Jan 24, 2006Sanjay Sarma, cofounder and former chairman of research of the Auto-ID Center at the Massachusetts Institute of Technology (MIT), speaking at an invitation-only conference of academics and end users of radio frequency identification technologies, said many opportunities still exist for RFID researchers and startups.
At MIT's RFID Academic Convocation, Sarma identified numerous areas where researchers might develop new technologies that could lead to business opportunities for providers of such items as tags, semiconductors, protocols, antennas, interrogators, middleware, databases and enterprise applications.
According to Sarma, although interrogator antennas are fairly simple, they could still benefit from innovation. "We need to eventually put readers on miles and miles of store shelves," Sarma explained. "How are we going to do that? There are probably a dozen answers we haven't thought of. There is something game-changing around the corner."
Sarma pointed out that current database and enterprise applications could one day be inadequate for a world with ubiquitous RFID tags and sensors, where data is captured in real time and needs to be turned into actionable information quickly. "I believe that we will have a totally new view of enterprise systems that is based on events," he said, referring to the movement of goods or the change in their status, rather than just sales transactions involving those goods.
Sarma is now chief technology officer at OATSystems, a startup that emerged out of studies on RFID middleware done at the Auto-ID Center. He suggested a great deal of research needs to be done regarding process change.
"Most of the attention so far has been on the supply chain," he said. "But that's just a start. Disaster relief could be transformed. One day, you could have something akin to a weather map of a natural disaster. The map would show what you need and where you need it. There is flooding in this district. There's an outbreak of disease over here. RFID would provide the visibility into the location of supplies and vaccines that lets you react to the crisis in a much more scientific way."
Sarma discussed a number of key areas, including protocols, privacy and security, where the problems look fairly straightforward, but the issues become more and more complex as you dig into them. For example, he explained how removing the serial number on a tag but keeping the manufacturer code and product category information will not necessarily protect the privacy of someone wearing clothes with embedded RFID tags. That's because an RFID system might still be able to identify a person through the creation of profiles of people and what they wear.
Sarma concluded his opening presentation by saying academic researchers need the input of end users to ensure their work is actually of value to companies deploying RFID. "At the Auto-ID Center, we had input from end user companies that supported our research, but we guessed at some things," he said. "We were, for the most part, right, but we were lucky. We can't depend on luck. We need to listen to the end user community."
