25 years ago: Smart tags, the distributed memory revolution
Contactless electronic identification is now something we take for granted.
Contactless electronic identification is something we take for granted now. When Peter Hewkin of Scientific Generics wrote the following article for E&T’s predecessor magazine IEE Review in June 1989 the idea of using data-storage badges to keep track of people, animals or car bodies on a large-scale basis was an exciting and potentially revolutionary prospect.
A combination of 'evolution' in electronics-related technologies and 'natural selection' in the marketplace is making practical a new philosophy of information management which is already being used in our factories — and may in the near future find its way into our cars, our warehouses and even our wallets. This 'revolution' is expected to be every bit as fundamental in its effects as was that caused by the acceptance of barcodes as an object-identification standard in the 1970s.
'Smart tags' are small devices containing a combination of memory, data processing and communications capabilities. They can communicate without physical contact with purpose-specific stations over a range from a few millimetres up to several metres. Their shapes and sizes are application specific, ranging today from the size of a brick right down to a capsule of a few millimetres diameter. Most importantly, as uptake is increasing, prices are falling. Sophisticated read/write tags can be purchased for £20-£30, simpler read-only ones for £5-£10. Extremely simple devices, holding only a few bits of memory, are priced in pence.
The smart-tag family tree is not particularly auspicious — the technologies used in smart tags have been familiar to us for many years. However, fine tuning and innovative combination of existing technologies have provided a radical new product which can profit from economies of scale previously realised for other industries (computers, digital watches and semiconductor manufacture, to name but a few). Smart tags typically incorporate the following components: semiconductor memories (ROM, EPROM or EEPROM) and processing; inductive communication, using either stamped, etched or wound coils; and sometimes sub-miniature high-energy batteries.
The need for on-board batteries (in so-called 'active' tags) has either been reduced by cutting down current drain in standby mode to an absolute minimum, or by eliminating batteries altogether (in 'passive' tags) where energy is transmitted from the station to power the smart tag during its answer-back phase.
One particularly innovative design has included a complete microwave transponder for both data and power transmission, together with memory and support circuitry, in a thick-film hybrid design. Such a device, with neither battery nor solder joints, can survive temperatures of up to 200°C — common in car-painting lines. The selection of spectrum for powering up and communicating with smart tags is influenced by the range required, by the sizes of the tag and reader head, by cost implications and by legislation. Low frequency (bands around 132kHz and 250 kHz) is preferred where large read coils are practical and non-line-of-sight contact is required, while high-frequency (up to 2.45 GHz) is more attractive for transmitting energy over significant distances in focused beams — but this carries with it problems of reflection and absorption. Legislation currently results in different frequencies being allowed in different countries — for example, 400 kHz is popular in Japan and the USA, and 250 kHz is more common in Europe (where 400 kHz is forbidden). Low-cost read-only tags have also been developed using SAW (surface-acoustic-wave) devices, which have characteristic resonant responses to a common radiofrequency excitation signal.
Some of the simplest devices sidestep silicon-engineering and spectrum-usage issues completely. They use an advanced understanding of the properties of magnetic materials to detect and quantify the presence of such material in an oscillating magnetic field. The material concerned can be in the form of one or more tiny strips, costing a few pence, and the detection system can tolerate the close proximity of a wide range of metallic objects.
In the ever-more-sophisticated world of information management, there is a fundamental argument taking place between centralisation and distribution of memory and of processing capability. It is clear that, for certain types of application — such as office networks, EFTPOS (electronic funds transfer at point of sale) and telephone-based services — greater distribution of memory and of processing capability is advantageous.
Smart tags provide an opportunity to extend this philosophy even further. Object-related data and (if necessary) data processing can be physically carried with the object, and allow for interrogation or update wherever the object may be.
From the information-management point of view, this provides a fundamental alternative to the 'classical' approach of using an 'all knowing' system to remember the position and status of all items for which it is responsible. One strength of this alternative is that interrogation can take place anywhere without needing online communication to (or a copy of) the all-knowing database. A further strength is that the data are highly available (always present with the object) and robust — centralised failure becomes impossible. Weaknesses include the difficulty in locating a particular item on demand and the 'per-item' incremental cost of the tag.
The subtleties of the centralised/ distributed argument, together with the constantly changing cost/ benefit profile of smart tags, are resulting in a phased acceptance of tags into commerce and industry. This has started with the applications where they can add most value. To date, the market sectors where tags have already gained acceptance include:
- automobile industry — car-body identification
- personnel security — intelligent ID badges
- agriculture — animal identification
- manufacturing industry — identification of tools and work in progress
- transportation – container identification.
According to a Cutter Information Corporation survey of 1987, the world market for RF identification will grow from US$124 million in 1988 to US$1340 million in 1991, when it will account for 28.4% of all automatic identification. (The rest of the automatic-identification market is accounted for by barcode, optical-character-recognition and magnetic-strip systems.)
The final assembly of automobiles is an extremely complex process which is compounded by the requirement for fast and flexible reaction to a market demanding a large number of made-to-measure automobile variants. As a result, the fear of any sort of centralised failure which could stop the entire factory is great. A second major concern is the need for flexible reaction to local irregularities, such as machine failures or out-of-specification products.
For the above reasons, most major automobile manufacturers have already moved to a production philosophy which is driven by smart tags. The tag acts as a 'birth certificate' for each vehicle (or major subsystem) built. It carries the vital statistics of the vehicle required (these often having been transmitted direct from the retailer's showroom only a few days beforehand). The tag is attached to a skid or hook carrying the body shell, and it determines and notes the progress of operations as the vehicle is manufactured. In the simplest case, the smart tag is a read-only device that transmits a serial number to each station to determine the manufacturing option. In this way, it is just like a barcode — indeed, in the 1970s, barcodes were used for this application — but the smart tag can communicate even when covered with dirt or paint, and sometimes from behind physical obstructions.
In more sophisticated cases, a read/write smart tag can actually send instructions to manufacturing stations; for example, requesting a particular engine variant or a particular paint finish. It can also note quality statistics and request corrective action for any defects that have been noticed. Finally, it can hold the distribution details which are needed to get it to its first owner — as fast as possible.
People and animals
To date, the largest single application for smart tags has been personnel security. A smart tag close to credit-card format forms a personal security-identification badge that is both difficult to forge and machine-readable without contact. In some cases, the badge can be used to identify users while it is being worn on the lapel, or even while it is in the owner's wallet. Underground, smart tags have been tested for identifying miners in dangerous proximity to machinery. Smart tags for this sort of application cost between £5 and £10 today.
It is a simple matter to extend the idea by allowing the security tag to carry a bit map which allocates the wearer certain privileges (and denies others). By using a read/write tag, these privileges can be changed — for example, on passing through a particular door. Use of privileges can also be monitored centrally.
Exactly the same principle is applied to farm animals, using tags hung round the neck or attached to ears. Here, automatic feeding or milking, as well as tracking and theft control, are of interest, and the privileges include access to particular stalls and delivery of feedstock. Sub-miniature read-only smart tags have recently been introduced which can be injected under the skin using a syringe. This reduces problems such as loss, damage and fraud.
Manufacturing and transportation
Smart tags are also gaining ground in manufacturing industry, e.g. by providing integral data for workpieces and tools. One common application is the tagging of cutters that are shared between a number of automated-machining centres. By knowing the exact history of each cutter, its setting, wear and the danger of breakage can be optimally assessed. This, in turn, extends the useful life of each cutter and reduces the number of cutters needed to service the machining centres.
More generally, smart tags attached to pallets are used to decide the fate of the items which the pallet is transporting. This can include executing goods-inwards formalities, determining destination in a warehouse, or ownership in a baggage-handling application. The smart-tag philosophy is also of particular value for tagging items that move over large geographical areas, where online communication with a database is not practical. An example is the tagging of containers and railway wagons to register movements between countries or other economic zones. The smart tag holding details of the containers' contents can be used automatically to generate paperwork, or to arrange for payment of tax or duty.
As the prices of smart tags fall further (target price for some read-only tags is £1), they will become regarded as 'dispersible' or even 'disposable', instead of 'reusable'. For example, in the vehicle-manufacture example described earlier, when one vehicle is completed the tag normally circulates on the hook or skid to pick up a new body shell. As prices fall further, the smart log could well stay with the vehicle and become an electronic log book (noting service requirements, ownership and chassis number).
Thinking even further ahead, the tag could be read out on the road as the vehicle passes a reading station — to pick up navigation data (or to register a speeding fine!). The EC 'Drive' project is currently addressing this possibility. A major opportunity will also arise for the elimination of the paper documentation which accompanies packages, containers etc. as they are distributed. By including the distribution data in a smart tag, it will be possible to access this automatically for identification, sorting, registration etc. This idea can extend from airline baggage handling to just-in-time supply of components to a manufacturer using a delivery van.
Another significant influence is the use of the smart tag in financial transactions. We are familiar with token cards and credit cards using holography or magnetic stripes to store read-only data. By including sophisticated processing (e.g. for data encryption or analysis of biometric data) and increased memory on so-called smart cards (superstructures with integral keypad and display are also under investigation), it may become possible to pay for services without even having to reach for your wallet — the card in the wallet could transmit the necessary information to the supplier of the service!
As with barcodes, the crucial issue for smart tags will be the creation of open standards for usage, communication and data storage. This is still some way off as smart tags are now gaining acceptance as a product genus which requires and deserves specially tailored legislation. It is to be expected that smart tags will be taken up most readily in the manufacturing and security sectors, where their robustness to rough environments is particularly attractive. Here they already offer valuable advantages over and above barcodes. Again, as with barcodes, a significant strategic move will be needed from a group of far-sighted companies to bring the revolution to the people.
Originally published in IEE Review, June 1989 (Volume 35, Issue 6)
|To start a discussion topic about this article, please log in or register.|
"Who's getting the best engineering education? And what did your careers advisor suggest you do when you leave school?"
- What to Specialise in Electronics Engineering?? [03:02 am 03/04/14]
- Britain to have just one remaining coal pit by the end of 2015 [01:11 am 03/04/14]
- LV Generator Star point earthing - UK [08:35 pm 02/04/14]
- East West Rail - the Oxford to Bedford route [07:33 pm 02/04/14]
- Small nuclear power [06:06 pm 02/04/14]
The essential source of engineering products and suppliers.
Tune into our latest podcast