Encryption method, cryptographic communication method, ciphertext generating device and cryptographic communication system of public-key cryptosystem6990200Abstract Two public keys including a random number term therein are prepared in a database for each divided plaintext in advance, a plaintext is divided into a plurality of 1-bit divided plaintexts, one public key is selected for each divided plaintext from the database according to the bit data of each divided plaintext, and the selected public keys are added to generate a ciphertext. This invention bases its security on free selection of a desired combination of public keys. Claims What is claimed is: Description BACKGROUND OF THE INVENTION
By multiplying a base-product b1b2 . . . bi by a random number term vi, a base Bi is given as shown by (1) below. Bi=vib1b2 . . . bi (1) Here, vi is set so that each Bi expressed by equation (1) has an almost equal size. However, the condition gcd(vi, bi=1)=1 must be satisfied. With the use of a random number w, the public key {ci} is found as shown by (2) below. ci≡wBi(mod P) (2) By performing the product-sum operation of messages {mi} obtained by dividing the plaintext into K pieces and the public keys {ci}, the ciphertext C is obtained as shown by (3) below. C=m1c1+m2c2+ . . . +mKcK (3) Decryption processing is carried out as follows. For the ciphertext C, an intermediate decrypted text M is found as shown by (4) below. M≡w-1C(mod P) (4) This intermediate decrypted text M is specifically given as equation (5), and it can be decrypted by a sequential decryption algorithm shown below. M=m1b1v1+m2b1b2v2+ . . . +mKb1b2 . . . bKvK (5) [Sequential Decryption Algorithm] Step 1 M1=M/b1 m1=M1v1-1(mod b2) Step i (i=2 to K-1) M1=(Mi-1-mi-1vi-1)/b1 m1≡Mivi-1(mod bi+1) Step K MK=(MK-1-mK-1vK-1)/bK mK=MK/vK Originally, such a public-key encryption scheme bases its security on the difficulty of factoring and the difficulty of solving a discrete logarithm problem, and various attacks against it have been proposed. Moreover, the present inventors proposed a new type of public-key cryptosystem encryption method which bases its security on such a point that a set of public keys can be freely selected among a very large number of combinations of public keys (Japanese Patent Application No. 11-269407/1999, hereinafter referred to as the "second conventional example"). This scheme is a modified scheme of the above-mentioned first conventional example. In this scheme, a plurality of public keys produced from the products of integers and random number terms are prepared in advance for each divided plaintext obtained by dividing a plaintext, an arbitrary public key is selected for each divided plaintext among these prepared public keys, and a ciphertext is generated by using the selected public keys. The following description will explain the encryption method and decryption method proposed in this second conventional example. The intermediate decryped text M during the first transfer by the encryption scheme of the second conventional example based on the scheme of the first conventional example is given by (6) below. M=m1′b1v1+m2′b1b2v2+ . . . +mK′b1b2 . . . bKvK (6) However, m1′ is encoded to establish (7) below module J for j given by adding jog2 J-bit redundancy to message (divided plaintext) m1, and the information indicating which public key among later-described plurality of public keys is to be selected for each divided plaintext is transmitted. mi′≡j(mod J) (7) FIG. 1 is an illustration showing a public key list indicating a plurality of public keys prepared for each divided plaintext. In FIG. 1, K represents a dividing number (class number) of plaintext. As illustrated in FIG. 1, the set {b1b2 . . . bivi(j)} provided by multiplying the base-product by a random number term is prepared as J pieces of public keys for each divided plaintext (each class). An entity as the recipient transforms these products of the base-product and random number term by a random number w and publicizes them. In other words, the products of the base-product and random number term shown in FIG. 1 are transformed as shown by (8) below, and the set {cij} thereof is publicized. b1b1 . . . bivi(j)w≡cij(mod P) (8) A set of public keys which is randomly selected by an entity as the sender is expressed as shown by (9) below. In this case, it is possible for the entity as the sender to select public keys in JK(>> 1) ways. [Eq. 1] (c1, j1, c2, j2, . . . , cK, jK) (9) According to a set of the selected public keys shown in (9) above, the entity as the sender lets mi′≡ji (mod J), and then generates the ciphertext C to the entity as the recipient as shown by (10) below. [Eq. 2] C=m1′c1, j1+m2′c2, j2+ . . . +mK′cK, jK (10) In order to decrypt the ciphertext C thus generated, the entity as the recipient predetermines the random number term vi(j) of FIG. 1 as shown by (11) below. vi(j)=wb, i+ri(j)bi+l (11) where each of wb, i, ri(j) is a random number. Further, the entity as the recipient has wb, i-1 that satisfies (12) below as a secret key. wb, i·wb, i-1≡1(mod bi+l) (12) The decryption processing by the entity as the recipient is carried out as follows. An intermediate decrypted text M0 is given as shown by (13) below. [Eq. 3] M0=m1′b1v1(j Therefore, decryption can be performed by the sequential decryption algorithm shown in (14) below. Incidentally, in (14), although bK+1 is a random number satisfying mK′<bK+i, it is not used as a base. In general, the random number term for ji in step i is expressed as shown by (15) below. ##EQU1## In the decryption method proposed in the above-described second conventional example, since public keys are arbitrarily selected, i.e., since the entity as the sender freely selects public keys and generates ciphertext, the selection pattern of the public keys is unknown to attackers, and thus making it difficult to attack. Besides, the present inventors are further researching on a more practical encryption method. BRIEF SUMMARY OF THE INVENTION An object of the present invention is to provide a public-key cryptosystem encryption method, cryptographic communication method, ciphertext generating device and cryptographic communication system which are capable of achieving high-speed processing while ensuring security by free selection of public keys, and a memory product/data signal embodied in carrier wave for recording/transmitting operation programs for these methods. According to a first aspect of the present invention, two public keys including a random number term therein are prepared for each divided plaintext in advance, a plaintext to be encrypted is divided into a plurality of 1-bit divided plaintexts, one public key is selected for each divided plaintext among the two public keys prepared, according to a bit pattern of the plurality of divided plaintexts, and a ciphertext is generated by using the plurality of divided plaintexts and selected public keys. In the encryption method proposed in the above-mentioned second conventional example, this first aspect limits the divided plaintexts to one bit and restrains the number of rows in the public key list to two rows (J=2). It is therefore possible to perform encryption and decryption processing at extremely high speeds. However, with the simple addition of such limitations, since a public key of the first row is selected when mi=0 and a public key of the second row is selected when mi=1, a 0, 1-knapsack cryptosystem with an extremely low level of security will result. Then, with the first aspect, the ciphertext is generated by determining which public key is to be selected for each divided plaintext, according to a bit pattern of a plurality of divided plaintexts. Hence, unlike the 0, 1-knapsack cryptosystem, high security is achieved. According to a second aspect of the present invention, 2s public keys including a random number term therein are prepared for each divided plaintext in advance, a plaintext to be encrypted is divided into a plurality of s-bit divided plaintexts, one public key is selected for each divided plaintext among the 2s public keys prepared, according to the bit data of each divided plaintext, and a ciphertext is generated by using the selected public keys. For example, when s=1, two public keys including a random number term therein (a public key list with two upper and lower rows) are prepared for each divided plaintext, one of the public keys is selected according to the bit data ("0", "1") of each divided plaintext, and all the selected public keys are added to generate the ciphertext. At this time, as an example, when the plaintext is "0", the public key of the upper row is selected, while when the plaintext is "1", the public key of the lower row is selected. With the second aspect, the ciphertext is generated simply by adding the public keys including a random number term therein, which are selected according to the bit data, and the encryption and decryption processing becomes extremely fast. The bit data of each divided plaintext used as a criterion to select a public key is unknown to the attackers and the selection pattern of the public keys can never be known, thereby achieving high security. With the present invention, it is possible to achieve high-speed encryption/decryption processing while ensuring security by free selection of public keys, and the present invention can largely contribute to the development and realization of practical use of public-key encryption schemes. The above and further objects and features of the invention will more fully be apparent from the following detailed description with accompanying drawings. BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS FIG. 1 is an illustration showing a public key list according to an encryption scheme of the second conventional example, FIG. 2 is a depiction showing a state of cryptographic communications of information between two entities according to a first embodiment, FIG. 3 is an illustration showing a public key list in a database according to the present invention, FIG. 4 is a depiction showing a state of cryptographic communications of information between two entities according to a second embodiment, and FIG. 5 is an illustration showing the structures of embodiments of a memory product. DETAILED DESCRIPTION OF THE INVENTION The present invention will be described below with reference to the drawings illustrating the embodiments thereof. First Embodiment The following description will explain a first embodiment in which a public key is selected according to a bit pattern of a plurality of divided plaintexts. FIG. 2 is a depiction showing a state in which an encryption scheme according to the first embodiment (first aspect) is used for information communications between entities A and B. The example shown in FIG. 2 illustrates a case where one of the entities, A, encrypts a plaintext X into a ciphertext C and transmits the ciphertext C to the other entity, B, via a communication path 1, and the entity B decrypts the ciphertext C into the original plaintext X. The entity A as the sender is provided with a plaintext divider 2 for dividing the plaintext X into a plurality of 1-bit divided plaintexts, a public-key selector 3 for selecting a public key for each divided plaintext from a database 6 storing a public key list as described later, and an encryptor 4 for generating the ciphertext C by using the selected public keys and respective divided plaintexts. Besides, the entity B as the recipient is provided with a decryptor 5 for decrypting the transmitted ciphertext C into the original plaintext X. In this example, the issuer of the public key list is the entity B as the recipient, and the user of this public key list is the entity A as the sender. Next, a specific technique will be explained. FIG. 3 is an illustration showing the public key list in the database 6 that stores a plurality of public keys for each divided plaintext in advance. FIG. 3 shows a public key list in accordance with the supposition that a public key for each divided plaintext is constructed by modular transformation by (w1, P1). In FIG. 3, K represents a dividing number (class number) of the plaintext X, two (upper row, lower row) public keys including a random number term therein are prepared for each of K pieces of divided plaintexts (for each class). In the encryption method proposed in the second conventional example, when mi=0, the component vi(0) of the upper row of the public key list of FIG. 3 is selected, while when mi=1, the component vi(1) of the lower row is selected. Thus, when the technique is simply applied to the encryption method of the second conventional example, a 0, 1-knapsack cryptosystem with an extremely low level of security will result. Then, in the first embodiment, it is determined which row of the public keys in the public key list is to be selected for each divided plaintext, according to a bit pattern of a plurality of divided plaintexts. In other words, after dividing the plaintext X into K pieces of 1-bit divided plaintexts, selection information (x1, x2, . . . , xK) indicating which row of the public keys is to be selected is determined for each divided plaintext, according to a bit pattern of the K pieces of divided plaintexts (m1, m2, . . . , mK). An algorithm for pre-coding the divided plaintexts to the selection information is as follows. [Pre-coding Algorithm] Step 1
Step i (i=2 to K-1)
For example, when the divided plaintexts are (m1, m2, m3, m4, m5)=(0, 1, 0, 1, 0), if the selection of the upper row is represented by 0 and the selection of the lower row is represented by 1, the precoded selection information of the upper and lower rows is (x1, x2, x3, x4, x5)=(0, 0, 1, 1, 0). The entity A generates the ciphertext C to the entity B as shown by (16) below, based on a set of public keys selected according to the bit pattern of a plurality of divided plaintexts. C=mivi(t1)w1+m22v2(t2)w1+ . . . +mK2K-1vK(tK)w1 (16)
For example, when the divided plaintext are (m1, m2, m3, m4, m5)=(0, 1, 0, 1, 0), since the selection information of the upper and lower rows of the public keys is (x1, x2, x3, x4, x5)=(0, 0, 1, 1, 0), the ciphertext C is concretely given as shown by (17) below. C=2v2(0)w1+23v4(1)w1 (17) The ciphertext C thus generated is transmitted from the entity A to the entity B via the communication path 1. Then, the ciphertext C is decrypted into the original plaintext X by the entity B. The decryption processing by the decryptor 5 of the entity B is carried out as follows. An intermediate decrypted text M1 found as shown by (18) below. M1≡C·w1-1(mod P1) (18) Let the selection information of the upper and lower rows be x1=0. Next, with the use of the component v1(0) of the upper row, m1 is found as shown by (19) below. m1≡M1·(v1(0))-1(mod 2) (19) The next intermediate decrypted text M2 is found as shown by (20) below. M2=M1-m1vi(0) (20) Supposing that x2=x1 xor m1, the next selection information x2 is found. Then, by considering that the upper row is selected when x2=0 and the lower row is selected when x2=1, m2 is found as shown by (21) below. m2=M2·(v2(x2))-1(mod 2) (21) Thereafter, in the same manner as for m2, the remaining m3, . . . , mK are decrypted. In the first embodiment as described above, the first base-product v1(1)w1 in the lower row of FIG. 3 is not used for decryption of the pre-coding. Since the number of rows in the public key list is 2 rows (J=2), in the first embodiment, the length of the input plaintext becomes twice longer, but the weight index=(average weight)/(concatenate plaintext length)=¼. Incidentally, the above-described algorithm for pre-coding divided plaintexts to selection information is merely an example and, needless to say, it is possible to use another example of algorithm for determining the selection information of public keys according to the bit pattern of a plurality of divided plaintexts. The following description will explain examples of the application of the first embodiment that achieve improved security. (Application of Multi-Stage Encryption) This is the application of the encryption method (the concept of multi-stage encryption) proposed in Japanese Patent Application No. 11-173338/1999 by the present inventors to the above-described encryption method, in which application ciphertext is generated by using the result of operating multi-stage modular-transformation by a plurality of random numbers on a public key selected for each divided plaintext. With respect to a base-product shown in FIG. 3, a plurality of sets (S sets) of a pair (w, P) of random number w and prime number P are set, multiplication by the random numbers are performed over S stages, and the result is used as a public key. Hence, by applying the multi-stage encryption technique to the basic encryption scheme of the first embodiment, it is possible to establish a scheme that achieves higher security. (Application of Product-Sum-Product Encryption) This is the application of the encryption method (the concept of product-sum-product encryption) proposed in Japanese Patent Application No. 11-205381/1999 by the present inventors to the above-described encryption method, in which application ciphertext is generated by setting a plurality of product-sum terms of the divided plaintexts and public keys selected for each divided plaintext and combining the plurality of the product-sum terms in the forms of product or sum. A part of divided plaintexts obtained by dividing plaintext and public keys selected for each of that part of the divided plaintexts are used to generate plural sets of product-sum terms as shown by (16) above, and multiplication and/or addition of the generated plural sets of the product-sum terms are further performed to generate ciphertext. Thus, by applying the product-sum-product encryption technique to the basic encryption scheme of the first embodiment, it is possible to establish a scheme that achieves higher security. As described in detail above, in the first embodiment, two public keys are prepared for each divided plaintext in advance, a plaintext to be encrypted is divided into a plurality of 1-bit divided plaintexts, one public key is selected among the two public keys prepared for each divided plaintext, according to a bit pattern of the plurality of divided plaintexts, and a ciphertext is generated by using the plurality of divided plaintexts and selected public keys. It is therefore possible to achieve high-speed encryption/decryption processing while ensuring security by free selection of public keys and to foster the development and practical use of the public-key encryption scheme. Second Embodiment The following description will explain a second embodiment in which public keys are selected according to the bit data of a plurality of divided plaintexts. FIG. 4 is a depiction showing a state in which an encryption scheme according to the second embodiment (second aspect) is used for information communications between the entities A and B. Like FIG. 2, the example shown in FIG. 4 illustrates a case where one of the entities, A, encrypts a plaintext X into a ciphertext C and transmits the ciphertext C to the other entity, B, via a communication path 11, and the entity B decrypts the ciphertext C into the original plaintext X. The entity A as the sender is provided with a plaintext divider 12 for dividing the plaintext X into a plurality of divided plaintexts, a public-key selector 13 for selecting a public key for each divided plaintext from a database 16 storing a public key list, and an encryptor 14 for generating the ciphertext C by using the selected public keys. Besides, the entity B as the recipient is provided with a decryptor 15 for decrypting the transmitted ciphertext C into the original plaintext X. In this example, the issuer of the public key list is the entity B as the recipient, and the user of this public key list is the entity A as the sender. Next, a specific technique will be explained. Note that the following explanation is given by illustrating an example in which s=1, i.e., each divided plaintext is one bit and two public keys are provided for selection with respect to each divided plaintext. FIG. 3 is an illustration showing the public key list in the database 16 that stores two public keys for each divided plaintext in advance. FIG. 3 shows a public key list in accordance with the supposition that a public key for each divided plaintext is constructed by modular transformation by (w1, P1). In FIG. 3, K represents a dividing number (class number) of the plaintext X, two (upper row, lower row) public keys including a random number term therein are prepared for each of K pieces of divided plaintexts (for each class). Besides, the random number vi(0) and random number vi(1) in FIG. 3 satisfy (22) and (23) below, respectively. vi(0)≡0(mod 2) (22) vi(1)≡0(mod 2) (23) After dividing the plaintext X into K pieces of 1-bit divided plaintexts, the entity A selects a public key according to the bit data of each of the divided plaintexts. In other words, when the divided plaintext is mi=0, a public key of the upper row, i.e., the base-product 2i-1vi(0), is selected, while when the divided plaintext is mi=1, a public key of the lower row, i.e., the base-product 2i-1vi(1), is selected. By sequentially adding the selected public keys, the ciphertext C to the entity B is generated as shown by (24) below. C=vi(t1)w1+2v2(t2)w1+ . . . +2K-1vK(tK)w1 (t1, t2, . . . , tK=0 or 1) (24) For example, when the divided plaintexts are (m1, m2, m3, m4, m5)=(0, 1, 0, 1, 0), the ciphertext C to the entity B is generated as shown by (25) below. C=v1(0)w1+2v2(1)w1+22v3(0)w1+23v4(1)w1 The ciphertext C thus generated is transmitted from the entity A to the entity B via the communication path 11. Then, the ciphertext C is decrypted into the original plaintext X by the entity B. The decryption processing by the decryptor 15 of the entity B is carried out as follows. An intermediate decrypted text M1 is found as shown by (26) below. M1≡C·w1-1(mod P1) (26) Here, it is apparent that the intermediate decrypted text M1 is expressed as shown by (27) below. Here, however, (28) shown below must be satisfied. [Eq. 5] M1+v1(m |2i-1vi(m Therefore, decryption can be performed by a decryption algorithm shown in (29) below. It will be appreciated that this decryption algorithm is extremely simplified. ##EQU2## The following description will explain the characteristics of the encryption scheme of the second embodiment by mainly discussing the comparison between this encryption scheme and a 0, 1-knapsack cryptosystem which is very close to this. There is a notable difference between the encryption scheme of the second embodiment and the conventional knapsack cryptosystem in that the encryption scheme of the second embodiment does not have Σmici form, i.e., is not of product-sum type but is of addition type. In the scheme of the second embodiment, the weight index=½ for the concatenate plaintext. For this sense, it would be considered that the scheme of the second embodiment is strengthened against concatenate attacks. The scheme of the second embodiment has the following significant characteristics in comparison with the conventional 0, 1-knapsack cryptosystem. In the scheme of the second embodiment, as the sum of ciphertext C shown in (30) below based on the public keys (c1, c2, . . . , cK) corresponding to the upper row of FIG. 3 and ciphertext C′ shown in (31) below based on the public keys (c1′, c2′, . . . , cK′) corresponding to the lower row of FIG. 3, ciphertext Cs is given as shown by (32) below. [Eq. 7] C=Σ{overscore (mi)}ci (30) C=Σmici (31) CS=C+C′ (32) For example, when the divided plaintexts are (m1, m2, m3, m4, m5)=(0, 1, 1, 0, 1), the ciphertext C and ciphertext C′ are given as shown by (33) and (34) below, respectively. [Eq. 8] C={overscore (m1)}c1+{overscore (m4)}c4 (33) C′=m2c′2+m3c′3+m5c′5 (34) The ciphertext C and ciphertext C′ are generated by multi-stage encryption, and designed to achieve higher security by different random numbers {vi}, {vi′}, respectively. The ciphertext Cs according to the scheme of the second embodiment is given as the sum of two apparently different ciphertexts of the knapsack cryptosystem. For this sense, is should be considered that a breakthrough was made in the 0, 1-knapsack cryptosystem. Regarding the LLL (Lenstra-Lenstra-Lovasz) attack, it is considered that it is extremely difficult for the attacker to attack since (the input plaintext lengh)/(ciphertext)=2 is satisfied. The following description will explain examples of the application of the second embodiment that achieve improved security. (Application of Multi-Stage Encryption) This is the application of the encryption method (the concept of multi-stage encryption) proposed in Japanese Patent Application No. 11-173338/1999 by the present inventors to the above-described encryption method, in which application ciphertext is generated by using the result of operating multi-stage modular-transformation by a plurality of random numbers on a public key selected for each divided plaintext. With respect to a base-product shown in FIG. 3, a plurality of sets (S sets) of a pair (w, P) of random number w and prime number P are set, multiplication by the random numbers are performed over S stages, and the result is used as a public key. Hence, by applying the multi-stage encryption technique to the basic encryption scheme of the second embodiment, it is possible to establish a scheme that achieves higher security. (Application of Product-Sum-Product Encryption) This is the application of the encryption method (the concept of product-sum-product encryption) proposed in Japanese Patent Application No. 11-205381/1999 by the present inventors to the above-described encryption method, in which application ciphertext is generated by setting a plurality of sum terms obtained by adding a plurality of selected public keys and combining a plurality of the sum terms in the form of product and/or sum. Plural sets of sum terms as shown by (24) above are generated with the use of a plurality of public keys selected according to the bit data of each divided plaintext, and multiplication and/or addition of the generated plural sets of the sum terms are further performed to generate ciphertext. Thus, by applying the product-sum-product encryption technique to the basic encryption scheme of the second embodiment, it is possible to establish a scheme that achieves higher security. Incidentally, in the above-described example, while the case where two public keys are provided for selection with respect to each divided plaintext (s=1) has been explained, it is possible to expand the application to the case where bi=2s(s: natural number no less than 2) by using a random number as shown by (36) below that satisfies (35) below. For example, when s=2, four public keys are prepared for each divided plaintext, a plaintext is divided into 2-bit divided plaintexts, one public key is selected for each divided plaintext among the four public keys according to the bit data of each divided plaintext, and a ciphertext is generated in the form of sum of all of the selected public keys. [Eq. 9] vi(m vi(m As described in detail above, in the second embodiment, 2s public keys including a random number term therein are prepared for each divided plaintext in advance, a plaintext to be encrypted is divided into a plurality of s-bit divided plaintexts, one public key is selected for each divided plaintext among the 2s public keys prepared for each divided plaintext, according to the bit data of each divided plaintext, and a ciphertext is generated by using the selected public keys. It is therefore possible to achieve high-speed encryption/decryption processing while ensuring security by free selection of public keys and to foster the development and practical use of the public-key encryption scheme. Further, while the above-described examples are illustrated for the cryptographic communication system, needless to say, it is possible to apply the encryption methods of the first and second embodiments of the present invention to the case where a ciphertext is by encrypting a plaintext and the generated ciphertext is simply recorded. Next, examples of a memory product and transmission medium of the present invention will be explained. FIG. 5 is an illustration showing the structures of embodiments of the memory product of the present invention. The programs exemplified here include a process for selecting a public key for each divided plaintext among a plurality of public keys stored in the database 6 (or 16) in advance, according to the data pattern of a plurality of divided plaintexts (or the bit data of each divided plaintext), and a process for generating ciphertext by using the selected public keys and divided plaintexts (or by using the selected public keys), or include a process for decrypting the ciphertext thus generated according to the above-described decryption algorithm, and are recorded in the memory product explained below. Besides, a computer 20 is provided for each entity. In FIG. 5, a memory product 21 to be on-line connected to the computer 20 is constructed by, for example, a WWW (World Wide Web) server computer installed at a distant point from the installation position of the computer 20, and a program 21a as mentioned above is stored in the memory product 21. The program 21a read from the memory product 21 through a transmission medium 24 such as a communication line controls the computer 20 to generate the ciphertext C, or decrypt the ciphertext C into the original plaintext X. A memory product 22 provided inside the computer 20 is constructed by, for example, a hard disk drive or ROM installed in the computer 20, and a program 22a as mentioned above is stored in the memory product 22. The program 22a read from the memory product 22 controls the computer 20 to generate the ciphertext C, or decrypt the ciphertext C into the original plaintext X. A memory product 23 which is used by loading it in a disk drive 20a provided for the computer 20 is constructed by, for example, a portable magneto-optical disk, CD-ROM or flexible disk, and a program 23a as mentioned above is stored in the memory product 23. The program 23a read from the memory product 23 controls the computer 20 to generate the ciphertext C, or decrypt the ciphertext C into the original plaintext X. As this invention may be embodied in several forms without departing from the spirit of essential characteristics thereof, the present embodiment is therefore illustrative and not restrictive, since the scope of the invention is defined by the appended claims rather than by the description preceding them, and all changes that fall within metes and bounds of the claims, or equivalence of such metes and bounds thereof are therefore intended to be embraced by the claims.
|
Same subclass Same class Consider this |
||||||||||