Technological Advancements

Technological advancements are the high-octane fuel that powers the continued acceptance and growth of new technologies. These advancements can provide the following advantages:

• Make existing applications easier to use

• Offer more functionality

• Drive deployment costs down

Technological advancements open the door for new applications that were not imaginable or possible before. In the following section, we explore some of the more significant technological advancements that are under development today.

New and Improved Tags

Innovation around the design and manufacture of RFID tags is an ongoing process. Some of the most promising new designs are covered in the following sections.

Alternative Tag Designs

Many factors, including physical and environmental, affect the readable range and accuracy of tags. Some examples are detection near metal or liquid and extreme weather conditions such as low temperature or high humidity. Besides simply improving on existing technology to overcome these limitations, alternative physics are being employed that can sidestep or leapfrog these limitations.

The majority of the work in the alternative physics area includes developments around chipless tags, introduced in Chapter 3, "Components of RFID Systems." Chipless tags promise to improve upon the physical limitations of radio frequency detection while potentially offering reduced costs due to the absence of integrated circuitry. Chipless tags can be more easily applied near metal and liquid or embedded in items like paper, thereby offering greater flexibility and functionality with their use. One chipless tag technology showing promise in supply chain applications uses Surface Acoustic Wave (SAW) technology. SAW technology involves the propagation of radio frequency acoustic waves on the surface of polished crystals. Other promising chipless technologies that have the potential to revolutionize RFID applications use nanotechnology, genomics, or even chemistry to achieve chipless tagging and unique identification of objects such as paper currency and product labels. You can find vendors that develop and supply chipless tag technologies at this book's companion Web site, www.rfidfieldguide.com.

When it comes to major advancements in IC-based tag design, Smart Active Label (SAL) technology is gaining momentum in the market. SAL offers enhanced range and accuracy attributes while being less vulnerable to liquid or metal. A SAL tag is essentially a semi-active smart label with its power source in the form of a thin, flexible battery. Using SAL tags, tagging and detecting cans of soda and bottles containing liquid can become more practical and economical.

Tag Packaging

Tag packaging plays a significant role in the applicability and practicality of specific uses of RFID. Expect to see tag and antenna packaging designs that will continue to push the envelope of creativity and ingenuity, much as injectable and ingestible tags have done in the past. Chipless tags based on nanotechnology will certainly be at the forefront of such developments.

Another entirely different approach to tag packaging that is very promising is related to printed electronics. This involves the process of "printing" antennae, transistors, or even integrated circuits using conductive ink and standard printing processes. The potential to inexpensively print a tag onto a box or the packaging of an item unlocks a new set of possibilities for the widespread application of RFID in everyday items. Already, several companies (identified in the vendor guide at this book's companion Web site, www.rfidfieldguide.com) have designed smart label antennae that use conductive ink instead of copper.

Sensory Tags

Tags whose packaging integrates them with sensors can monitor, record, and even react to all sorts of environmental conditions. Known as sensory tags, these tag types promote an entirely new set of applications. The major advancements here will be around the coupling or combining of RFID tag technology with sensor technology in very small form factors. Smart Dust is one such combination that offers the functionality of tiny environmental sensors known as MicroElectroMechanical Sensors (MEMS) with active RFID tag-like capabilities. Each such device is expected to be one cubic millimeter in size. The potential applications of this technology span a wide area, from monitoring battlefield activities in a military operation to tracking the facial movements of the disabled to control their wheelchairs.

Architecture for the New Network

RFID systems generate mountains of new data that need to be synchronized, filtered, analyzed, managed, and acted upon, often in real-time or near real-time. Each tag is essentially a single computing device, albeit a very simple one, that acts as one node in a network of, eventually, billions or even trillions of such devices. This new network is dramatically different and in many ways more complex than even the Internet, the most complex network ever known. This fact is due primarily to the number of nodes that could exist in the expanded model of a worldwide RFID network, which will be several orders of magnitude larger than the number of nodes on the Internet. This simply means that traditional computing architectures and infrastructures will not be adequate to handle the dramatically higher data volumes expected in a network of RFID tags. Here, we discuss two different approaches under development that address the requirements of this new network from both hardware and software perspectives.

Where will all this RFID data come from?

Consider the scenario where a major retail chain will be tagging all its goods in all its stores, at the single item level. The number of tagged items in this scenario can easily reach 10 billion or more. This means that the data identifying the 10 billion items amounts to 120 gigabytes (10 billion x 12 bytes per tag). If these items were read once every 5 minutes somewhere in the supply chain, they would generate nearly 15 terabytes of tracking data every day (120 gigabytes x 12 times per hour x 10 hours per day). That's 15 terabytes of additional data generated by one retail chain every day. Using this formula, 10 major retailers tagging and tracking every item will generate 150 terabytes of data. This is bigger than the estimated 136 terabytes of data from 17 million books in the U.S. Library of Congress^[1]. Obviously, a great majority of this RFID data is duplicate and will likely be discarded. However, all the data needs to be processed, examined, and acted upon, even if such action means simply ignoring much of it.

We use item-level tagging (a more distant scenario) to demonstrate the eventual avalanche of RFID data. However, you can apply a similar formula to calculate the amount of data for a more immediate scenario: case- and pallet-level tagging. Although the volume of data in this case is an order of magnitude smaller, it still represents several orders of magnitude more data than a pre-RFID scenario.

^[1] Source: University of California, Berkeley: How Much Information 2003? http://www.sims.berkeley.edu/research/projects/how-much-info-2003/

Microprocessor Design

Several computer giants are revising their microprocessor development roadmaps in favor of a new microprocessor architecture called Chip Multi-Threading (CMT). One of the pioneers in this area is Sun Microsystems, which has already introduced the first design of this new architecture. This is just in time for the expected volume spike in RFID data as the U.S. Department of Defense (DoD) and major retailers around the world go into full deployment mode with their mandates. Simply put, CMT architecture bucks the trend of traditional microprocessor design and architecture that primarily seeks to perform single tasks faster and faster. Instead, CMT is an architecture that allows the efficient execution of many tasks simultaneously. This is parallel computing taken all the way to the core of the microprocessor.

Peer-to-Peer Computing

Although the data generated by RFID systems can easily reach trillions of bytes that need to be processed almost instantaneously, much of the data is disbursed across one or more enterprises, and often across the globe. This suggests that local processing of data, by RFID readers, before passing it along to a centralized computer can dramatically reduce the burden placed on centralized computing resources. This is an excellent scenario in which to apply Peer-to-Peer (P2P) programming techniques to perform RFID-related data processing locally. P2P technology is a type of distributed computing technique that decentralizes computing tasks across several less powerful cooperating computers (peers) within a network.

Expect RFID readers to become increasingly more "intelligent." Readers will perform many of the data processing, analysis, and management tasks within a local network of cooperating tags and readers. They will accomplish what today is mostly done by centralized computers.

Falling RFID Tag Price

With RFID technology, cost of components, especially cost of individual tags, will play a major role in determining its ultimate success and ubiquity. From an economic perspective, the cost of tags is expected to continue to drop as the volume production goes up to meet demand. However, both alternative chipless tag designs and advances in fabrication and manufacturing of integrated circuits (IC) are expected to drive the cost of tags dramatically lower. The 5 cents tag, as it has been called, has been widely viewed as the inflection point where wide adoption of RFID will quickly occur. To be clear, the supply and demand equation alone is unlikely to drive the price of IC-based tags down to the 5 cents mark. Today, tag prices barely dip below 25 cents, even in high volumes. Therefore, alternative tag designs and more efficient tag manufacturing are likely to be important factors in driving the cost of tags down by another factor of five.