systems every 1.5 metres) !!

In relation to this, a typical RTLS use case will require the future system in providing very fine granular localized information of its asset. Engines such as those from Ekahau, AeroScouts and NIST can provide tracking average accuracy of mere metres. Walking along corridors, Ekahau Positioning Engine in particular performs localization on average accuracy of one to two metres.

A key deciding factor on using RTLS, as the name implies concerns timing. Active tags being powered by batteries can be programmed to repeatedly send out signals (packets) in very short intervals of time. The location engine consumes these packets calculates the device’s approximate location. Typically active tags is able to “contact” a nearby beacon or access point in distances ranging from 25 metres to 100 metres, depending on the grade of access points used. In stark contrast, passive tags relies fully on RFID readers to power itself. With the uncertainty that a passive tag may be sufficiently close enough to a reader, there is considerably low chances the tag can even be detected to begin with.

In summary a typical RTLS use case will have at least one of the following requirements :

1. No pre-determined path. Need to “find” assets or people who can be anywhere at any point of time.
2. Application needs to track the exact position of an asset or person down to mere metres within the facility
3. The word is real-time. Use case requires real-time monitoring and reporting of the asset or person’s locality, in very short intervals of time (seconds or minutes)
4. Reading distances further than 3 -5 metres on average

Share This