Introduction

Intuitively, a Smart Card is a plastic card with an embedded microprocessor and a memory large enough to store programs made by the card-issuing company. The exact structure of a smart card is specified by interational standards: the plastic card must have dimensions of 85.60mm x 53.98mm x 0.80mm and must be able to bend a specified amount without damage for instance. A printed circuit and an integrated circuit chip (microcontroller) are embedded on the card. As silicon cannot handle bending very well, the chip must be very small. The printed circuit is a thin gold plate that provides electrical contacts to the outside world and also protects the chip from mechanical stress and electrical static.

Surprisingly, Smart Cards are not a recent invention. In 1974, the French journalist Roland Moreno devised a revolutionary payment system - an early electronic stored-value application mounted on a ring. (Curiously enough, almost a quarter century later, manufacturers returned to the idea of a smart ring, this time as Java Ring). Later the same year, the first epoxy cards were produced. However, they were not planar yet.

In 1975, the first card in the credit-card format with the chip and its contacts on one side was made by the French company CII-Honeywell-Bull. Its first customer orders were shipped two years later.

All Smart Cards have memory and a processor, but some are almost exclusively the former. These 'memory cards' are not the subject of this site as they aren't very 'smart'.

Smart Cards do not have an internal power source but need power to operate. Therefore they only operate when in the presence of a Card Accepting Device (CAD) which supplies their power requirements. Most Smart Cards come into physical contact with CADs while others do not. However, they have practically the same architecture and will not be distinguished in this site.

A fundamental change in Smart Cards is in the language they are programmed. In a manner analogous to the development of computer systems, there has been a move from assembly language to higher level languages, especially Java.

Life-cycle of a Smart Card

• Stage 1. The processor is made and tested by a chip-manufacturing company. A fabrication key (FK) is added to protect the chip from fraudulent modification until the next stage. The FK of each chip is unique and is derived from a master manufacturer key.
• Stage 2. The chip is mounted on a plastic card and the FK replaced by a personalization key. This is protected from later change by a personalization lock and physical memory access instructions are disabled.
• Stage 3. The card issuer (e.g. a bank) writes data files and application data to the card. Information regarding the identitiy of the card holder (e.g. a client of the bank) is stored as well.
• Stage 4. The card is used by the card holder. Access to its information is limited by the application's security policies.
• Stage 5. Card is invalidated either by blocking all PINS or by writing an invalidation lock. In the second case, only read instructions may remain active for analysis purposes. In the first case, neither writes nor reads are allowed.

References:

Cardshow.com
Orlin Grabbe's Smart Cards and Private Currencies article
Litronic's Introduction to Smart Cards