WO2001017407A1 - Electronic mailbox for receiving a package containing an electronic postage stamp - Google Patents

Abstract

Packages contain circuitry which stores a virtual stamp and other information, such as identity of the sender, identity of the recipient, contents of the package, and so on. A locker at a recipient's location comprises a transceiver to communicate with the circuitry, to secure information such as the identity of the sender or the contents of the package and means to transmit a message to the recipient, informing the recipient of the arrival of the package.

Description

ELECTRONIC MAILBOX FOR RECEIVING A PACKAGE CONTAINING AN

E LECTRONIC POSTAGE STAMP The invention concerns an electronic circuit, resembling a so-called "smart card," which is affixed to an envelope or package, and replaces several types of printed matter ordinarily applied to the package.

Envelopes and boxes, together called packages herein, are commonly used to transport articles. Both (1) government agencies, such as postal services, and (2) private carriers transport the packages from origin to destination.

The packages are equipped with numerous types of printed material, or markings, and each serves a different purpose. One group of printing identifies the destination. Another group identifies the origin, so that, if the destination cannot be located, the package can be returned to its origin.

A third group indicates a serial number which the transporter assigned to the package, for the transporter's internal record-keeping. A fourth group may indicate an identifying number assigned by the originator.

A fifth group may indicate whether, and how much, money has been paid to cover the cost of transport. So-called "stamps," printed either by a postal service or a privately owned postage meter, provide examples of the fifth group.

A sixth group may state a characteristic of the contents of the package, such as "fragile," "flammable," and so on. A seventh group may indicate shipping instructions, such as "expedite." Additional types of markings can be envisioned.

All these markings must be applied to the package, and the application steps represent a consumption of time and effort. In addition, in many instances, humans must interact with the package at both the origin of its transport, and at the destination. Further, a pair of humans is generally required in each interaction. For example, when a person ships a package using a private shipping service, the person delivers the package to an agent, and completes paper forms. When the shipping service delivers the package to the recipient, a delivery agent generally requests a signature from the recipient.

It is not seen that involvement of these two pairs of persons is either necessary or desired.

In one form of the invention, a type of "smart card" is affixed to a package or envelope. The smart card carries all information required by the originator, carrier, and recipient of the package. Various cryptographic techniques are implemented to assure that the data contained within the smart card remains intact and correct.

In another form of the invention, an automated locker accepts packages from a shipper, communicates with the smart card affixed to the package, and begins the shipping process automatically.

In another form of the invention, an automated locker located at the destination of the package accepts the package and issues a receipt for it.

The invention accordingly provides apparatus for receiving packages, comprising: a transceiver for communicating with incoming packages and a delivery agent; a locker containing a door and a lock on the door; means for authenticating the delivery agent and for opening the door when authentication occurs; means for receiving data from the package; and means for transmitting information about the package to a remote location.

Preferably, the means authenticates the delivery agent by receiving a code from the delivery agent and validating the code with a remote party. Preferably, the information transmitted includes information about the contents of the package.

Preferably, the transceiver communicates using infra-red radiation as a carrier.

Preferably, the infra-red radiation is transmitted to a second transceiver associated with the package, and the radiation must travel through at least one sheet of paper to reach the second transceiver.

Embodiments of the invention will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 illustrates a package 3, a smart card 6, and a station S which communicates with the smart card.

Figure 2 illustrates data which is stored in the smart card 6 of Figure 1.

Figures 3 A, 3B, and 3C show three views of a surface-mount integrated circuit, IC.

Figures 4 and 5 illustrate mounting systems for mounting the IC of Figure 3 to a package 3.

Figure 6 illustrates another type of smart card C2 which may be attached to the package 3.

Figure 7 is a cross-sectional view of transistor pair fabricated atop a silicon wafer, and shows the depth of the wafer occupied by the pair.

Figures 8A and 8B illustrate fabrication steps used in producing a thin IC.

Figure 8C illustrates a carrier in which the IC of Figure 8B is mounted.

Figure 9 illustrates another mounting system for mounting the IC of Figure 8B.

Figure 10 illustrates a package 3 which carries, either internally or externally, a single-board computer 100, together with a station S which communicates with the computer. Figures 11, 12, 12A, 12B, 12C, 12D, 12E and 13 illustrate approaches to delivering power to the invention.

Figure 14 illustrates a communication network used by the invention.

Figure 15 illustrates an approximate computation of a radius of curvature for carrier 105 when it bends.

Figure 16 illustrates an approximate computation of the maximum angle B allowed before point P contacts the IC 100.

Figure 17 are scale drawings which show the relation between thickness T of a carrier and aperture A of the carrier, in which the IC is mounted.

Figure 18A illustrates a cantilevered silicon wafer SIL.

Figure 18B is a model illustrating how a small bending force FB can require a large

opposing force FO.

Figures 19, 20, and 21 apply the same model of Figure 18 A, in showing how a cube possesses greatly improved resistance to the bending force FB.

Figure 22 illustrates a cube C of silicon, and a flexible sheet 230 containing conductive traces TR.

Figure 23 illustrates the cube C of Figure 22, sandwiched between the sheet 230 and a

plate 235.

Figure 24 is similar to Figure 23, but shows multiple cubes.

Figure 25 illustrates another mounting system for the IC 100.

Figure 26 is a cut-away view of the apparatus of Figure 25.

Figure 27 is a perspective view of the apparatus of Figure 25.

Figure 28 illustrates another embodiment.

Figure 29 shows the embodiment of Figure 28 in partially assembled form. Figure 30 shows the embodiment of Figure 28 is cross-sectional view.

Figure 31 shows an apparatus 410, such as that of Figures 23, 25, or 28, attached to a card 405, and inserted into an envelope 400. The apparatus communicates through the paper walls of the envelope, using infra-red or radio frequency energy.

Figure 32 illustrates multiple ICs 100 fastened to a membrane M.

Figure 33 illustrates column-spacers COL attached to the membrane M of Figure 32.

Figure 34 shows the apparatus of Figure 33, supported in a frame 305.

Figures 35 and 36 are cross-sectional views of the apparatus of Figure 34.

Figure 37 illustrates another embodiment.

Figures 38 and 39 illustrate how a gripper GR can be equipped with contacts CON2, which mate with contacts CON contained on the smart card 500.

Figure 40 is a functional diagram of one architecture which implements the invention.

Figures 41 - 44 illustrate a mounting system for mounting a cylindrical wafer of silicon, about 1/4 inch in diameter.

Figures 45 and 46 illustrate a pedestal mounting system. A dual-pedestal mounting system can be used, wherein an additional pedestal 775 or 780 is placed on the top side of the IC 100.

Figure 47 illustrates a three-point mounting system.

Figure 48 illustrates one form of the invention, in exploded view.

Figure 49 is similar to Figure 47, but illustrates components 100, 805, 810, and 815 in assembled form.

Figure 50 illustrates the components of Figure 47 in assembled form.

Figures 51 - 53 are cross-sectional views of the apparatus of the type shown in Figure Figure 54 illustrates an envelope modified for carrying the smart card 6 of Figure 1.

Figure 55 illustrates a smart card 6 enclosed within an envelope 980 attached to a package.

Figures 56 and 57 illustrate one form of the invention.

Figure 58 illustrates the invention of Figures 56 and 57 contained within an envelope 990.

Figures 59 - 62 illustrate a storage box 1000, which may be located at a package depositary, for receiving packages from senders, or at a destination, for receiving packages from shipping agents.

Figure 63 illustrates the storage box 1000, with added components.

Figures 64 - 67 are flow charts illustrating logic implemented by one form of the invention.

One Form of Invention

Figure 1 illustrates a package 3 to which is affixed a "smart card" 6. Smart cards are credit-card sized objects, which contain digital memory and a simple logic, or a microprocessor. Smart cards are known in the art, and at least one industry standard, International Standards Organization (ISO) 7816, has been developed stating one set of specifications for smart cards. The following U.S. patents describe some of their features, and are hereby incorporated by reference.

PATENT NO. LEAD INVENTOR ISSUE DATE

4,605,844 Haggan Aug 12, 1986

4,795,898 Bernstein Jan 3, 1989

4,798,322 Bernstein Jan 17, 1989 ,825,052 Chemin Apr 25, 1989 ,827,113 Rikuna May 2, 1989 ,831,245 Ogasawara May 16, 1989 ,841,133 Gercekci Jun 20, 1989 ,845,740 Tokuyama Jul 4, 1989 ,849,615 Mollet Jul 18, 1989 ,864,110 Guillou Sep 5, 1989 ,874,935 Younger Oct 17, 1989 ,888,773 Arlington Dec 19, 1989 ,916,296 Streck Apr 10, 1990 ,924,075 Tanaka May 8, 1990 ,931,991 Cvijanovich Jun 5, 1990 ,968,873 Dethloff Nov 6, 1990 ,982,069 Kayanakis Jan 1, 1991 ,200,601 Jarvis Apr 6, 1993 ,241,161 Zuta Aug 31, 1993 ,286,957 Defrasne Feb 15, 1994 ,311,396 Steffen May 10, 1994 ,322,989 Long Jun 21, 1994 ,412,253 Hough May 2, 1995 ,457,747 Drexler Oct 10, 1995 ,480,842 Clifton Jan 2, 1996 ,517,460 Yamaguchi May 14, 1996 ,613,001 Bakhoum Mar 18, 1997 5,640,306 Gaumet Jun 17, 1997

In Figure 1, a reader R located at a station S, and preferably of the non-contact type, communicates with the smart card 6, in order to read and write data to, and from, the smart card 6. Such readers are described in the patents identified above. The station may be located at an intermediate point on the route of the package 3, or may be located at the beginning of route of the package, where the package is deposited by a customer to a transport service.

Figure 2 illustrates the types of data which can be stored within the smart card 6. The data includes (1) a serial number for the package, (2) a description of the contents of the package, (3) an identification number which identifies the sender, and (4) cost information, such as the cost of shipping, and whether the cost was paid, and (5) size and weight. In addition, the other types of data discussed in the Background of the Invention can be included. Specific uses for the data are described in a later section. These data can be loaded into the smart card 6 by the reader R of Figure 1.

The smart card is preferably of the non-powered type. However, battery-powered smart cards can be used.

Second Form of Invention

In some situations, the full power of a smart card may not be required, so that a card containing only memory is used. Memory chips have become sufficiently small that a chip may be affixed directly to the package 3.

Figure 3 illustrates three views of a surface-mount integrated circuit, IC. The thickness T, shown in Figure 3C and corresponding to dimension D in Figure 3A, is about 1/10 inch, or about 2.5 millimeters. As a point of reference: an ordinary sheet of typing paper is about 0.004 inch thick, so that the thickness of this IC equals the thickness of about 25 sheets of paper.

This IC represents a memory chip, such as the 256K serial EEPROM (Electrically Eraseable Programmable Read Only Memory) surface mount IC available from Microchip Technologies, Inc., Phoenix, Arizona. The IC of Figure 3 is diagrammatic only, and the actual EEPROM used will not necessarily contain the fourteen pins shown in Figure 3B.

The IC is mounted in a carrier C as shown in Figure 4. The carrier C comprises a flat, stiff, bottom surface B, and a protective ring RR, which surrounds the IC, forming a well W. The pins of the IC, labeled 1 - 14 in Figure 3B, are connected to respective conductive pads P, and one connection is represented by dashed line L. The carrier C is mounted to the package 3, as by a suitable adhesive (not shown).

The pads P are used to read, and write, data to the IC. For example, contact probes PR, connected to external read/write logic, such as that contained in a microcomputer (not shown), are applied to the pads P. Two of the probes PR1 and PR2 apply electrical power. Other probes handle addressing, handshaking, and data transfer. When the probes are removed, the IC, of course, loses power. Figure 5 illustrates a variation, in which three pads P are shown. Two pads PI and P2 receive electrical power, and the third P3 represents an infra-red detector, connected to an infra-red transceiver (not shown). Such transceivers are discussed in a later section.

Third Form of Invention

Figure 6 illustrates another form of the invention. A circuit C2 is attached to the package 3. The circuit C2 contains an infra-red transceiver XCVR 1. Such transceivers are available, for example, from Calibre, Inc., 1762 Technology Drive, Suite 226, San Jose, California, as model numbers CHX 1000 and CHX 1010. The latter transceiver measures about 10 mm in length by 2.7 mm in height. The size of the transceiver can be reduced by applying procedures described later. These transceivers consume about 120 milliamps at 3 volts, and have an operating range of about 2 meters. They utilize the RS-232 data protocol, so are directly compatible with standard computers.

The transceiver XCVR 1 reads data from, and writes data to, a memory MEM, such as the 256K serial EEPROM described above. The memory MEM includes a DECODER. The DECODER acts as a multiplexer, allowing a very small number of I/O lines to connect to the very large number of memory cells.

The circuit C is powered by a solar cell SOL, such as the SA series manufactured by Solarex, and available from Digit-Key Corporation, Thief River Falls, Minnesota. One such cell measures 172 x 350 x 13 mm (6.8 x 13.8 x 0.5 inch) and produces two watts in sunlight. One may reduce the size of the cell by using artificial light which is more intense than sunlight. Preferably, the cell is about 1 x 2 inches, or the size of a large postage stamp, and powered by a light source producing appropriate output. The size of the solar cell SOL may be reduced my another expedient. The cell SOL may be used to collect light over a period of time and charge a capacitor CAP. When data transfer is to take place, the capacitor is used to power the circuitry C. So-called "super capacitors" are available, providing very large capacitance values, such as one Farad, in small packages. These are used as back-up power supplies for computer memory.

As a simple example of this expedient, assume that the components within the circuit C consume one watt. Assume that the data transfer rate is 9600 bits per second, and one second of transfer is required. If the two-watt solar cell of the type identified above is used, it can be reduced to 1/20 of its size, by illuminating it for 10 seconds, and charging capacitor CAP during that period. The capacitor CAP then delivers the single watt for the single second of transmission.

In Figure 6, the station S is equipped with a matching transceiver XCVR 2, which communicates with the other, XCVR 1. The station also contains a light source LS, for illuminating the solar cell SOL. Preferably, a baffle B prevents the light from interfering with the infrared transceivers.

Alternately, power can be applied to the circuit by a direct, hard-wired connection, using probes of the type shown in Figure 5 (not shown in Figure 6).

Packaging Considerations

As stated above, the IC of Figure 3 has a thickness equaling that of a stack of 25 sheets of paper. Thus, the carrier C of Figure 4 is constrained to such a thickness, as a minimum. A thinner carrier C may be desired in certain situations as, for example, when the package 3 takes the form of a paper envelope. A brief digression will be made to describe the actual thickness of modern integrated circuits. Figure 7 illustrates, in cross-sectional view, a CMOS transistor pair. An N+ silicon wafer, 675 microns thick (one micron equals one-millionth of an inch) supports the transistors. The pair comprises an N-well and a P-well, as indicated, associated with each well are layers of thermal oxide, phosphorus glass, P-glass, and an overlying layer of silicon nitride, SiN. In addition, N+ and P+ sources, and drains, are shown, as well as polysilicon gates. Aluminum contacts are indicated.

The actual thickness of the transistors lies in the range of a few microns. For simplicity, a thickness of 25 microns will be assumed, as indicated. The remainder of the wafer, 650 microns, is not needed for electronic reasons, and acts primarily as mechanical support for the transistors.

Thus, the actual thickness of the integrated circuits used by the invention is extremely small: thinner than a single sheet of paper. (A sheet 0.004 mils thick corresponds to about 100 microns in thickness. Thus, the 25-micron thickness indicated in Figure 7 is 1/4 the thickness of such a sheet.)

Consequently, the thickness of the electronic components of a memory circuit is insignificant. However, since those electronic components are constructed of materials, such as silicon, which resemble glass in brittleness, a 25-micron-thick circuit cannot be directly attached to an envelope. The following discussion will set forth approaches to packaging this very thin structure for attachment to an envelope.

Additional Form of Invention

In one form of the invention, a normal memory IC is manufactured, diced into its final size, ground to a very small thickness, and supported in a carrier which does not transmit bending loads to the circuit. To explain this in greater detail, the memory IC is indicated in Figure 8. After manufacture, either the individual IC, or the six-inch diameter wafer on which it was manufactured, it is attached to a vacuum chuck VC, and ground to a suitable thickness, yet without damaging the transistors, which are indicated in Figure 7. The final thickness is preferably in the range of 1 - 30 mils. (One mil is one milli-inch, or 1/1000 inch. One mil equals 25.4 microns.) If the wafer is used, it is then diced into individual dies, and the memory IC, as one of the dice, is selected.

The now-thinner IC is bonded to a backing sheet of stainless steel, aluminum, or other rigid substrate having a similar coefficient of thermal expansion as the IC, and labeled BACK in Figure 8B. At this time, the IC is still attached to the vacuum chuck VC, which imparts rigidity to the IC. The rigid backing sheet is preferably a few mils thick. Stated in more prosaic terms: the backing is about the thickness of one, or two, playing cards. The backing sheet adds stiffness to the IC.

In Figure 8C, the IC, with its stiffening sheet BACK, is fastened to a mound of flexible silicon adhesive AD, which is itself supported by a somewhat flexible substrate SUB. The overall assembly in Figure 8C has a thickness T in the range of about 5 - 30 mils.

Wires W lead to contact pads P, which are supported by ring RR. Under this arrangement, even significant bending of the substrate SUB will not be communicated to the IC.

In another form of the invention, the IC, either as fabricated or after the thinning operation, is supported on a type of trampoline, as shown in Figure 9. A flexible web, or membrane, M is supported by the walls comprising the carrier. The IC is supported by the web M. Again, very little stress or strain is applied to the IC if the carrier C should bend. Preferably, the IC is connected to the web by a dot of adhesive (not shown), applied at a single point P, as indicated.

Additional Embodiment

Figure 10 illustrates a package 3 containing a single-board computer 100. The computer 100 is shown positioned inside the package, but can be attached externally, in the manner of carrier C in Figure 4. The computer 100 contains non- volatile memory, such as EEPROM (Electrically Eraseable Programmable Memory) which retains data, even when electrical power terminates.

One such computer is based on the so-called "personal computer" architecture, which was developed by IBM Corporation, Armonk, New York, and utilizes the 8XX86 family of microprocessors, designed by INTEL Corporation, Santa Clara, California. A single-board implementation of this architecture is available, for example, from Octagon Systems, 6510 West 91st Avenue, Westminister, Colorado.

Alternately, the computer 100 can take the form of a microcontroller, such as one of the PIC series of controllers, available from Microchip Technologies, together with nonvolatile memory.

The computer 100 is equipped with an interface 105, such as the infra-red transceiver XCVR 1 shown in Figure 6, which communicates with a compatible transceiver 110 contained in station S.

Figures 1 1 - 13 illustrate three approaches to delivering power to the computer 100. In Figure 11, a time-changing magnetic flux B is induced in an iron core Fe by a wire coil Cl energized by an alternating-current source S 1. The field B induces a time-changing current within a second coil C2. In effect, coils Cl and C2 represent a transformer which has been cut in half. The induced time-changing current is fed to a rectifier R, whose output is fed to a ripple filter RF, which produces nearly constant DC voltage at the terminals labeled POWER

OUT.

In Figure 12, a light source L transmits light to a photocell P which produces power at the terminals labeled POWER OUT.

In Figure 12A, a flat coil of wire 200 surrounds a time-changing magnetic field B. In Figure 12B, a flat radio-frequency antenna ANT is shown. Microwave energy is applied to the antenna ANT. In both Figures 12A and 12B, the power extracted from the device is applied to a rectifier and filter (not shown), as in Figure 11.

In Figure 12C, a piezoelectric film PIEZ is stretched over a hoop (not shown), forming a diaphragm. Power is extracted from two contacts: contact Cl on the front, and contact C2 on the back. Such piezoelectric film is available under the name KYNAR from AMP Corporation, Valley Forge, Pennsylvania. An acoustic signal, such as a 60 Hz tone produced by a loudspeaker (not shown) mechanically displaces the film PIEZ, thereby causing an oscillating voltage to be produced at the terminals POWER OUT. That voltage is rectified and filtered, as in Figure 11.

In Figure 12D, a capacitor comprising plates PL3 and PL4, separated by a dielectric D, produces the power. Those plates are connected to plates PL2 and PL5, respectively, as indicated by wires Wl and W2. The latter plates PL2 and PL5 are excited by a sinusoidal electric field generated by plates PL1 and PL6. The power delivered to the terminals POWER OUT is rectified and filtered in the manner of Figure 11. Figure 12E illustrates the system in cross-section. The smart card, containing all plates except plates PL1 and PL6, is indicated as CARD. In Figure 13, two wires, provided by the station 15, are connected to two terminals Tl and T2, provide the POWER OUT.

The data described above is stored in the memory of the computer.

The Data

The data of Figure 2 is loaded, used, and verified, by computers, or other processors, which are networked together, as by the INTERNET. Figure 14 provides an example.

Using the INTERNET, a first computer Cl contacts a governmental postal service, or its agent, which maintains computer C2. The latter C2, in effect, "sells" postage stamps to computer C 1 , by transferring a unique number to computer C 1. This number may, for example, be a 200 byte number. Computer Cl pays for the postage in any convenient manner.

Computer C2 loads that number into the memory M of the invention. The package 3 is deposited with the postal service. Thereafter, whenever question is raised as to whether the proper postage was paid for package 3, the memory M is read, and the number is verified with computer C2, who issued the number, using the network. For example, computer C3, located at a postal processing station, may read the memory M, and obtain the verification.

In addition, the memory M is loaded with the name and address of the recipient. Computers at the various postal processing centers read that information, and route the package accordingly.

It is possible that a thief may read the number from memory M, and load that number into the thief s own package 3, to thereby issue counterfeit postage. However, this can be prevented by known cryptographic techniques, such as those discussed in Applied Cryptography, by Bruce Schneier (John Wiley & Sons, New York, 1996, ISBN 0 471 12845

7). Some techniques are described below. This text is hereby incorporated by reference.

An example of one preventive technique will be described. Since the memory M also contains the name and address of the intended recipient of the package, and also probably contains the name and address of the sender, at the time of issuance of the 200-byte number, computer Cl gave those names and addresses to computer C2, which stores it, in association with the 200-byte number.

At a later time, whenever the 200-byte number is read from the memory M on the package 3 at a processing station (not shown), these names and addresses are also read, and all three pieces of data (200-byte number, sender, and recipient) are transmitted to computer C2. If a name or address does not correspond to that stored in computer C2 at the time of issuance of the 200-byte number, the package is rejected. Thus, a thief can only send packages from the authentic sender's location to the authentic recipient's location, which reduces the value of a stolen 200-byte number to almost zero.

Further, a time-stamp indicating the time of issuance can be added to the 200-byte number. The 200-byte number expires a stated time after issuance, such as five days. The expiration further reduces the value of a stolen 200-byte number.

Further still, at the time of issuance, the computer C2 is informed of the postal codes of both the sender and recipient, called ZIP codes in the U.S. The computer C2 then computes an expected route for the package, which includes the major postal handling stations which will handle the package, and which will read the 200-byte number.

The computer C2 then encrypts the identities of those stations, perhaps using public- key encryption, and adds the encrypted identities to the 200-byte number. When the number and the encrypted identities are stored in memory M, the package 3 thus "knows" which major stations it will visit. Alternately, computer C2 only stores those identities within itself.

When a major station reads memory M, it inquires whether the package 3 should, in fact, be located at the station, either by de-crypting the identities, or by asking computer C2.

If not, the transport of the package is suspended, and an investigation is undertaken. If so, the package is forwarded without hindrance.

Bending of Envelope and Countermeasures Bending

As stated above, in an integrated circuit, the thickness of the actual electronics residing upon a silicon substrate is very thin. Sheet 100 in Figure 15 represents a silicon substrate, which has been ground to a thickness of 1 mil. Assume that the sheet 100 is mounted in a membrane M, as in Figure 9, which is supported by two sheets of cardboard 105 in Figure 15. Sheets 110 represent protective sheets, such as paper or sheet plastic. The overall thickness T is in the approximate range of 5 - 20 mils.

As shown on the right side of Figure 15, the cardboard 105 is bent at a 15-degree angle, rotating about points P2. The radius of curvature R existing when the sheet 110 contacts the IC 100 can be approximated by the expression given, namely, R = A/(2 x sin 15), wherein A is the spacing between the cardboard sheets, as indicated. This is the maximum radius of curvature allowed. If this radius is exceeded, then bending forces will be applied to the IC 100, which is not allowed.

As the expression R = A/(2 x sin 15) indicates, as A increases, the radius R increases. Thus, the larger the spacing A becomes, the larger the allowed radius becomes, meaning that the less bending is allowed. Figure 16 illustrates an approximate calculation of the maximum angle of bending B allowed, the maximum angle B being that at which the bottom sheet 105 contacts the IC 100.

Parameter T is the thickness of the cardboard. As indicated on the right of the Figure, T/2 is the altitude of a triangle, and A/2 is the hypotenuse. Thus, as indicated, the angle B is determined by the expression sin B = T/A.

Thus, as the thickness T increases, sin B increases, so angle B increases. With a larger T, a larger angle B can be tolerated, for a given A. Conversely, as the spacing A increases, sin B decreases, so angle B decreases. With a larger A, a smaller angle B can be tolerated, for a given T.

Figure 17 illustrates this graphically, and is drawn approximately to scale. Clearly, with a large spacing A, such as 100 mils, and a small thickness T, a small amount of bending will cause the sheet 110 to contact the IC 100. But with a smaller spacing A, and a larger thickness T, such as 12 mils, the allowed angle of bending allowed becomes larger.

Thus, one preferred ratio of T to A is 12/50, as shown in Figure 17. The dimension T corresponds to the depth of the well W shown in Figure 4. The dimension A corresponds to the width of the well W. Other numerical values of T and A can be used, such as 6 and 25, but the ratio 12/50 is preferred, as well as deviations from that ratio by up to 200 percent. Of course, a higher ratio gives greater protection against bending.

Figures 18 A and 18B illustrate why bending is a problem in a flat structure. Figure 18 A shows a thin sheet of silicon SIL held in a vise V. Figure 18B illustrates a model explaining how a small bending force FB can break the sheet SIL. That force FB creates a moment about point P. That moment must be opposed by a tensile force FO, shown in the enlargement. The moment arm of FO is one mil, while that of FB is 500 mils, or one-half inch.

Since the moments must be equal, the product 0.00 l(FO) must equal the product .500(FB).

Force FO must be 500 times greater than force FB.

Of course, this conclusion is based on a simplified model. Force FO is not the only force opposing force FB. Numerous forces within region R in Figure 18B form the opposition. However, in a glass-like structure, numerous grain boundaries exist. That is, just as grains of sand in a glass only contact each other at specific points, the grains in a glass-like structure only contact each other at similar points. Thus, the forces FO are applied at these points, and not continuously within region R.

Therefore, the bending causes very high tensile forces to exist in the silicon sheet SIL. These tensile forces must be minimized. Further, the tensile forces vastly exceed the shear forces within the sheet. That is, two sets of systems must be in balance: the moments and the forces themselves. The bending force FB must be balanced by an opposing force, and those two forces together place the sheet SIL into shear. However, the opposing force need only equal the bending force FB; it is not 500 times greater, as is force FO.

For these reasons, a cube-like structure is preferred. Figure 19 shows a cube C. Applying the analysis of Figure 18, Figure 20 shows that if a bending force FB is applied, the opposing force FO need only equal the bending force FB, because the moment arms Dl and D2 are equal. If the shear is resisted in a single plane, such as plane PL in Figure 21, then the resisting shear force FR need only equal the bending force FB.

Thus, in the cube-like structure shown in Figure 20, the opposing tensile force FO equals the bending force FB. If the moment point P is moved to the left, the opposing force FO will increase, but can only attain a value of 2 x FO, even assuming that such a situation can exist. Therefore, in a cube-like structure, the maximum tensile force possible, assuming that only a single layer of atoms on the surface (represented by force FO) opposes the bending force FB, is twice the bending force. Restated, the maximum possible tensile force is limited to twice the resisting shear force.

Figure 22 illustrates a cube C of silicon. Surface SI represents the surface of a silicon wafer, from which the cube C is cut. Surface SI carries the electronic components of the invention. Contacts CON are located at the corners of surface S 1.

The cube is about 1/16 inch on a side. Alternately, the cube can be 675 microns on a side, which is about 1/40 inch. In the latter case, the cube can be cut directly from standard silicon wafers, which are supplied in that thickness.

A flexible insulating strip 230, such as MYLAR or KAPTON supports traces TR. When the strip 230 in Figure 23 is placed atop the cube C, traces TR (not shown) align with the contacts, and connect to contact points CP. These contact points CP allow external probes (not shown) to make contact with the cube C. The sheet 230 is bonded to a rigid backing 235.

Multiple cubes C can be placed between the sheet 230 and the backing 235, as shown in Figure 24. Spacers can be also added, to cause elements 230 and 235 to run parallel to each other.

Countermeasures

Figures 25 - 27 illustrate one form of the invention. One, or two, silicon dies 100, or integrated circuits, are supported by a flexible membrane M. The dies 100 contain the electronics used by the invention. The sandwiching of the membrane M between two dies 100, which are attached by adhesive (not shown) to the membrane M, provides a stiff laminated structure.

The membrane M is supported by a frame 305. Structural columns COL are provided to prevent films 300 from contacting the dies 100 when bending occurs. Figures 15 and 16 illustrated such contact.

Figure 25 is drawn approximately to scale. Dimension Dl represents a distance of 1 to 10 mils. Thus, since the overall structure is about six units Dl in thickness, the overall structure ranges between about 6 and 60 mils in thickness. The die 100 is preferably square in plan view, that is, when viewed from the top. The width, WID, ranges from about 1/40 inch to about 1/2 inch.

The die 100 either connects with interface circuitry (not shown), such as the infra-red transceivers described above, or the interface circuitry is fabricated directly on the die 100.

Figures 28 - 29 illustrate another form of the invention. A frame 305 is provided in Figure 28. A flexible membrane M supports silicon die 100, which carries the logic required by the invention. An infra-red transceiver, of the type described above, may be provided separately as unit TR, or may be integrated onto the die 100. The die 100 is of a size described in connection with Figures 25 - 27.

A flat battery is provided, such as the POLAPULSE P80 or PI 00, manufactured by Polaroid Corporation, Waltham, Massachusetts. The battery provides electrical power for the die 100. This battery measure approximately 3 x 3 x 0.19 inches. However, batteries in general comprise electrodes separated by electrolyte. For example, two flat plates of the appropriate metals, separated by a cloth saturated in appropriate electrolyte, provide a simple battery. Accordingly, battery BAT is represented as a laminate of multiple films, which can be of the type found in the POLAPULSE battery just identified. The sequence is: conductor, electrolyte, conductor, insulator, and this sequence repeats a sufficient number of times to provide (1) the needed voltage and (2) the required thickness. Appropriate connections are made between the conductors, as known in the art. The battery may be either "primary," that is, non-rechargeable, or "secondary," that is, rechargeable.

Power is delivered to the die 100 by the battery 100 through wires W shown in Figures 29 and 30. An antenna ANT is shown in Figure 28, and a wire W connects to it in Figure 29. The antenna is used if an rf transceiver is utilized instead of the infra-red transceiver. Such transceivers are discussed below.

Window WIN in Figure 29 may be provided in film 350. However, since infra-red light will readily pass through several sheets of paper, film 350 may be constructed of paper, and the window WIN eliminated.

Dimension T in Figure 30 represents thickness, and ranges preferably from about 10 mils to about 60 mils. Width W in Figure 29 ranges preferably from about 1/2 inch to about 4 inches. Length L ranges preferably from about 1/2 inch to about 5 inches.

Figure 31 illustrates one form of the invention. Block 410 represents the memory, processor, and transceiver, or the smart card 6 of Figure 1. It is supported by a card 405. Both block 410 and the card 405 are sealed into envelope 400. Block 410 may be self- powered, or externally powered. As stated above, infra-red light will penetrate ordinary paper, so communication is possible using the infra-red transceiver identified above, even though the transceiver is completely surrounded by envelope material.

Figure 32 illustrates another form of the invention. Membrane M supports four dies 100, which are interconnected by traces TR. This arrangement is somewhat analogous in structure to a set of mosaic tiles, fastened to a flexible backing. The backing-tile system can bend. However, a single tile of the same size cannot. The structure in Figure 32 can bend. Electrical connections between the traces TR and the contacts (not shown) on the dies

100 can be made using conductive adhesives. Such adhesives, the membranes M, and the processing available to construct the apparatus of Figure 32 are available from Republic Technologies, Charlottesville, Virginia.

In Figure 33, columns COL are added. The apparatus of Figure 33 is supported by a frame 305 in Figure 34. Figure 35 provides a cross-sectional view. Figure 36 shows protective films F added. The columns COL prevent the films F from touching the dies 100 if bending occurs. The dimensions of the structure are approximately the same as given for those of Figures 25 - 31.

Figure 37 illustrates a variation, wherein the membrane M contains traces (not shown) on both sides, and dies 100 on both sides as well. Representative dimensions are given.

Figure 38 illustrates an envelope ENV carrying the invention 500. The invention 500 bears contact fingers CON for communication. A gripper GR is shown, such as that used in post offices to grasp envelopes. The gripper GR contains mating contacts CON2. When the gripper GR grasps the envelope ENV, as in Figure 39, the contacts CON and CON2 become mated. Communication occurs during the time the gripper GR moves the envelope from one location to another.

For example, human operators commonly operate a workstation, at which a gripper presents one envelope after another to the operator. The operator reads part of the address on the envelope, such as a postal zone (called ZIP codes in the U.S.), and enters the zone into a computer terminal. The gripper then deposits the envelope into an appropriate chute.

During the time the gripper contacts the envelope, certain of the steps described herein can be undertaken. One Form of Invention

Figure 40 is a functional block diagram of one form of the invention. A card-like structure 600 supports the components indicated. The card is preferably the thickness of a common playing card, about 5 - 10 mils, and about matchbook size, about 1 x 1 to 2 x 2 inches.

The components include a power source 610, which can take the form of a flat battery, or a receiver of externally generated power, as illustrated in Figures 11 and 12A - 12E. Contact terminals TE are provided in Figure 40, for use if external power is to be directly supplied. In such a case, diode Dl may be added, to isolate the battery (if used) from the terminals TE. Alternately, if the power source 610 is rechargeable, diode Dl may be eliminated.

Circuit 620 contains the logic which implements the operations utilized by the invention. A processing block 630 and an interfacing block 640 are shown. These may be separate, or fabricated onto a single die. Interface block 640 is preferably an infra-red transceiver as discussed above, or a radio-frequency transceiver.

Processing block 630 contains memory, a decoder/controller for accessing that memory, a simple microprocessor, and one, or more, programs which run on the microprocessor. A permanent serial number 631 may be provided. This will, in general, be a large number, because large numbers of the invention are expected to be used. The serial number is readable by the processor, and can be transmitted using the interface 640. Alternately, the serial number 631 is not provided, but users generate their own serial numbers, as by using one of the 200-byte numbers described below.

In one form of the invention, a limited set of functions is implemented, and no others. Block 650 indicates one such set of functions. As indicated in item A, the invention receives and stores a small quantity of 200-byte numbers. A quantity often is indicated, and 200-byte numbers are chosen because they are "large" numbers. Other quantities of numbers, and other sizes of each number can be used. The 200-byte number is illustrative of the "keys" used in the art of cryptography. Smaller, or larger, keys may be used. These numbers can also represent data.

As indicated in item B, the invention will transmit a selected number, using the transceiver 640. For example, if a message arrives stating, "Transmit number 5," the invention will transmit the 200-byte number labeled 5. The serial number can be selected for transmission.

As indicated in item C, the invention can ascertain whether it is being addressed by an external party. Reasons for this are explained later. In brief, if one instance of the invention is present with several other instances at the same location, an external agent may wish to communicate with each individually. However, if the agent addressed all instances at once, then all instances would attempt to respond at once.

To prevent this, the agent can "sort" the instances according to their serial numbers. For example, assume serial numbers ranging from 1 to 10. The agent may address all instances, and state, "Instance having serial number 1 report." If such an instance is present, it will report. The agent will ask that instance to transmit its 200-byte number 4, and the instance will do so. The agent will proceed to serial number 2, and so on.

In addition to the limited functionality of block 650, another form of the invention will perform the additional functions of block 660, but no others. As indicated in item D, the invention will transmit a specific 200-byte number to any party who asks for it. This number may represent the serial number 631. As indicated by item E, the invention requires that inquiries for all other numbers be authorized by a cryptographic key. That is, only agents who present a proper key, or code, will be granted access to the 200-byte numbers. Further, when access is granted, the 200- byte number given to the agent is given in ciphertext. No plain text is given to the agent.

Additional Embodiments

Figure 41 illustrates one form of the invention. Cylinder CYL, preferably about 1/4 inch in diameter, was cut from a silicon wafer, and carries circuitry 700 which was fabricated on the wafer. The circuitry performs part, or all, of the functions described in connection with Figure 40.

An elastomeric support 705 contains three slits SI, S2, and S3, which extend from face FI to face F2. The elastomeric support also contains a triangular hole 710. The hole 710 is sized such that, when cylinder CYL in inserted into the hole 710, the walls bulge slightly, as indicated in Figure 42. The deformation of the walls applies pressure to the cylinder CYL, thereby keeping it in place.

The hole may also be cylindrical, as shown in Figure 43. Whether cylindrical or triangular, the walls of hole 710 may carry electrical contacts 720. These mate with contacts 730 carried on the surface of the cylinder CYL. The contacts 720 on the wall connect to another component, such as a power supply or infra-red transceiver, generically indicated as circle 750 in Figure 44.

Figures 45 and 46 illustrate a cross-sectional view, of the type shown in Figure 15. In Figure 44, semiconductor die 100 is supported by a single pedestal 775. The die 100 is about 1 or 2 mils thick. Bending of the frame 305 will not, in general, be transmitted to the die 100. In Figure 46, the semiconductor die 100 is thicker, and is supported by a shorter pedestal 780. The thickness is about 3 - 10 mils. The thicker die has greater bending resistance.

In Figure 47, the die 100 is viewed from the top. It is supported at three points, by three dots of elastomeric material, such as silicone adhesive. The three-point suspension provides a unique feature. Three points define a plane. Thus, if any of the points move, as may occur if the frame 305 bends, the points will still define a plane. The movement may apply compressive or tensile loading to the die 100, but not bending.

The die 100 may contain a transceiver, so that external connections are not required, except possibly to supply power. Further, those connections can be eliminated if the expedients of Figures 11 and 12A - 12E are implemented.

Figure 48 illustrates another form of the invention. The die 100, also shown in plan view as 100A, contains conductive traces 800, for making connections with the outside world. Three flexible sheets 805, 810, and 815 are shown, of the type used in making flexible printed circuit boards, are also shown. Sheet 805 contains traces 820, which mate with traces 800, when the elements are assembled into the stack 930 shown in Figure 49. The stack is inserted into a channel 935 cut into frame 305, as shown in Figure 50. A flexible pedestal 940 supports one end of the die 100.

The stack 930 can be fabricated using the approaches available from Republic Technologies, discussed above.

Figure 51 is a cross-sectional view of the assembly. Figure 52 shows an alternate configuration, wherein the stack is positioned mid-way in the height of the frame 305. Figure 53 shows the pedestal 940 absent. The die 100 is cantilevered.

In Figures 50 - 53, the die 100 is substantially isolated from bending of the frame 305. Figure 54 illustrates one form of the invention. An envelope 950 is divided into two sections 960 and 965 by a partition, such as a bead of adhesive 970. A folded letter 975 is contained in section 960, and a smart card 6 is contained in section 965. In this approach, the smart card 6 does not add its thickness to that of the letter 975.

Figure 55 illustrates another form of the invention. A smart card 6 is contained in an envelope 980, which is attached to a package 3. If the smart card 6 contains an infra-red transceiver, it communicates through the paper of envelope 980. Alternately, a window transparent to infra-red light can be provided.

Figure 56 illustrates another form of the invention. The smart card 6 contains a groove 985 on at least two sides. The groove mates with the edges of a card 990, as shown in Figure 57. Transparent adhesive tape 991 may secure the smart card 6. The assembly is placed into an envelope 995, as shown in Figure 58. Other documents are placed onto the envelope in addition. Infra-red communication is undertaken through the paper of the envelope, or a transparent window can be provided.

Figures 59 - 62 illustrate schematically a storage box 1000, which receives packages 3, and also envelopes (not shown). The packages 3 contain a smart card 6, or other embodiment of the invention described herein. A one-way door 1005 is provided, which allows packages to be deposited, but not withdrawn. The door is shown schematically. Numerous approaches to one-way doors are possible, including revolving doors; the doors commonly used on post office deposit boxes; and "airlock" type double doors, wherein an outer door is only allowed to open when an inner door is locked, and the inner door is only allowed to open when the outer door is locked.

Figure 63 illustrates the storage box 1000, with the addition of a lock 1010, and two local transceivers 1012 which are capable of communicating with the smart card 6, or other embodiment, shown in Figure 59. Two transceivers in Figure 63 are not strictly necessary, but are shown to allow the option of communication prior to deposit of a package, and also to allow communication after the deposit.

Protocols

Figures 64 - 67 are flow charts illustrating logic implemented by various forms of the invention.

Figure 64 describes a typical journey undertaken by the invention, from beginning to end. In block 1100, a virtual postage stamp is obtained. The stamp can be obtained using the apparatus shown in Figure 14. The stamp is actually data, such as a 200 byte binary number, written into the memory M of Figure 14.

The stamp preferably possesses three properties, as indicated in block 1100 in Figure 64. It cannot be copied, it is self-canceling, and it cannot be re-used. Procedures for achieving all three properties are described in the Schneier text described herein. Some simple approaches for attaining the properties are the following.

The triplet of (1) the stamp (a number), (2) the recipient's address, and (3) the date are applied to a one-way hash function, which produces an output. All items are stored within the memory M in Figure 14. The date may be encrypted, as may be the others.

In the science of cryptography, the inputs (stamp, date, and recipient's address) are called the "pre-image" and the output is called the "hash value." Numerous types of one-way hash functions are available. In general, they are complex mathematical algorithms. They possess the property that, even if the algorithm is known, it is impossible to determine the pre-image from a given hash value. Hence, the designation "one way." When the validity of the stamp is to be ascertained, a station, such as S in Figure 1, reads the four items (date, stamp, address, and hash value) and checks whether the pre-image (date, stamp, and address) hash to the hash value. If not, the stamp is rejected.

With this approach, the stamp is effectively tied to the recipient and the date, and cannot be used with others.

It may be desirable, at the issuance of the stamp, to treat the stamp as a hash value of other pre-images. For example, the sender may maintain an account with the postal service C2 in Figure 14. That account is assigned a unique number. The sender is also assigned a second number, which, in effect, is an encrypted version of the unique account number. The latter number is stored within the memory M in Figure 14, together with the stamp, which is a hash of a pre-image which includes the unique account number.

When the validity of the stamp is to be ascertained, the encrypted account number is deduced, the pre-image is generated, and the hash value is determined. If that value does not equal the stamp's value, the stamp is rejected.

In a more general sense, when the validity of the stamp is to be ascertained, the question to be asked is this: Was this stamp issued by the postal service ? This question is essentially the same asked when messages are to be authenticated, or signatures are to be authenticated. Message Authentication Codes, MACs, and digital signature authentication are known in the art, and described in the Schneier text.

In addition, the "stamp" is essentially cash, which is destroyed in the transportation process. Digital cash systems exist.

A second approach to validating the stamp is the following. The sender buys stamps in groups of 100 stamps. At the time of purchase, a random number is generated. Then a hash of that number is generated. A hash of that hash is generated, and so on, 100 times. Thus, one begins with R, produces f(R), then f(f(R)), f(f(f(R))), and so on. Label these numbers x( 1 ) through x( 100). These are given to the sender.

The postal service also computes x(101), and stores it next to the sender's name.

When the sender mails a package, x(100) is loaded into the memory M in Figure 14. When the postal service validates the stamp, it computes x(101), using x(100) read from the memory M. If that x(101) matches that next to the sender's name, the stamp is validated. The postal service replaces the x(101) next to the sender's name with the x(100) received from the sender.

When the sender sends the next letter, x(99) is used, and the process repeats.

Returning to Figure 64, once a stamp having the desired properties is obtained, it is loaded into memory M in Figure 14, as indicated in block 1105. In block 1110, a serial number for the parcel is loaded into memory M. In block 1115, the destination of the parcel is loaded into memory M. In block 1120, the parcel is delivered to the transporter, such as a postal service or a private carrier.

In block 1125, the transporter reads the destination from the memory M, and may also perform validation of the stamp. In block 1130, the transporter routes the parcel to the destination. In block 1135, the transporter delivers the parcel to a box of the type shown in Figure 63, which is present at the destination. In block 1140, a transceiver or computer at the box reads the memory M, and transmits a message to the owner of the box, based on the data read. For example, the data may identify the contents of the parcel, and the computer informs the owner, by a message transmitted over the network in Figure 14, or by a pager message, of the contents which have arrived.

Figure 65 illustrates a protocol undertaken by a transceiver at the box located at the recipient's location. The transceiver controls the lock on the box. The transceiver communicates with the agent delivering the parcel. The transceiver, in essence, asks: Are you a person who has authority to deposit articles into the box ? Restated, the agent is to be authenticated. Standard procedures for authentication are known, and are described in the

Schneier text. In general, the agent presents a code, which the transceiver validates, either locally, or by contacting a remote party, such as the postal service.

For example, each delivery agent can be assigned a number, and the number can change daily. The agent presents a number, and the transceiver telephones the postal service, or contacts the postal service over the Internet, to validate the number. The validation step is represented by blocks 1200 and 1205 in Figure 65.

If the delivery agent is validated, the door of the box is opened, as indicated in block 1210. If not, an alert is sent to an interested party, such as the postal service, the police, or the owner of the box, as indicated in block 1220.

Figure 66 illustrates steps occurring within the recipient's box, when packages arrive. In block 1300, deposit of a package is detected. This detection is straightforward, since the local transceiver will know of the deposit of a package. If more than one package is deposited, additional steps, described later, can be taken.

In block 1305, the package is hailed, as by a radio-frequency signal, and a special suppressant code is transmitted. As will be explained in later steps, all previously delivered packages have stored this code in their memories. They are programmed not to respond to a hail containing that suppressant signal. However, since the arriving package does not possess that suppressant, it responds.

In block 1310, information is obtained from the package, such as the sender, contents, and so on. In block 1315, the suppressant-code is transmitted to the package, and it is told to store it. Thus, the package will not respond to future hails containing that code. In block 1320, the local transceiver processes the information obtained from the package, as by generating a log of received packages, and transmitting the log to the owner of the box. The transmission can take the form of an e-mail message.

A paper receipt may be issued to the delivery agent. However, since many package transporters are evolving toward a paperless system, the local transceiver can issue a virtual receipt. The virtual receipt should be non-repudiable. The following example illustrates how such a receipt can be generated.

Three parties are involved: (1) the local transceiver, (2) the delivery agent, and (3) a trusted neutral, who is remote, and contacted by the Internet, as in Figure 14. The local transceiver reads the serial number of the package, and transmits it to the trusted neutral, indicating that the local transceiver has received the package. The trusted neutral sends a confirming message to the delivery agent, as by using a paging service.

Figure 67 illustrates action undertaken by the smart card 6 on the package when it is hailed. In block 1350, the package detects the hailing. In block 1355, the logic inquires whether it has stored the suppressant-signal. If so, the hail is ignored, as in block 1360. If not, the logic responds to the hail, in block 1370.

If multiple packages are deposited at once, they will respond to the hail simultaneously, and communication will probably become impossible. In such a case, the local transceiver detects the fact that more than one package is responding at once. The transceiver then hails with a different code. In effect, the transceiver performs a sorting routine. The transceiver first asks packages having a serial number between zero and 100 to respond; then it asks packages numbered between 101 and 200 to respond, and so on. If two packages respond at this time, the transceiver performs another sort, as by requesting numbers between zero and 50, then 51 and 100, and so on. Additional Considerations

1. Infra-red transceivers were discussed above. In another approach, radio-frequency transmission can be used to carry data from the carrier C in Figure 4 to the station S.

Apparatus for performing such transmission is commercially available. For example, a cellular modem, model CM 900, as well as a product designated as The RF Module, model number RFM 900, are available from International Data Communications, 4051 Clipper Court, Fremont, California 94538. The RF Module has dimensions 1.6 x 1.0 x 0.35 inches, including its antenna, a useful range of 500 feet, and will carry 19,200 bits per second.

2. The following protocols, which are publicly known, and described in the Schneier text identified above, are specifically usable in connection with verification of, and communication with, the invention: MACs (Message Authentication Codes), public-key cryptography, hashing functions and one-way functions, digital signatures, digital signatures with time-stamps, signing documents using public-key cryptography, digital signatures which are not repudiable, SKEY functions, and authentication generally.

In the verification generally, the question arises whether a given 200-byte number (or "stamp") contained within the memory M attached to the package 3 was, in fact, issued by the postal service. Similar verification questions can arise as to other data. Any of the preceding protocols can answer the question to such a high degree of certainty that the answer is considered absolute.

In addition, the invention may keep secret some of the items of data by encryption techniques. For example, if the package contained valuable jewelry, the contents should not be identified within the memory. Further, if the contents are identified, they should be identified in encrypted form, thereby concealing their identity. Thus, the invention specifically teaches the storage of a list of contents, in encrypted form.

One definition of "encryption" is the process of disguising a message in such a way as to hide its substance. Accordingly, translation of English text into ASCII numbers is not considered "encryption," because no actual disguising occurs.

3. Figure 2 indicates types of data stored within the smart card 6. The data can be classified into two groups: protected and freely available. The freely available data will be given to anyone who asks for it. However, the protected data requires a key. In Figure 40, a key (not shown) is stored within circuit 620 for each 200-byte number. If a user wishes to obtain a 200-byte number, the user must present the correct key.

4. The system, such as that in Figure 40, is constructed so that all access to the memory, programs, and logic is granted through the transceiver, or interface 640. That is, no bus lines are exposed which allow direct reading of the memory or programs. All interaction between the circuit 620 goes through the transceiver, and is filtered, or controlled, by the program controlling I/O operations. That program requires keys for access to certain data.

5. The invention may contain a display D, shown in Figure 1, which allows direct reading of data stored within the smart card 6. The display is actuated by shorting together two electrical terminals Tl and T2. Logic within the smart card 6 detects the shorting, and scrolls data through the display which is allowed to be seen by the general public. Shorting is preferred because mechanical push-button switches can be actuated accidentally by normal processing of the package 3. Further, shorting can be accomplished by readily available objects, such as paper clips.

If secure data is required to be retrieved from the card 6, a transceiver is used, and an access code is given to the card 6.

The display may present graphical images, such as bar codes, for reading by bar code readers.

6. Figure 2 indicates that weight and size data are stored in the card 6. This data can be read by local transceivers 1012 in Figure 63, at the time of deposit of the package. The data is communicated to a central office, using the network of Figure 14. This data is used for preliminary processing of the package.

That is, the transport agency needs to know the sizes and weights of the packages, to arrange for vehicles to carry them. However, those arrangements do not require that the packages be present at the locations where the arrangements are being made. Thus, the packages may rest in the storage locker 1000 in Figure 63, while the arrangements are being made, and will then be later retrieved.

7. The invention is specifically intended to be re-usable. For example, the apparatus of Figure 57 can be used in one envelope, and then re-programmed and used in another envelope.

8. When a depositary locker 1000 as in Figure 63 receives a package from a sender, the local transceiver ascertains the address of the recipient, and transmits a message to the recipient, in the manner of Figure 14. Electronic mail messages can be used, a paging service, or other approaches.

9. Issuance of a receipt upon delivery to a recipient was discussed. A similar receipt can be issued to the sender at the time of deposit. This could allow the sender to request return of a package if desired, since, by stipulation, the receipt can be verified as to

authenticity.

10. Under the invention, postage need not be paid in advance. That is, the same system which examines the packages to determine whether adequate postage is carried by each may take another approach. Since that system is performing a computation of needed postage, that system can also bill the sender for the postage at that time, rather than require the sender to affix the postage in advance. Further, if the sender does not pay the assessed bill immediately, the postal service can delay delivery of the package until the bill is paid.

Numerous substitutions and modifications can be undertaken without departing from the true spirit and scope of the invention. What is desired to be secured by Letters Patent is the invention as defined in the following claims.

Claims

WHAT IS CLAIMED IS:

1. Apparatus for receiving packages, comprising: a) a transceiver for communicating with i) incoming packages and ii) a delivery agent; b) a locker containing a door and a lock on the door; c) means for: i) authenticating the delivery agent, and opening the door when authentication occurs; ii) receiving data from the package; and iii) transmitting information about the package to a remote location.

2. Apparatus according to claim 1, wherein the means authenticates the delivery agent by iv) receiving a code from the delivery agent and v) validating the code with a remote party.

3. Apparatus according to claim 1, wherein the information transmitted in paragraph (c)(iii) includes information about the contents of the package.

4. Apparatus according to claim 1 , wherein the transceiver communicates using infra-red radiation as a carrier.

5. Apparatus according to claim 4, wherein the infra-red radiation is transmitted to a second transceiver associated with the package, and the radiation must travel through at least one sheet of paper to reach the second transceiver.