Method and Apparatus for Secure Identification and Verification

This invention relates to methods and apparatus for secure identification and verification and more particularly to such a system which permits identification or verification by the mere proximity to a checkpoint.

In U.S. Pat. No. 4,720,860, issued Jan. 19, 1988, and in U.S. Pat. No. 4,885,778, issued Dec. 5, 1989, systems are disclosed which permit the secure verification of an individual by providing the individual with a card displaying a unique nonpredictable or pseudorandom code for the individual which code changes at regular time intervals, for example each minute. With this system, the individual either (a) keys in the number appearing on his card and the system identifies the individual by recognizing this number as being one which is present for an identifyee in the system at the given instant; or (b) the system provides verification by having the individual key in his personal identification number (PIN) or a public ID number which the system then uses to retrieve the nonpredictable code which should be present for the individual in the system. In an improved version of the system, described in copending application Ser. No. 07/341,932, filed Apr. 21, 1989, a personal password or personal identification number (PIN) is entered into the card by the user keying a keypad on the card and this PIN is utilized in generating the nonpredictable code. This further enhances security in that it makes it more difficult for someone other than the person to whom the card was issued to use the card, (requires two factors rather than only a single factor for security; something known and something possessed rather than only something possessed) and it also results in the PIN being transmitted to the verification station in a secure coded fashion so that the PIN cannot be surreptitiously observed, electronically eavesdropped or learned.

While the systems described in these patents and applications provide a high level of security, they require that the user key in the number appearing on the card, which may be a five to ten digit number, in order to obtain identification or verification. While this is useful when the user is at a remote location, for example to gain access to a computer system by telephone or terminal, it can cause annoying delays when the user is for example trying to enter a physically secure facility where the user may have to enter his code three or four times to pass through various security barriers or checkpoints at the facility. It would therefore be desirable if the nonpredictable code could be available on a card, badge or other suitable device or unit carried by the user and be presented such that the code could be automatically sensed or read by the system, permitting the user to pass through various checkpoints without the need for keying in the current code appearing on the unit at each such location. It would also be desirable if the location of an individual in the facility could be tracked without requiring any active input on the part of the user.

While devices are currently available which permit a coded output to be obtained from a unit, these systems are used primarily for nonsecure applications such as identifying livestock, railroad cars, pallets or trucks. An example of such devices is the Nedap GIS RF identification system available from Nedap USA, Sunnyvale, Calif. Someone either gaining possession of such a device or capturing its electronic radiation could easily determine the code stored therein and surreptitiously generate such code. Any individual in possession of such device could also gain access to the facility even if such individual were not the individual to whom the unit was issued.

A need therefore exists for an improved identification and verification method and apparatus which provides highly secure identification and verification while permitting such verification or identification to be achieved with either no user input, or with the user inputting only a few, usually memorized, secret code characters, such as his PIN. If a PIN is entered, it is desirable that it need be entered only once to gain full access to the facility rather than being entered for each checkpoint, provided access is completed within a predetermined time period.

This invention provides a method and apparatus for performing identification and/or verification at a predetermined location or site. A selected radiation, such as for example an RF signal or an ultrasonic signal at a predetermined frequency, is transmitted at the site. The identifyee has a unit such as a card, badge or other token which detects the predetermined radiation. The unit also stores a predetermined coded value, at least a predetermined portion of which is changed at selected time intervals in accordance with an algorithm, the algorithm being such that the value of the predetermined portion of the stored coded value at any given time is nonpredictable. In response to the detection of the selected radiation, the unit presents an indication of the current stored coded value to the site, the site containing a means responsive to the produced coded value for identifying the identifyee of the unit. For one preferred embodiment, the detection means at the unit is a tuned circuit which absorbs energy from an RF source at the frequency thereof. The currently stored coded value may be used to control the absorption state of the tuned circuit as the bits of the stored value are sequentially read out, or the sequentially read out coded bits may be utilized to control a transmitter. In either event, equipment at the site may automatically detect the current stored coded value at the unit. For another preferred embodiment the detection means is an ultrasonic detector, such as, for example, a piezoelectric crystal.

In an identification mode, a processor at the site compares the received current coded value with current coded values which are stored or generated for each identifyee in the system, and identifies the identifyee when a match is detected. If no match is detected, the identifyee may be rejected for access to the facility or resource. By placing transmitting and receiving equipment at a plurality of sites at a facility, the movements of identifyees through the facility may be tracked. In order to permit the system to be used for verification, some portion of the coded material is fixed to be used for a public ID or index.

The unit may also contain a keypad on which the user may input a short code, such as the identifyee's personal identification number (PIN). The PIN is mixed with the changing nonpredictable code in the processor in accordance with a predetermined algorithm so that the correct current coded value will appear for an identifyee only if the identifyee has inputted the proper PIN into the unit. To permit the system to be used for verification, a selected number of bits in the coded value, for example the bits for six decimal digits or characters, may be fixed in a register, with the remaining bits being used to represent the generated nonpredictable code. The six characters represent a public ID code for the identifyee which may be recognized by a site processor and utilized to retrieve the appropriate current nonpredictable coded value for such identifyee. This value may then be compared against the remainder of the received bits. Again, a match signifies acceptance with a mismatch signifying rejection.

The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of preferred embodiments as illustrated in the accompanying drawings.

FIG. 1 is a schematic semiblock diagram of a system incorporating the teachings of this invention.

FIG. 1A is a schematic block diagram of a portion of the system shown in FIG. 1 for a first alternative embodiment of the invention.

FIG. 1B is a schematic block diagram of a portion of the system shown in FIG. 1 for a second alternative embodiment of the invention.

FIGS. 2-1 and 2-2 when combined form a flow diagram illustrating the operation of a system operating in accordance with preferred embodiments of the invention.

FIG. 1 shows an identification and/or verification system in accordance with the teachings of this invention. The system has a site 10 at which identification or verification is being performed and a unit 12 which is carried by the individual to be identified or verified. The unit 12 would typically be a card, badge or other token but could take other forms. For example, it could be embedded as part of a watch, pin, pen or other item or device worn or carried by the identifyee. At the site 10, an RF transmission line 14, which includes an RF energy source generate an RF field 16. Typically RF field 16 would be continuously generated. However, if desired transmission line 14 may be energized to generate field 16 in response to the detection of an individual or other identifyee approaching the site 10. The individual could be detected using standard technique such as a photodetector, pressure sensitive pad, or the like.

When the unit 12 gets close to the field 16, the field is picked up by antenna 18. The output from antenna 18 is applied as one input to gate 20 and is an input to a tank circuit 22 formed, for example, by coil 24 and capacitor 26. The values for coil 24 and capacitor 26 are selected such that the resonant frequency of tank circuit 22 is equal to the frequency of RF transmission line 14. Thus, when antenna 18 picks up RF field 16, the tank circuit 22 becomes a load on the transmission line causing the standing wave ratio of the transmission line to be reduced. The absorption of energy by tank circuit 22 also results in a charging of capacitor 26 which is detected by detector 28. Detector 28 may for example be a standard diode detector. The output from detector 28 is applied to processor 30 indicating that the unit 12 is being scanned.

When the person or object to be identified receives the unit 12, a predetermined coded value unique to the identifyee is stored in register 32. The value stored in register 32 may for example contain sixty-four binary bits which may represent sixteen decimal characters. For the preferred embodiment, the first six of these decimal characters stored in portion 32A of the register represent a public ID code for the user and are permanently burned into or otherwise stored in a register 32. The remaining bits, representing for example ten decimal characters, are stored in portion 32B of the register. These bits are, as will now be discussed, changed at periodic intervals in accordance with a predetermined algorithm so that the code in portion 32B of the register at any given time is pseudorandom and nonpredictable. The manner in which such nonpredictable codes are generated is discussed in greater detail in the beforementioned U.S. Pat. Nos. 4,720,860 and 4,885,778.

Processor 30 has as inputs, in addition to the input from detector 28, an input from a real time clock 34 and inputs from a keypad 36. Clock 34 may for example indicate the current hour and minutes and the value in this clock may change every minute. Keypads 36 may for example be pressure sensitive pads representing the digit 0-9 which may be operated by the identifyee to for example key in his secret PIN. The contents of area 32B of register 32 are also applied as an input to processor 30. Processor 30 has a predetermined algorithm stored therein which is secret and which, in response to the inputs from register 32B, clock 34 and, when present, keypads 36, generates at periodic intervals, such as each time there is a change in the minute value in clock 34, a new nonpredictable code which is stored in register 32B.

Gate 20 is normally open so that the RF field 16 received by antenna 18 is applied to tank circuit 22. However, when detector 28 applies a signal to processor 30, processor 30 generates a sequence of shift pulses on line 38 which cause successive bit positions in register 32 to be connected to gate input line 40. When a bit is present on line 40, gate 20 is enabled, shorting any RF signal received at antenna 18 to ground. This prevents tank circuit 22 from presenting a load to RF transmission line 14. However, when no bit (i.e., a zero bit) appears on line 40, gate 20 is disabled, permitting tank circuit 22 to receive the RF field signal and thus to load RF transmission line 14. Power for controlling processor 30, clock 34, and the other components of unit 12 may be provided by a suitable conventional battery (not shown).

The change in standing wave ratio in RF transmission line 14 caused by tank circuit 22 is detected by bit detector 44. Since gate 20 is normally open, there will be drop in the RF standing wave ratio as the unit 12 approaches the site. When the unit 12 is close enough to site 10 for tank circuit 22 to provide a predetermined load to the transmission line, detector 44 becomes operative. This would typically be at a distance of a few feet, although the exact distance will vary with the equipment used. To synchronize the site and the unit, detector 44 may, for example, have a greater range than detector 28 so that it is operative when detector 28 generates an output. The first bit in register 32 may always be a one so that a transition occurs in the absorption state of the unit when a shifting operation begins. Detector 44 may detect and sync on this transition, either alone or in combination with processor 48. Other standard synchronization technologies may also be employed including either one or both of the site and unit transmitting a sync cord.

Once bit detector 44 determines that a unit 12 is shifting out a code, it starts loading bits into code storage register 46 at the same rate that shift pulses are being applied by processor 30 to line 38 Since the state of the standing wave ratio applied to detector 44 varies as a function of whether gate 20 is enabled or disabled during each such shift interval, and the state of gate 20 in turn varies as a function of whether a bit is present or not in the currently scanned position in register 32, the code stored in register 46 corresponds to the code stored in register 32 of the unit 12 being scanned.

The time required to scan register 32 is very short compared to the time between changes in values stored in register portion 32B, so that the contents of register 32 is easily scanned between change cycles of the values stored in register 32B. For example, it may take only 75 milliseconds to scan register 32 while the value in the register is changed only every minute. However, to avoid erroneous outputs, processor 30 will not typically change the contents of register 32 during a scan cycle. Further, the processor 48 at site 10 will always us the nonpredictable code for the time interval at which a scan began in order to determine a code match. Checksum or parity information may also be transmitted to insure system integrity and reliability, and in particular to protect against transmission errors.

When a code from unit 12 has been stored in storage 46, this code is applied to processor 48 which also receives stored ID codes from a store 50. The coded values inputted to processor 48 are compared in a manner to be discussed in greater detail later to either identify the object or individual with the unit or to verify the identity or validity of such individual or object. If identity is successfully established, processor 48 may generate an output to an access release mechanism 52 to, for example, permit a door or vault to be opened o to automatically open a door through which the identifyee needs to pass.

While the embodiment of the invention shown in FIG. 1 operates satisfactorily, since it depends on a detection of change in standing wave ratio in RF transmission line 14 in order to transmit bit information, it requires that antenna 18 be relatively close to RF transmission line 14, for example less than a few feet, in order to function. In some applications, a system which operates with the unit 12 at a greater range from the site 10 may be desirable. FIG. 1A therefore shows a portion of the circuit for an alternative embodiment of the invention wherein gate 20 is replaced by a transmitter 60 and RF transmission line 14 is replaced by an RF transmitter/receiver 62. The remainder of the circuit could be the same as for the circuit of FIG. 1.

For this embodiment of the invention, tank circuit 22 would still be utilized to energize detector 28. However, before processor 30 starts delivering shift pulses to line 38, it would deliver a signal or pulses to line 64 enabling transmitter 60 to start generating an output, for example an RF output at a frequency different than the frequency of RF transmitter 62, which transmission is modulated, enabled or otherwise controlled by the bits being scanned from register 32. The modulated RF signal is picked by the receiver portion of circuit 62 and applied to bit detector 44' which syncs the site and unit and determines the state of each transmitted bit from the received signal, generally in the manner previously described. For example, processor 30 may cause transmitter 60 to generate a syncing code sequence before transmission of data from register 32 through transmitter 60 begins. The range for the device shown in FIG. 1A would be limited only by the power available for transmitter 60 and could typically be several yards.

FIG. 1B shows another alternative embodiment of the invention wherein ultrasonic signals rather than RF signals are utilized to transmit information between the site 10 and unit 12. Thus, for this embodiment of the invention, RF transmitter/receiver 62 is replaced with an ultrasonic transponder 70 which may, for example, be a piezoelectric crystal generating an output at a predetermined ultrasonic frequency. Antenna 18, tank circuit 20, and transmitter 60 are similarly replaced by an ultrasonic transponder 72. The output from the ultrasonic transponder 72 is applied to an ultrasonic detector 74 which, in response to the detection of a signal at the frequency from transponder 70 generates an output on line 76 to processor 30 indicating that the unit 12 is being scanned. Processor 30 then enables transponder 72 to start transmitting any control and/or sync bits and the shifted outputs from register 32. While in FIG. 1B, the transponders 70 and 72 have been indicated as being ultrasonic transponders, the invention may also be practiced utilizing transponders (or transmitters and receivers as appropriate) for frequencies in other portions of the electromagnetic spectrum, for example, gamma rays, X-rays, or the like. While it may require the user to hold the unit up to be scanned, rather than merely wearing or carrying the unit, it may also be possible to practice the invention using visible light, infrared or the like. Further, while for the preferred embodiments of the invention described above the transponders 70 and 72 are both operated in the same frequency band of the electromagnetic spectrum, this is also not a limitation on the invention, and there may be applications where it is desirable for the transponders to operate in different frequency bands with, for example, the site transmitting RF energy and receiving ultrasonic energy with the unit receiving the RF energy and responding thereto while transmitting in the ultrasonic frequency band.

FIG. 2 is a flow diagram illustrating the operation of a system of the type shown in FIGS. 1 and 1A. Referring to the right hand side of this figure, it is seen that initially a coded value is stored in register 32 during step 70. This is done when the unit is issued to the identifyee or before. No later than the time the unit 12 is issued, the unit is also activated so that clock 34 starts operating and the time the unit starts operating is recorded at site 10. As indicated previously, all of the coded value in register 32 may be subject to change by processor 30. However, for the preferred embodiment, a portion 32A of this register contains a fixed public identification code which is not changed by processor 30 and is utilized in the manner to be discussed shortly.

The stored coded value in register 32 is applied to processor 30 and, during step 72, is utilized in conjunction with the clock signal from clock 74 to generate a nonpredictable code which is stored at least in portion 32B of register 32. The manner in which this nonpredictable code is generated is discussed in greater detail in the beforementioned United States patents. During step 74, the new code generated by processor 30 at each time interval is stored in register 32, or at least portion 32B thereof, replacing the previous values stored therein. As is indicated by dotted line 75, this coded value is used during the identification or verification operations to be discussed later. During step 76, the system checks to determine if the time interval between updates to the nonpredictable code has occurred and, when the time interval has passed, steps 72, 74 and 76 are repeated with a new clock value.

This sequence of operations continues until the identifyee in possession of the unit 12 approaches the site 10. At that point, the identifyee keys in his PIN, which may for example be a predetermined three to six digit number, on keypad 36. The keying in of the PIN is detected during step 78 and the keyed in PIN value is utilized by processor 30 during step 72, in addition to the stored coded value and the clock value, in generating the new nonpredictable code. The new nonpredictable code may be generated when the PIN is keyed in or at the next clock interval. In the later event, the identifyee should key in the PIN when he is at least one minute away from site 10 so that the value stored in register 32 when he reaches the site and is scanned includes the PIN value. This is important since, as will be discussed shortly, the site will respond to the generated nonpredictable code only if such code includes the PIN in the generation thereof.

As was previously discussed, one reason for keying in the PIN is so that someone cannot gain access to the facility with a lost or stolen unit 12, since the person having such unit would not know the identifyee's secret PIN, and the code being generated by unit 12 would therefore not be the proper code for admittance. However, in order for this objective to be achieved, the PIN must not remain in the unit on a permanent basis. However, in order to enable the identifyee to pass through a number of checkpoints without requiring the rekeying of the PIN at each such checkpoint, it is desirable that the PIN remain in the unit long enough to afford the identifyee full access to the facility. These objectives are achieved in accordance with the preferred embodiment of the invention by having the PIN remain in the unit 12 (i.e., in a predetermined storage location in processor 30) for a predetermined duration which is long enough to normally permit the identifyee full access to the facility. Thus, once a PIN has been inputted and detected, processor 30 starts counting down a predetermined duration, for example by setting a predetermined value in a register and counting that value down during selected clock intervals. During step 80, a determination is made as to whether the predetermined duration has expired. If the duration has not expired, the PIN remains in the unit for use in generating the nonpredictable codes and the system continues to check to see if the duration has expired. When the PIN retention duration expires, the system proceeds to step 82 to erase the PIN or otherwise remove the PIN from the unit 12. Thus, the security of the system is maintained, with two independent factors, something known and something possessed, being required for identification and authentication.

During step 86, RF transmission line 14 or RF transmitter 62 at the site is either continuously transmitting an RF field or starts transmitting such field in response to the detection of an identifyee approaching the site. Either in response to the same detector or in response to a detection by detector 44, the site detects the approach of a unit 12 (step 87), thus enabling the syncing of the site and unit. When antenna 18 is close enough to the site so that tank circuit 22 starts absorbing energy sufficiently to charge capacitor 26 to a level triggering detector 28, an RF signal is detected at the unit as indicated by step 88. This results in processor 30 causing the contents of register 32 to be scanned a bit at a time during step 90. Register 32 may be scanned in a number of known ways. During step 92, the next step in the operation, the bit at the currently scanned register position begins to control either gate 20 to passively present the bit to site 10 or to modulate transmitter 60 to actively transmit or present the bit to site 10. During step 93 site 10 syncs on the transition caused by the first bit, this bit being selected, as previously indicated, to cause such transition, or the site and unit are synchronized in some other standard manner.

During step 94, which is performed at the site, detector 44 or 44' detects and stores the bit received from the unit. The operation then proceeds to step 96 to determined whether all bits from register 32 have been received. Since the number of bits stored in register 32 is known, this is a simple determination which again can be accomplished by counting down a preset value or by other standard means. If all bits have not been received during step 96, the operation returns to step 94 to detect and store the next bit. When, during step 96, it is determined that all bits have been received, the operation proceeds to step 98.

For purposes of illustration, the system in FIG. 2 is shown as having two different modes of operation, namely an identification (ID) mode and a verification mode. In the ID mode, public bits are not provided in section 32A of register 32 and processor 48 functions to compare the received nonpredictable code with the current nonpredictable code for each identifyee in the system. In the verify mode, the public ID bits are present, permitting the system to select the current nonpredictable code for the identifyee and to compare this code with the received code. For purposes of illustration, both modes of operation are shown as being available in FIG. 2 although, in a typical system, only one or the other of such modes of operation would generally be present. During step 98 a determination is made as to which mode the system is operating in.

Assume initially that the system is operating in the ID mode. In this mode, the system proceeds from step 98 to step 100 during which a current code in the system is provided to processor 48. In order for the processor to function in this mode, it is necessary that it update all of the values in ID code store 50 during the same time intervals that codes ar updated by processor 30, so that the values stored in ID code store 50 are always the current coded values for each identifyee in the system. Since each update operation can be accomplished in a few milliseconds by existing processors, this requirement does not impose a serious limitation so long as the number of identifyees in the system is no excessively large.

During step 102, the coded value from ID code store 50 is compared in processor 48 with the received coded value stored in register 46. If these two coded values do not match, the operation proceeds to step 104 during which a determination is made as to whether all codes in store 50 have been used. If all codes in store 50 have not been used, the operation returns to repeat steps 100, 102 and 104 for a new coded value from store 50.

This sequence of operations continues until either, (a) during step 102, a match is obtained, in which event the system proceeds to step 106 to accept the identifyee and to for example activate access release 52; or (b) if no matches are obtained during step 102, and during step 104 it is determined that all codes in the system have been used, the operation proceeds to step 108 to reject the individual possessing unit 12. Step 108 would normally involve denying access to the facility and might also trigger an alarm to alert a guard or other individual that someone is seeking to improperly gain access to the facility.

If an identifyee is rejected, it may merely mean that he has not entered his PIN or that he has been sufficiently delayed in entering the facility so that his PIN duration has expired and the PIN has been erased. Another potential problem which might lead to rejection is the problem discussed in U.S. Pat. No. 4,885,778 where the clock 34 at the unit becomes slightly out of synchronization with the clock at the site. A method for solving this problem is discussed in the prior application, and the technique discussed in this prior application may also be utilized in conjunction with this invention to avoid spurious rejections.

Two other potential problems exist, particularly when operating in the ID mode. The first is that, since the codes are nonpredictable, it is possible that at a given time interval the same nonpredictable code may exist for two or more individuals in the system. Thus, even though a match is obtained during step 102, it may still be desirable to continue comparing the code in storage 46 with the codes in store 50 to be sure that a second match does not occur to assure that the individual has been correctly identified. If a second match does occur, rather than rejecting or falsely identifying the individual the identification may be delayed for the standard time interval, for example, one minute, until the code in the unit changes so that a second check can be made. The likelihood of two individuals having the same nonpredictable code for two successive time intervals is so infinitesimally small as to be almost nonexistent, and this would permit a unique identification of the individual.

A second possibility is that, with a reasonably large number of individuals in the system, an identifyee may be correctly identified even though he inputted the wrong PIN and should be rejected. While there is a possibility of this occurring, with sixty four bits, the number of potential code combinations is 2.sup.64, so that even with one thousand individuals in the system, the likelihood of a false hit (i.e., a false positive or so called "type 2 error") is very low. The verification mode to be now described virtually eliminates the possibility of a false hit occurring.

If during step 98 it is determined that the system is in a verification mode, the operation proceeds from step 98 to step 112 during which the public ID bits portion of the received code, the portion of the code stored in register portion 32A, is looked at by processor 48. This code may be used as an address to access memory 50 to obtain a code for that identifyee. If memory 50 is continuously updated so that the values stored in the memory are the current nonpredictable codes for the identifyees, then this is the value retrieved from memory 50. However, if these values are not continuously updated, the originally inputted code or last updated code for the individual could be retrieved and processed using the last known code and the known time since the last update to obtain the current nonpredictable code for the identifyee. These operations are performed during step 114. During step 116, the next step in the operation, the current nonpredictable code for the identifyee obtained from processor 48 is compared with the current nonpredictable code in store 46 from unit 12. If these codes do not match, the individual is rejected during step 118, while if these codes match, the identifyee is accepted during step 106. The operations and options during either acceptance or rejection would be the same as those previously described when in the ID mode.

In the discussion so far it has been assumed that all of the bits in register 32 are code bits used for identifying the individual. However, in a system, one of more of these bits may be used as checksum or parity bits to be sure that the code generated is a valid code and to protect against transmission errors. Thus, after step 96 in the operation, a checksum or parity check might be performed. If the received code were determined to be invalid, the code could be reread during the same or a subsequent time interval, or other appropriate action taken. Further, while RF transmission and reception has been utilized for the preferred embodiment of the invention, and ultrasonic transmission and reception are shown for an alternative embodiment, it is apparent that the objects of this invention, particularly for the embodiments of the invention shown in FIGS. 1A and 1B, could be achieved using other forms of radiation from the transmitter at the site and other types of radiation detection. Thus, the transmitter could be generating a low level ultrasonic signal, a low level microwave signal, or any other type of radiation which could be easily detected at unit 12. Similarly, transmitter 60 could be generating information as bursts of energy at something other than the RF frequency o could be modulating something other than an RF signal.

Further, while for the preferred embodiment the system is being used to control passage into a secure facility, into a vault, or the like, with a number of transmitters and receivers positioned at strategic locations throughout a facility, the system could also be utilized to passively monitor the location of identifyees throughout the facility. For example, a watchman having a unit 12 could walk through the facility and have the system record centrally his passing each desired checkpoint, rather than utilizing the current more cumbersome clock system. In any system where there are multiple checkpoints at a site, a single processor 48 and stores 46 and 50 would normally be shared by all checkpoints. Since the system could also identify individuals entering and leaving a facility, it could eliminate the need for attendance time clocks. It could also be used in a variety of other applications where secure identification or verification of individuals or objects is required.

Thus, while the invention has been particularly shown and described above with reference to preferred embodiments, the foregoing and other changes in form and detail may be made therein by one skilled in the art without departing from the spirit and scope of the invention.