Identification system - Patent Review 4864292

Description Submit all comments and votes

BACKGROUND

The present invention relates generally to the field of electronic locking or access-control systems where a user wishing to gain access to a restricted area, or to some service such as the use of a cash-dispensing machine, presents to a receiving apparatus a token which transmits identification data indicative of the user's authority to gain such access. More particularly, the invention relates to a contactless-identification system of the kind where the receiving apparatus includes an alternating magnetic field generator and the token includes a transponder powered by induction from said field.

SUMMARY

As essential component of the transponder in a system of the kind indicated above is a coil or antenna through which the power for operation is induced and through which the identification data is transmitted. In prior art locking system in which this principle of operation has been used at least the antenna (and possibly other components, in particular capacitors) is manufactured and assembled as a discrete component separate from the rest of the transponder circuit, to which it must be electrically connected when assembled to the token. The present invention, on the other hand, proposes in one broad aspect a token for use in an electronic locking or access-control system of the kind indicated above of which the transponder is manufactured as a complete integrated circuit including the antenna. By eliminating the need to make connections between at least a separate antenna and other parts of the transponder circuit the reliability and manufacturing cost of a token in accordance with the invention should be respectively increased and decreased in comparison with those used in prior art systems.

BRIEF DESCRIPTION OF DRAWINGS

This and other features of the present invention will now be more particularly described, by way of example, with reference to the accompanying drawings, in which:

FIG. 1 is a schematic block diagram of a system operating in accordance with the invention as applied to physical access-control;

FIGS. 2-6 are schematic block diagrams of various circuits illustrating the development of a suitable transponder in accordance with the invention;

FIG. 7 illustrates schematically an example of the implementation of transponder as an integrated circuit; and

FIGS. 8A-8D include several schematic block diagrams illustrating the development of a suitable transmitter for use in the system.

DETAILED DESCRIPTION OF DRAWINGS

Referring to FIG. 1, the illustrated access-control system comprises a receiving apparatus 1 to which users present an individual identification token 2 when entry through a normally-locked door 3 is required. When a token is presented to the apparatus its presence is sensed by a detector 4 which instructs the central logic unit 5 to activate a transmitter 6. The latter energises the primary coil 7 of an alternating magnetic field generator comprising a ferromagnetic core 8 which forms a magnetic loop with a small gap 8A. When the token 2 is presented to the apparatus an integrated circuit transponder 9 carried by it is placed within the gap 8A and is subjected to the alternating field concentrated by the core 8. As more particularly described below the transponder is thus caused to transmit identification data which modulates the magnetic field in the core 8 in amplitude or frequency, or adds a magnetic field with a different frequency, and can be detected by a receiver 10 associated with the transmitter 6. The received data is compared by the logic unit 5 with data stored in an associated memory 11 and if found valid an electromagnetic door release 12 is actuated to permit entry.

The transponder circuit 9 may operate on a principle as illustrated in FIG. 2. An antenna coil L receives the magnetic energy from the generator in the apparatus 1 and provides the circuit with power to function. Diodes D1-D4 rectify the induced voltage in L, capacitor CS buffers the supply voltage VDD, while zener diode ZD limits the maximum supply voltage. The frequency of the alternating magnetic field, suitably divided, acts as a clock for logic circuitry 13 which drives a shift register 14 containing the identification data from a memory 15. This information is transmitted by intermittent closing and opening of a solid state switch S. When S is closed the resonance frequency of the input circuitry containing capacitors C1, C2 changes. This leads to a variation of the amplitude of the magnetic field and can be detected by the receiver 10.

Problems arise, however, in integrating a circuit of this kind with capacitors C1, C2 and CS onto a single monolithic device. In general only very small capacitors, with a relatively poor tolerance on the capacitance value, can be practically integrated onto a chip because of the large surface area they require, and in the present instance the coil L also will occupy a considerable proportion of the chip area. This means that the supply decoupling will be very limited because of the small possible values of CS while the resonance frequency of the input will have a wide tolerance in the absence of any economical form of "on chip" trimming.

To avoid the use of C1 and C2, a non-resonating input as shown in FIG. 3 could be used. Switch S short-circuits coil L, which leads to an amplitude modulation of the magnetic field, which can be detected by the receiver 10. In addition, if the transmitter 6 consists of an oscillator of which the frequency is (partly) determined by the inductance of coil 7, modulating the magnetic field amplitude will also modulate this inductance and therefore the frequency of the magnetic field. Detection of this data within receiver 10 can therefore also by realised by using FM-detection instead of AM-detection.

Two drawbacks occur, however, in that when S is closed the circuit will no longer be provided with power and also the clock signal will disappear. Even when a series resistance R is used, the supply voltage will still decrease significantly; otherwise, by increasing R, a significant decrease of the modulation of the magnetic field will occur. A solution to this problem is shown in FIG. 4. Switch S now short-circuits the coil only during one half of a period of the magnetic field. In this way, no significant decrease of the supply voltage occurs when S is closed. By connecting the clock input of the logic circuit with point A, the clock signal will constantly be available whether S is closed or not.

An alternative scheme for data-transmission from the responder is illustrated in FIG. 5 where the coil L is short-circuited during the first half period of the magnetic field or during the second half period of the magnetic field, dependant on the logic state of the output. In this way a magnetic field with the same frequency as the output signal is added to the existing magnetic field instead of modulating this field, so that the AM- (or FM-) demodulator situated in the receiver 10 can be replaced by a high-cut filter.

The power supply capacitor CS can be kept small by using a high supply frequency--say 10 MHz--and by keeping the power consumption low. The static power consumption can be very low if the circuit is realised in a CMOS technology, but at high frequencies (like the used input clock frequency) the power consumption of CMOS becomes considerable. This considerable power consumption is due to the transitions of the logic levels inside the circuit, so if it is arranged that transitions only occur on the positive going edge of the input clock, the power during the transitions can directly be derived from L instead of the supply capacitor CS. During the transitions the supply current will first be obtained from CS, but because CS is small, the supply voltage VDD will decrease rapidly, until VDD is equal to the momentary voltage on A minus the forward voltage drop of D4. From that moment L will provide the circuit with supply voltage and current during a large part of the positive period of the voltage on A. A voltage divider R1, R2 might be used (as indicated in FIG. 4) to provide a higher voltage on A at the moment the clock input is 'triggered' and multiple transitions inside the circuit occur.

A schmitt-trigger clock input will also enhance reliable operation.

A further improvement which makes detection of the transmitted data easier is the use of a modulator such as indicated at 16 in FIG. 6. To explain, when token 2 is detected by detector 4, fast retrieval of the identification data is desirable for reasons of fast door release and minimising power consumption. When the transmitter 6 is switched on a lot of low frequency components with a large amplitute occur for a while in the received signal. This makes easy and reliable detection impossible for some time directly after switching on the transmitter. To minimise this time delay a low-cut filter should be used with a cut-off frequency as high as possible, but not higher than the lowest frequency components of the signal transmitted by the responder. Therefore the lowest frequency components of the transmitted signal should be as high as possible. If the data output of shift register 14 is applied directly to S, as indicated in FIGS. 4 and 5, it will happen that S is closed or opened for a long period if the data contains a succession of one's (or zero's next to each other. Therefore the transmitted data will contain low frequency components, which restrict the cut-off frequency of the low-cut filter inside the receiver 10. If instead the data is modulated onto a subcarrier (using a phase or frequency modulation principle), and this modulated subcarrier is used to switch S, the difference between the maximum and minimum time that S is closed or opened can be highly reduced. Thus, by using the modulator 16 the lowest frequency components contained by the transmitted signal can become more than an order of magnitude higher; therefore the cut-off frequency of the low-cut filter inside the receiver 10 can be chosen an order of magnitude higher which leads to a much faster and reliable detection of the transmitted data.

For reasons of power consumption of the transponder, VDD must be as low as possible. This makes it desirable that the input clock divider and therefore the whole final circuit is realised in a high speed CMOS technology. If for other reasons, standard CMOS would be highly preferred, an alternative circuit is possible with an on chip oscillator. Both alternatives are indicated at 17 and 18 respectively in FIG. 6.

Turning to FIG. 7, this indicates one possible implementation of the entire transponder circuit 9 on a chip which may measure approximately 4.times.4 mm. The active components of the circuit are provided in the central area 19 surrounded by the turns of the antenna coil L. Contact pads for use in programming the identification data into the memory 15 are indicated at 20. If, however, programming can be achieved by a contactless method, making use of the coil L to provide the circuit with sufficient power and to transfer the data to be programmed, area 20 can be used for the corresponding programming logic.

The transponder circuit 9 is schematically illustrated in FIG. 1 as mounted in the shank of a token 2 shaped to resemble a conventional key and this represents a convenient implementation of a personal identification device. In principle, however, the structure of the substrate by which the transponder is carried is open to considerable variation--and could for example be in the form of a card--so long as the token and receiving apparatus are appropriately mutually configured to place the transponder correctly in relation to the field generator when used.

A transmitter 6 configuration which would normally be applied is shown in FIG. 8A. The magnetic fieldstrength inside the airgap 8A is to a first approximation a linear function of the current Ip multiplied by the number of primary windings Np. Because the system is battery operated, the total supply voltage and current are limited, while the power dissipation must be kept to a minimum. This has led to the following considerations:

(1) The number of windings of the coil L on the transponder chip is restricted and must be kept to a minimum for reasons of minimising chip area and series resistance of the coil. On the other hand the total induced (secondary) voltage in the coil on the chip must be maximised. Because the secondary voltage is inversely proportional to the primary number of windings Np, Np should be minimised. The smallest practical value of Np is one (FIG. 8B).

(2) When Np is chosen to be one, the inductance and therefore the impedance of the primary coil is very low and Ip and the power to be delivered by the battery becomes excessive, which is not allowable. A solution is found by using a high frequency (say 10 MHz) and by tuning the inductance of the primary coil using a capacitor Cp (FIG. 8B). Although Ip does not change, the current Ig to be delivered by the generator and therefore also the current and energy to be delivered by the battery is decreased by an order of magnitude.

(3) To obtain the stated severe decrease of the generator current Cp must be tuned, which is undesirable during mass production. A solution is to make the tuned network LpCp the frequency determining part of a generator, formed by LpCp and a transconductance amplifier (FIG. 8C). In this way the frequency is automatically tuned to LpCp.

(4) Because the excessive current Ip still flows through the capacitor Cp, this capacitor must be a high quality, low loss and therefore an expensive and large device. A solution to this problem is shown in FIG. 8D, where the primary coil is divided into two parts. The inductance seen between point A and B is highly increased which leads to a much lower value for Cp and Ip. Therefore Cp can become a standard type capacitor.

* * * * *