Waiting for the Internet of Things

More than a decade ago, radio-frequency identification (RFID) technology pioneer Kevin Ashton coined the term "Internet of Things." The idea was that every item, product, or "thing" would have a unique identifier just as every computer does on the Internet. RF tags, of course, would provide the means by which these things could be tracked and identified.

For logistics managers, the Internet of Things would be a game changer. Among other benefits, it would make it possible to track the flow of goods into and out of a warehouse at the item level. Some retailers and consumer packaged goods manufacturers are already experimenting with item-level tracking. Nonetheless, it appears that the ability to track everything is still several years away.

Why? A recent report from Frost & Sullivan ("Analysis of the Active RFID and Sensor Networks Market") offers some insight into the barriers to making the Internet of Things a reality. One of the top challenges, it notes, is getting more companies to buy the type of tags necessary to make this possible.

An Internet of Things would require the use of active tags—tags equipped with a transmitter and their own power source, typically a battery. These tags periodically transmit their identifying information—in effect, saying "Here I am, notice me." That continuous flow of information regarding an object's presence at a particular time and place provides visibility into the movement of goods as they travel through the supply chain.

Frost & Sullivan's research, however, indicates that while users are generally aware of active tags and their capabilities, they still prefer passive RFID tags. As the name implies, a passive tag has no battery or power source. In order to transmit its identifying information, the tag must be "pinged" by an outside energy source.

As for why users are shying away from active tags, there are a couple of reasons. First, there's the lack of common industry standards. While passive tags use data standards developed by the EPCglobal consortium, there's no such system in place for active tags. At the moment, makers of active tags use different technology protocols, such as Wi-Fi, Rubee, Zigbee, ultra wide band, infrared, and ultrasound. All of those protocols require different standards, hindering widescale adoption of the technology.

"Market industry participants, industry associations, and governments across all nations need to work toegether to bring a common standard to boost the adoption [of active tags]," said Frost & Sullivan Senior Research Analyst Nandini Bhattacharya in a press release announcing the report's publication. She notes that the Institute of Electrical and Electronics Engineers (IEEE) is now working on an 802.15.4f standard that would enable interoperability of various frequencies and technologies used by vendors.

Although a standard would hasten the adoption of active tags, there's still another obstacle—cost. Bhattacharya says a passive basic tag goes for $2 to $5 per unit,while an active tag costs between $10 and $15. And that's the low end of the range. If those tags are embedded with sensors and support multiple technologies, the cost of an active tag can top $100 per unit.

At those prices, active tags are still too costly for most companies to justify their deployment. "Although the prices of various active RFID technologies have [dropped] in the last couple of years, companies with budgetary constraints still find them prohibitively priced," says Bhattacharya.

Despite those impediments, the Frost & Sullivan report notes that more companies are looking to migrate from traditional active tags to more advanced technologies such as real-time locating systems (RTLS) and active RFID-based sensors that allow users to track the location and condition of objects in real time. In its report, Frost & Sullivan predicted that the total market for all types of RFID and sensors, which stood at $964 million in 2010, would reach $8.39 billion in 2017.

Just how accurate that prediction will prove remains to be seen. But it looks like it will be at least another five years before the Internet of Things arrives.