4.8 Smart Cards and Terminals

John M. Gill and Kelvin Currie

Introduction

Telecommunication terminals are becoming increasingly sophisticated and can now offer a bewildering number of choices to the user. To handle this increased number of choices, the terminal often incorporates a sophisticated interface which can cause problems for users who are elderly or have a disability. However some of these terminals give the potential for modifying the interface to meet the needs of the individual user.

To select a preferred interface, the user could simply press a button or select from a menu on the screen. This is unlikely to find favour with service providers if it significantly increases the time taken to undertake the transaction, but it may be viable for simple operations such as increasing audio amplification on a public telephone. The user's preferences could be stored on a card. With the magnetic stripe or optical cards there is very limited spare capacity for storing this information, but this method has been used successfully for storing the user's preference for language (e.g. English or French). A smart card has fewer restrictions on storage capacity so it appears to be ideal for this purpose, as long as some international standard is agreed for the coding of this information on the card.

A smart card is a credit card sized plastic card incorporating an integrated microcircuit. This circuit holds information in electronic form that can be easily, securely and accurately accessed by a range of terminals.

Smart cards are one of three main types:

• memory only - often used as pre-payment cards for public telephones;
• microprocessor - this adds the possibility of in-corporating security features (e.g. for banking applications);
• contactless - eliminates the need to put the card in a reader which makes this type attractive for public transport applications.

In an ideal world, the user would be able to select and store their preferred interface any time they use the card at a terminal. However practical constraints may restrict this choice to being at the time of issuing the card.

Preferred customer verification method:

• Personal Identification Number (PIN) - this can be a problem for people with dyslexia since they may remember the correct digits but in the wrong order;
• fingerprint;
• hand geometry;
• voice recognition;
• dynamic signature verification;
• retinal scan;
• visual recognition of the user (e.g. face).

Although some of these biometric methods are used in specialized application areas, it is likely to be some years before they are used on telecommunication terminals for use by the general public. However, when they do come into general use, it is highly desirable that the user has a choice of verification method since some groups of disabled people will be excluded if there is no choice.

Preferred input:

• more time - this is particularly important for many elderly persons. On a public telephone, this could involve storing the number being dialled and then sending it at a press of a function key;
• keyboard only - for instance, a blind person may prefer not to use a touchscreen;
• touchscreen only - this may be easier for people who are not used to operating keypads;
• speech recognition - this may be restricted to giving basic instructions to the terminal.

Preferred operation:

• reduced functionality - this is for limited use of the terminal such as automatically dialling a pre-stored number on a telephone card (a useful feature for some intellectually impaired persons);
• pre-set amount (for an Automatic Teller Machine) - after entering the PIN, the cash dispenser would automatically issue say £100 (this is a mode which might be surprisingly popular among many elderly users);
• simple instructions on touch screen - this is a feature which might be useful for customers who are technophobic.

Preferred output:

• large characters on screen - this includes size of characters as well as foreground / background colours or the choice of font;
• audio - this can vary from beeps to indicate the acceptance of an instruction, to speech feedback on key pressed (but not for the PIN or password), to speech prompts, to speech output of information normally displayed on the screen. Speech prompts can usually be achieved by using stored speech which is inexpensive and of good audio quality; full vocabulary speech synthesis is of lower quality which may be unacceptable to the occasional older user;
• for audio systems such as the telephone, it would be desirable to be able to select the amplification for each frequency band - this would go some way towards compensating for a hearing loss in a parti-cular frequency band;
• maximize use of icons - this may be desired by persons who are illiterate or who do not understand any of the languages available on that terminal;
• Braille display - these are expensive (adding a few thousand pounds to the cost of a terminal) and it is estimated, for example, that only 19,000 people can read Braille in the UK. However there may be some special application areas where the cost is justified.

If close contactless card:

• No card insertion - this would help people in wheelchairs who cannot reach the slot for the card reader, those with hand tremor or arthritis, and blind persons.

If distant contactless card:

• Locating appropriate terminal (e.g. audible location signal) - this would help blind persons find the terminal
• Trigger message - for instance, the user's card could trigger an audio message at the entrance to a bus giving the destination of that bus;
• Preferred height of keypad - a wheelchair user would find it very helpful if the keypad could automatically adjust to a suitable height;
• Remote input device - a personal terminal with a radio or infra-red link to the fixed terminal would permit the custom designing of special interfaces. This could provide an appropriate interface for a severely disabled individual, but the customization would be expensive.

Applications

Expanding on some of the points given above, with particular reference to the use of telephones, it can be seen that one of the major barriers to using the telephone is hearing impairment. Loss of hearing acuity is probably one of the most widespread disabilities in the general population, and is often age related.

The traditional answer to hearing loss which falls short of profound deafness, i.e. hard of hearing people with a loss up to 80dB or so, is to provide additional amplification. This method has its limitations, but can prove useful for a large number of people.

The limitations of straight amplification are well known. In particular, it is not sufficient to regard any hearing loss as being a simple attenuation. It is more usual for there to be a frequency dependent component to the loss, or other distortions to the perceived sound. Indeed most hearing aids have a gain frequency adjustment to meet the individual needs of the wearer. By using digital signal processing, it is now theoretically possible to produce a telephone that could have infinite variability of its receive frequency response in a similar way to a graphic equalizer on a hi-fi.

Naturally there are limits to the permissible, or achievable permutations, in particular, the received speech is conditioned by the limitations of the transmission network. However there may be great benefits here for people with quite severe hearing impairments if frequency/gain variation were to be married with additional gain (and the inductive coupler).

Unlike a hearing aid, the telephone is a mass-market instrument, serving a market of approximately 26 million connections in the UK alone. Specialist telephones serve a much smaller market, approximately 500,000 hard of hearing people in the UK, and consequently suffer the prospect of higher prices.

The market size improves if the totality of the European Union market is considered, giving a comparable figure for hard of hearing people of 40,000,000. However the market here is not yet truly barrier free or totally liberalized, and as a consequence is still fragmented. Smart cards may be able to help this situation to a large degree by removing the specialism from the telephone and placing it in the hands of the customer.

Although the technology and electronic components exist to produce variable response telephones, how would it be possible for the customer to use this vast range of adjustment. It is very unlikely that the supplier would (or should) be in a position to provide a customization service. Even if this were to be the case, it is likely that such a service would be very costly.

This is where the uses of smart cards may be introduced. If all telephones were to contain a smart card reader and an intelligent Digital Speech Processor (DSP), then it could be customized by the insertion of the card at any time. Without the card, the telephone would operate to the standard parameters that all telephones are set to. However, with the insertion of a suitably programmed smart card, the gain/frequency response curve could be adjusted to meet the needs of the card holder. The required information could be programmed on to the card by a qualified audiologist at any time, and could be kept up to date in the same way.

The card need not only hold information on the customer's hearing requirements. There may be a facility for storing data like his or her most commonly called numbers. This would give real customization of any telephone whether public or private.

For customers with sight difficulties, it may be possible to use the card to customize the telephone instrument in other ways:

• for example, a larger display character may be appropriate, or back lighting may be activated automatically;
• voice guidance in public payphones, or those in hotels, may be made available by the customer requirements on the card;
• voice confirmation of the number dialled, or complete compose and send, (as in mobile telephones) may be activated by the card. These functions remaining dormant if not required by the customer.

Even outgoing speech may be "pre-conditioned" by a telephone, under the instruction of the card, so that it becomes more intelligible to the person called. All of this may be handled by non-specific terminals so that the facilities can be provided in any location at any time, without additional cost to the customer requiring the extra facilities.

Display telephones or "screenphones" are due to arrive in large numbers soon. The reason for this change of direction in telephony is the proposal to provide services like home shopping and home banking via the telephone. These services may even make use of smart cards for cash purposes, such as the Mondex electronic cash system which is being trialled in Britain during 1995.

The major change here is the provision of a visual display on a large number of telephones, both private and public, which can provide text transmission in one direction at least. This will be the first step toward a more universal distribution of text telephones in the mass market.

The part that the smart card may play here is manifold. There are a number of difficulties with the provision of effective text communication today, not the least of these being the difference in line standards. The customers card could be used to program the "screenphone" to provide the line standard that he or she requires, assuming that the hardware is capable of handling it.

In addition, the card could ensure that if the local relay service is used, e.g. Typetalk in the UK, the call could be routed via the relay automatically. In effect the card could contain the relay number and the customer's log on details so that the call could be dialled by the customer in the normal way. If combined with a payment option, the card could relieve some of the relay billing problems for people on the move.

Lastly, there are network services. These can take a number of forms, such as network announcements, or voice guidance instructions for chargecards etc. The use of the card in a suitably equipped telephone would be able to pre-condition the network to provide the network information in the form required by the customer. For example it may be required in text, or in voice but spoken more slowly, or in another language.

At the end of the day, the smart card can effectively customize telephony for everybody, but at no extra cost to those who need additional facilities (the cost of the telephone instrument being standard to all with the extra facilities being activated when required). The major task is to get the parameters into the design of the card at an early stage so that the extra facilities are not overlooked, and unnecessary costs are thus avoided at a later stage.

Conclusions

Smart cards offer exciting possibilities for improving access to telecommunications services by disabled and elderly persons. However it will soon be essential that there is some agreed standard for recording these preferences on the smart card.

References

COST 219, (Ed.) Gill, J.M. (1994). Proceedings of the Seminar on Smart Cards and Disability, November 1994, Edinburgh. ISBN 1 86048 003 9.

GILL, J. M. (1994). Smart Cards: Meeting the Needs of Elderly and Disabled People. ISBN 1 86048 002 0.

SILVER, J. H., GILL, J. M. and WOLFFSOHN, J. S. W. (1994). Text Display Preferences on Self-Service Terminals by Visually Disabled People. ISBN 1 86048 001 2.