Section 4.2. Emerging Application Types | RFID Sourcebook (paperback)

4.2. Emerging Application Types

The following are some of the application areas of RFID that hold rich potential for the future:

Anti-counterfeit
Smart tags

These application types are not listed in any particular order signifying its importance or the degree of its applicability in an RFID context. It is also possible that the benefits and characteristics of two or more applications types can overlap. The following sections discuss these application types in more detail. For each class, at least one concrete example is provided. For each such example, the benefits and caveats are discussed. An implementation note accompanying each such example provides deployment-specific details.

4.2.1. Anti-Counterfeit

A wide range of items is susceptible to counterfeiting. Some of the items most frequently counterfeited are prescription drugs; currency bills; and high-value items such as perfume, electronics, and watches. Billions of dollars of revenue are lost annually as a result of counterfeiting, in addition to the pain and suffering inflicted on unsuspecting users of counterfeit items (drugs, for instance). An accurate estimate of counterfeiting is difficult because it can be hard to detect and investigate. In addition, counterfeiters are getting sophisticated and technologically savvy, making it impossible for a single technology or method to provide the silver bullet for its prevention. RFID can provide a solution to this problem; to be effective, however, several methods must be used in combination. Some of these measures may be nontechnical, such as licensing policies, education and awareness, law enforcement to tackle counterfeiting, and so on.

Member Applications

This application class type has a large overlap with item tracking and tracing. Indeed, an item to be tracked can be viewed as an asset that can be monitored. Some of the important examples of this application type are as follows:

Drug anti-counterfeit. This is discussed in the next subsection.
Currency anti-counterfeit. This is discussed next.
Brand-name and luxury item anti-counterfeit. RFID can be used to prevent counterfeiting of brand name items (sportswear, for example). Unique item codes can be embedded inside authentic sportswear and tracked. When such a sportswear garment surfaces outside an authorized selling zone, the tag can be used to detect who the original reseller was responsible for distributing it to the gray market. In addition, if a sportswear garment does not have a valid RFID tag (for example, damaged or containing an invalid item code), it can be assumed to be a counterfeit. This can also help to automate inventory in a store, replacing the current manual methods. However, tagging an individual item such as sportswear with an RFID tag is fraught with privacy issues.

The next two subsections discuss what are probably the two most important uses of RFID in preventing counterfeiting.

4.2.1.1. Drug Anti-Counterfeit

The International Federation of Pharmaceutical Manufacturers Association (IFPMA) estimates that 2 percent of the drugs sold per year globally might be counterfeit. The World Health Organization (WHO) estimates this number to be between 5 percent and 8 percent. Based on these estimates and a global drug market of $327 billion annually, the dollar value of counterfeit drugs ranges from $7 billion to $26 billion annually. A drug can be counterfeited in several ways, including the following:

Missing active ingredients
Labeled with an extended expiration date
Labeled with incorrect potency information
Labeled with a different drug name
Contaminated with possible lethal impurities
Diluted

According to a recommendation by the Healthcare Distribution Management Association, a nonprofit organization for drug distributors, RFID (EPC) tags need to be put on cases by 2005 and applied at the unit level by 2007. However, industry-wide adoption might be as long as 10 years away.

One RFID solution to this crisis is to associate a unique item code or an electronic signature (for example, an EPC [see Chapter 10]) with a case or an individual unit (drugs, medical equipment, and so on) so that it can be tracked through the supply chain. Such an electronic pedigree could help to detect counterfeit items entering distribution and to handle returns and recalls. Suppose, for example, that for any particular prescription bottle, an EPC tag must be associated with it for authenticity. Then one, and only one, of the following situations is possible:

The bottle does not have an EPC tag.
The bottle has an EPC tag that does not exist in the company database.
The bottle has an EPC tag that exists in the company database and matches the product details (such as the name; expiration date; when the product was manufactured, distributed, and received).
The bottle has an EPC tag that exists in the company database but does not match the product details (such as when the product was manufactured, distributed, and received). Suppose, for example, that the expiration date of a drug reads "February 20, 2006," but the data pulled out from the company database indicates an expiration of "March 15, 2005;" in this scenario, either the drug is a counterfeit or its expiration date has been tampered with. In any case, this drug should be discarded.

Clearly, except for situation 3, the bottle is a counterfeit. A basic assumption underlies this solution: The company databases are hosted securely by the pharmaceutical companies in such a way that an unauthenticated entity cannot tamper with them and that only authenticated entities (for example, a human operator or a computer program) with proper permissions can update them.

Benefits

Besides anti-counterfeiting, this application offers better supply-chain management and inventory control.
Drug returns can be facilitated by validating the authenticity of the returned drugs and the purchaser. Similarly, drug recalls can be targeted to customers who bought this drug. (It is assumed that the drug's EPC data will be associated with customer data at the time of purchase.)
A global EPC infrastructure can provide an effective deterrent to the current counterfeiting problem.

Caveats

For this type of solution to be effective, a comprehensive EPCglobal type of infrastructure is necessary (see Chapter 10). For this to happen on a global scale, massive investment is needed to set up the hardware, software, and network infrastructure (which might take several years to happen).
Challenges related to tagging every drug type in its various forms of packaging need to be solved. Most probably, specialized EPC tags will be needed to tag several types of drugs. These might also slow down the application adoption time and increase the cost of implementation.
The strength of EPC (that is, the company database) is also its weakness. If internal personnel or anyone external can tamper with the company database, EPC might actually aid in counterfeiting. In this case, counterfeited drugs gain "legal" status just by having the associated spurious data residing in the company database. In addition, simply attaching a valid EPC tag to a drug is not a watertight anti-counterfeit solution (see the following Implementation Notes); the introduction of counterfeit drugs into the distribution system is still possible.
If a valid EPC tag associated with a drug gets damaged through mishandling, its authenticity cannot be verified. In this case, the drug must be discarded, resulting in lost revenue. Note, however, that RFID tags are generally sturdy and can withstand environmental conditions to a fair extent.
Government permission might be needed in some cases to tag certain types of drugs or medical equipment. FDA recommendations on RFID might also contradict or make redundant some directives on unit-level bar coding for patient-safety initiatives. These might impact the adoption time.
A wide range of personnel must be trained to use the application.

Implementation Notes

Passive tags in the HF (13.56 MHz), UHF (868870, 902928 MHz), and microwave (2.45 GHz) frequencies have the potential to be used for this application type. Pallets and cases most probably will be tagged with UHF frequency tags. The actual drugs can be tagged with HF, UHF, and microwave tags. It might seem like a counterfeit drug can be made "valid" by stripping off a valid tag from a valid drug and putting it on this counterfeited drug. However, the tags will be manufactured in such a way that they will break if such an attempt is made, making it evident that some kind of tampering occurred. How about reading the valid unique electronic signature (such as the EPC number) from a legitimate drug, creating a tag with this data, and putting it on a counterfeit drug?

Consider a concrete example. Suppose that a drug called foobarlene is being distributed in cylindrical plastic bottles that are 3 inches high, 1 inch in diameter, and weigh 3 ounces. The drug is a pink liquid, and has an expiration date of December 28, 2006. Assume that the valid EPC tag from one of the foobarlene bottles is read by a counterfeiter, and that this person creates and attaches an identical tag using this data to a counterfeited drug. First, the physical characteristics of the drug have to match exactly. For example, the counterfeited drug has to look like a pink liquid packed in a cylindrical bottle 3 inches high and 1 inch in diameter, and must weigh 3 ounces. The bottle must have an expiration date of December 28, 2006. That way, even if someone compares the physical characteristics of this drug, using the EPC tag data, as stored in the company database, those characteristics will match with the counterfeit one. So comparing the physical characteristics in this case is not sufficient to determine conclusively whether this drug is indeed a counterfeit.

Next, the intended distribution region, as stored in the company database for this drug bottle, is compared with the counterfeit one. Suppose, for this example, that the drug was intended for distribution in South Africa, but actually shows up in China instead. Clearly, something is wrong, and this bottle of foobarlene is a counterfeit suspect. However, what happens if this drug shows up in South Africa instead? Because the intended destinations match, it cannot be determined for sure that this is a counterfeit drug (although in this case, it is). Therefore, if on its way to South Africa, this bottle of foobarlene is replaced with a counterfeit one that has exactly the same physical characteristics of that of the actual one, the company product data associated with the EPC tag will not be sufficient to determine conclusively that this drug is a counterfeit. Some other measures will be necessary (for example, opening a bottle of foobarlene randomly and testing the chemical properties of the drug to see whether it matches the composition specified in the company database).

As you can understand from the preceding example, one of the soft spots of this unique electronic signature solution seems to be between tag reads. That is, the counterfeits can be introduced by replacing the valid drugs between tag reads (for example, between two valid distribution points). Counterfeit drugs can be introduced, for instance, during transit between two valid distribution points, by replacing a valid lot of drugs with counterfeit drugs having the same EPC tag data, the same EPC case and pallet tag data, and the same physical characteristics as the replaced lot. Admittedly, pulling this feat off will require some very sophisticated and resourceful counterfeiters. In this case, the solution can do very little to identify these counterfeit drugs. The stolen valid drugs can then be sold on the black market to individuals who are not at all concerned whether a bottle of this stolen drug has a unique electronic signature. Note that in this case, no tampering with the company database is needed.

Dishonest sellers can also abuse electronic signatures to purvey counterfeit items. In one case, the seller can just put duplicate electronic signatures (for example, EPCs) on counterfeit items and bypass recording the sale of any of these counterfeits (as well as bypass the recording of the sale of the original item that had the genuine EPC associated with it). As a result, a duped customer can "authenticate" the counterfeit item's EPC from reliable sources (except the item information might indicate that it has not sold yet!) and might think that he indeed bought a genuine item. In another simple scheme, the seller of a counterfeit item can scan an EPC of a counterfeit item and show the customer the associated electronic pedigree and product-specific data on his computer to verify its "authenticity." However, instead of getting this data from the manufacturer/product database, the seller can spoof it locally using a simple program on his computer! The customer might not be able to recognize this ploy easily because the header, manager number, and object class of the counterfeited item's EPC, together with valid product information and the electronic pedigree, can easily be copied from a valid EPC of a similar product. In this case, the buyer might be easily convinced that he is indeed buying a genuine product! The seller can create further confusion, if confronted (after the buyer confirms it is a counterfeit via a valid source), by accusing technology glitches. ("Didn't I show you that the EPC was valid? How can it be invalid now? Let me show it to you again. That's the problem with this technologyit doesn't work all the time, and look who is getting blamed for using it!")

These scenarios clearly do not represent a weakness of the technology or the EPC scheme, because parts of the technology are selectively ignored to misuse and misrepresent the technology to sell counterfeit items. These scenarios do show that security and anti-counterfeiting are not simple issues that can be solved just by attaching RFID tags to items. This is where nontechnical measures, such as tracking down and prosecuting counterfeit sellers and black marketeers, strict control of transportation, physical security of the transported lot, and so on, are necessary to ensure that the technology delivers its potential benefits to the customers and end users. Then, the RFID solution coupled with these nontechnical measures will provide a solution that is close to bulletproof.

4.2.1.2. Currency Anti-Counterfeit

Currency counterfeiting is one of the oldest crimes in history. According to one U.S. Secret Service official, about $63 million in U.S. counterfeit currency was seized in 2003, of which $10.7 million was seized in the United States.^[6] Columbia accounted for more than $31 million, being the single largest producer of U.S. counterfeit currency. The Secret Service estimates that more than 42 percent of counterfeit currency passed domestically in 2003 was produced outside of the United States, whereas 46 percent of this amount was produced in the United States. Organized crime networks exploit the opportunities associated with "dollarization," which refers to the process by which a foreign country (for example, Ecuador) adopts the U.S. dollar as its national currency. These countries are prime targets for counterfeit U.S. currency. In the United States, an individual is solely responsible for the authenticity of the currency he carries. Authorities can seize counterfeit currency from an individual without any legal requirement to compensate this person. Likewise, the financial institution that dispensed the counterfeit currency to this individual is also not legally liable to compensate him.

^[6] Statement of Mr. Bruce A. Townsend, deputy assistant director, United States Secret Service, before the Committee on Financial Services, April 28, 2004.

RFID can be used to provide authenticity of paper currency bills. Very small RFID tags can be concealed in a currency bill. Such a tag can carry EPC data or some kind of unique identifier that can be read by special readers. If a tag is absent or if its data cannot be matched against the currency database, it can be assumed to be a counterfeit. Multiple tags, each carrying its unique data, can be inserted into a single currency bill so that the combination of the data from these tags uniquely identifies and authenticates this currency bill. This, in turn, will also make the counterfeiting difficult.

Benefits

Thwart nonexpert counterfeiters. An RFID solution would eradicate the vast majority of current and would-be counterfeiters who are not technically savvy.
Track illegal transactions. A currency bill can be tracked at a transaction level, which makes this solution an effective deterrent of illegal transactions. For example, when an ATM dispenses cash, the tag data of the currency can be read and associated with the account from which it is withdrawn. Similarly, when a currency note is deposited in the bank, the tag data can be associated with the account into which it is deposited. In both of these cases, the person(s) holding the account can be associated with handling this currency note.
Effective way to track blackmailers. The currency notes can be read and marked in the database before these are handed out to blackmailers whose identities cannot be disclosed at the transaction time. Next time the current notes are traced (from the previous benefit), the data can be used to identify the blackmailers.

Caveats

For this type of solution to be effective, a comprehensive EPCglobal type of infrastructure is necessary (see Chapter 10). For this to happen on a global scale, massive investment is needed to set up the hardware, software, and network infrastructure (which might take several years to happen).
If a valid EPC tag associated with a currency note is damaged through mishandling, its authenticity cannot be verified. In this case, the note might have to be marked as destroyed. Note, however, that RFID tags are generally sturdy and can withstand environmental conditions to a fair extent.
Associating currency note EPC tag data with a consumer records might raise privacy concerns.
Undue harassment of innocent persons might result as a side effect of associating transactions with a currency note. Suppose, for example, that a person A receives a valid currency note from a bank ATM (which records the tag ID of this currency note with this person's account). Assume that this note is subsequently counterfeited, without any knowledge of person A, and the authorities track the counterfeited note. Because the last transaction using the valid currency is associated with person A, he might become a target of the counterfeiting investigation. In another situation, person A uses this currency note to pay his taxi fare to the airport. The taxi driver is subsequently robbed, and the robber uses the currency bill to pay for drugs. The police catch the drug seller and retrieve the currency note. Person A could be in deep trouble now! The main issue in the second example is that not all cash transactions can be recorded to link the currency note with a transaction. These gaps can reduce the effectiveness of such tracking applications to a major extent.
A wide range of personnel must be trained to use the application properly.
Finally, the tag cost has to come down substantially from where it is today.

Implementation Notes

The possibility of tampering with the currency database, although theoretically possible, is remote. After all, it would probably be among the most secure databases in existence. Passive tags in the microwave (2.45 GHz) frequency range will be used because these require only small antennas to communicate with the reader (a crucial element that allows a tag to have a very small form factor so that it can be embedded in paper currency).

4.2.2. Smart Tags

The smart tags application type almost seems like an unfair catchall classification of an emerging application type! A smart tag is essentially an active (or a semi-passive) tag that has a battery and on-board electronics, and can therefore perform custom tasks besides just storing and transmitting data unique to an attached object. The key phrase here is custom tasks, which can be anything! For example, a custom tag can monitor and report its surrounding temperature, humidity, radioactive emissions, and so forth, among other unlimited task types. In other words, RFID can be combined with sensor technology to create a variety of smart tags. The application members belonging to this application type are only limited by one's imagination. Note that a smart tag can be of any physical dimensions as long as it can be deployed. It can be the size of a pack of cards, laptop, or a suitcase, provided it can be attached properly to the targeted object. Therefore, "anything" seems to be fair game when it comes to smart tags! It is the author's personal opinion that this application type holds the most potential for RFID technology of the future.

Member Applications

These application types do not exist today. Several of the application members (current and future) that belong to this application type may also belong to a different application type. For example, today, the electronic toll collection system uses smart (semi-active) RFID tags that can indicate whether the account balance of a person's toll account is at an acceptable level. This application is thus a member of both smart tag and electronic payment application types. Some example applications belonging to this type are as follows:

Toll payment. This is described earlier in the discussion of prevalent application types.
Smart expiration-date determination. Every perishable item has an expiration date that is determined at the time of its manufacture by assuming certain storage conditions, such as temperature, humidity, and so on. In reality, however, the "safe" consumption period can vary considerably. Consider meal packets to be consumed by soldiers in the field. Assume that these packets are stored in local storage areas near battlegrounds. The environmental conditions of these storage areas might be unpredictable. For example, a storage area might be located in the middle of a desert or in a high-altitude mountain cave, where temperature and humidity variations are extreme. As a result, the original expiration dates that were calculated based on average environmental conditions are not valid anymore. In this situation, smart tags attached to cases of such packets can determine environmental changes and can automatically calculate new expiration dates. These tags can then transmit this information to a central inventory system, which can then direct army personnel to discard meal packets whose new expiration dates have passed. The whole process of determination and notification is automatic and accurate.
Smart weapon. This is discussed in the next section.

Because of the high number of members of this application type and their substantial variations, the following section covers a sample member (a smart weapon) to provide a glimpse of the potential of this application type.

4.2.2.1. Smart Weapons

Suppose that an army is at war with enemy forces in hostile terrain.^[7] It is difficult to send a large number of forces through unexplored and dangerous terrain to secure it. It is equally difficult to keep the terrain impassable for enemy forces. However, control of the terrain might prove very crucial to winning the war. Currently, land mines are used to make a terrain inhospitable to enemy forces. However, land mines make the terrain equally dangerous for both forces. In today's high-technology warfare, military planners want a "smarter" solution.

^[7] The author does not support or condone war in any form. This is for illustration purposes only.

Enter RFID active tags used to design a new breed of smart weapons that can recognize a friendly force from an enemy force and can modify their behaviors accordingly. The military can airdrop a large number (perhaps several thousand) of these smart weapons (for example, high explosives), with an active RFID tag attached to each, on a particular terrain to make it inaccessible. A friendly force passing through this terrain can instruct, securely, the RFID active tags attached to these weapons to deactivate the weapons. After passing through the terrain, the force can, conversely, turn the weapons back on, rendering the terrain inhospitable to the enemy force.

The possibilities do not stop here. These weapons could, in real time, sense the external conditions of the terrain (for example, vibration of the ground or a loud sound) to signal the approaching enemy, and relay the information securely back to the command center. The command center can then instruct these weapons to turn on their more sophisticated features to fine-tune the monitoring and action (for example, specify conditions when these weapons can detonate). In addition, such a tag can communicate with other similar tags, forming a huge ad hoc wireless network (for example, a sensor network) to exchange data among themselves. For example, a shutdown command received by some of the smart bombs at the edge of the network can be transmitted to the remaining nodes, even if they are located deep inside. In other words, these smart weapons can replace human soldiers when the situation is too risky or the requirements are so demanding that it is impossible or very difficult for a human to perform the tasks satisfactorily for a long period of time in a continuous manner.

Benefits

Flexible. Smart weapon feature combinations are virtually unlimited.
Effective. A smart weapon, unlike a human, does not tire (unless, of course, the battery runs out) when performing tasks continuously. Even if enemy forces capture such a weapon, they cannot use it as a negotiating chip.
Possible to mass produce quickly. The production of smart weapons is only limited by the production capacity of the manufacturers.

Caveats

Could be expensive. Depending on the type of features supported by the active tag of a smart weapon, the cost of producing these tags will vary. More features generally means more complexity, and hence a higher production cost.
Extensive field testing required. The active tags for smart weapons must be field tested thoroughly under different operating conditions to make sure they function reliably, which increases implementation time and cost.

Implementation Notes

Smart tags will most likely operate at a 2.45 GHz or 5.8 GHz frequency range and be equipped with a battery (typically, a 5-year battery) and have specialized on-board electronics geared toward performing specialized tasks. A typical read distance of such a tag will be more than 100 feet (about 30.5 meters). Typical data transmission rate (to its surroundings, irrespective of the presence of a reader) will be once every few seconds to once every several hours. On the other hand, a smart tag could also be a semi-active tag that can be made to transmit its data in the presence of a suitable reader. A smart tag will most probably have a variety of built-in sensors to monitor its surroundings and be able to communicate with other similar tags (thus forming an ad hoc wireless network).