Login | Register   
LinkedIn
Google+
Twitter
RSS Feed
Download our iPhone app
TODAY'S HEADLINES  |   ARTICLE ARCHIVE  |   FORUMS  |   TIP BANK
Browse DevX
Sign up for e-mail newsletters from DevX


 
 

Location-Based Computing: Letting Computers Know Where Things Are

Posted by Russell Jones on May 19, 2008


Business computing today works largely on a flawed vision of reality. For example, consider an inventory-management application. The system holds a list of items in a database, such that:

  • When items arrive at the store, the item counts are incremented with the number of items that get stocked.
  • When an item gets sold, the system decrements the count for that item.
  • When an item count falls to a specific pre-set level, the system marks that item for ordering, or possibly even orders more stock automatically.

But the system doesn't really know how many items the store has. It can't, because the items themselves have no connection to the computer. For example, the system can't—by itself—account for misplaced items or stolen items. The system is forced to rely on a human to give it that type of information.

All that's going to change, because one of the next big advances in computing is location-based computing, which uses one or more methods to let the computer know and track the exact location of objects or people.



Location-based computing, so far, has been primarily concerned with GPS information: geographic locations—positions—of roads, buildings, trucks, and, in some cases, people. Trucking companies, for example can attach a GPS system to their trucks, which can then wirelessly send their location back to company headquarters every second or so, letting the company track the actual position of the truck. If it stops, the company will know. If it gets lost, the company will know. If the driver speeds, or drives too slowly, or takes a detour to visit his family, the company will know.

Similar applications let you track individuals via their GPS-enabled cell phones, or indeed, any cell phone, because software can triangulate the position of a cell phone by the relative speed with which it responds to various cell phone tower signals. That's how the 911 emergency service can locate a cell phone caller.

GPS tracking is fine for large objects that move large distances outdoors, but it's not cheap. A GPS receiver is relatively large, and it's not well-suited for tracking objects that are indoors, where they don't have clear access to satellite signals or objects that move only small distances, such as items in a store, books in a library, or children in a school. Another technology, called Radio Frequency Identification (RFID), is better suited for tracking small objects. Several years ago, Walmart, in a rare lucid action that might actually benefit employees, customers, and the corporation, announced it would require all its Sam's Club suppliers to tag shipping pallets with RFID tags by 2010. At that time, RFID technology was in its infancy, tags were battery-powered, relatively large, and expensive.

Tagging pallets to trackmerchandise uses RFID at the macro level, but still doesn't help computers track individual objects—the goods after they're removed from the pallet. But we're about to see an explosion in location-based computing at the micro level. Newer generations of RFID tags are far smaller, have (mostly) dropped the battery and draw power from the tag reader's electrical signal, can hold more data, and are more reliable. You can easily buy tags today that are about the size of a grain of rice—but much smaller tags are in development; some are the size of a dust particle, almost invisible to the naked eye. 
RFID tags do cost money, but the price is coming down rapidly. Costs vary depending on the type of tag (powered or unpowered), how close a reader must be to read the tag, the amount of information the tag will hold, the tag's frequency, the tag's size, and of course, volume. Common unpowered tags currently cost between 7 and 15 cents per tag. They're already being used on some big-ticket items, particularly those that are small and easy to misplace or steal. Readers cost money as well, running from $100 to $750 or so, with the average reader probably running around $300. At these prices, the tags obviously aren't ready to be used to track inexpensive items. However, they're already reaching the price point where they're useful for tracking more expensive items, where keeping the item's location in sync with software is critical.

If you're already working with location-based computing technologies, let me know. If you're not, but you have ideas about how you plan to use location-based computing in your business or your applications, let me know that as well.


 

TAGS:

RFID, location-based computing


Comment and Contribute

 

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

 

Sitemap