RFID, Sensors and the Internet of Things

By Mark Roberti

The use of a wide variety of terms to describe RFID technologies—and the lack of precision among those terms—confuses companies seeking solutions to their business problems.

Pity the poor businessperson. He or she is interested in taking advantage of new technologies that enable a company to monitor the location or condition of assets in real time. But technology providers, users and the media throw about so many terms to describe these technologies—including radio frequency identification, sensors, mesh networks, Internet of Things devices and the industrial Internet—how does anyone begin researching possible solutions?

Articles in the mainstream press often conflate all these concepts. Writers sometimes refer to RFID transponders as "sensors" or "wireless sensors," either to avoid the use of radio frequency identification, which seems like jargon to editors, or perhaps because they aren't sure what the different terms actually mean.

Last November, for example, The New York Times published an article about General Electric's efforts to gather data from the many items it manufactures (see Looking to Industry for the Next Digital Disruption). "Today, GE is putting sensors on everything, be it a gas turbine or a hospital bed," the article said. It included no explanation of the different types of sensors. The "sensor" on the hospital bed is most likely an active RFID tag that broadcasts the identity of the object and its location. It's not clear what GE monitors within its gas turbines or which technology it uses, but the company's Web site refers to the use of passive RFID tags for asset management.

The term "Internet of Things" was coined in 1999 by Kevin Ashton, then executive director of the MIT Auto-ID Center. It helped explain the concept of putting a low-cost RFID transponder on, say, a case of shampoo to enable tracking of the product from manufacture through sale. But some now use the term interchangeably with sensor networks and mesh networks. To confuse matters more, the term has broadened over time to include sensors inside medical equipment or photocopiers that can report on a machine's condition through an Ethernet connection to the Internet, and almost any other technology that connects a machine to the Internet.

Whether RFID tags are sensors is open to debate. The Oxford English Dictionary defines a sensor as "a device which detects or measures a physical property and records, indicates, or otherwise responds to it."

A basic RFID tag detects the presence of an object and can tell you its identity. It could be argued that this makes an RFID tag a sensor. Ashton addresses the issue in his column, "Making More Sense." "RFID tags enable computers to sense identity, and knowing what something is almost always is a prerequisite to being able to use other sensory information, such as temperature," he says. "I consider an RFID tag a 'sensor' because it can detect something about the physical world remotely and by proxy."

But passive RFID can also be used simply as a tool for counting objects, such as clothing items on store shelves. When used like a bar code, RFID does not seem like a sensor.

There are active RFID tags with on-board sensors that can detect humidity, moisture, motion, pressure and temperature. There are also battery-assisted and a few passive tags with sensors. RFID is certainly part of a broader ecosystem of data-capture technologies, and the many forms of RFID will exist alongside and in conjunction with sensors that monitor environmental and other conditions.

The debate about whether RFID tags are sensors is not simply academic. The bandying about of these terms results in confusion among businesspeople seeking to purchase solutions to solve their companies' problems. They don't understand the various types of RFID or the difference between RFID and other technologies for monitoring machines.

To help keep our readers informed and clear about what different technologies do, RFID Journal has adopted the following approach to these terms:

Basic RFID transponders, whether active or passive, are not sensors—they are RFID tags or transponders. Their purpose is to identify an object and determine its location. Tags are often used to count objects in known locations.

Wireless sensors, whether they communicate via mesh networks or conventional RFID readers, are RFID sensors—provided they include an ID that allows you to differentiate one sensor from another. Some argue that mesh-networking nodes, sometimes called motes, are not RFID because they have a central processing unit and run an operating system. That's like saying a laptop isn't a computer because it isn't a mainframe.

Most of our articles discuss a specific type of technology used to identify and track something. The Internet of Things is a term we use to refer to the broad set of wired and wireless systems that enable objects and machines to connect to the Internet and share information. RFID Journal covers smart appliances that rely on RFID to collect and transmit data (see NFC-Enabled Refrigerator Shares Data With Mobile Phones), but does not focus on sensors built into machines to monitor their condition and report on it via a wired connection to the Internet.

We believe that clarity in the use of these terms, even if some might consider our definitions arbitrary, will help companies seeking to use these technologies determine precisely what they need.

Illustration: iStockphoto