Multiple hashing in a cryptographic scheme

CA 02792575 2012-09-28

MULTIPLE HASHING IN A CRYPTOGRAPHIC SCHEME

BACKGROUND

This specification relates to generating hash values in a cryptography system. Cryptography systems enable secure communication over public channels. For example, digital signature schemes can be implemented in a public key cryptography system. In some cryptography systems, users verify the authenticity of other users’ digital signatures based on certificates issued by a trusted third party.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram of an example communication system.

FIG. 2 is a schematic diagram of an example hash-based function.

FIG. 3 is a signaling and flow diagram showing operations in an example implicit certificate scheme.

FIG. 4 is a signaling and flow diagram showing operations in an example digital signature scheme.

FIG. 5 is a signaling and flow diagram showing operations in another example digital signature scheme.

Like reference numbers in the various drawings indicate like elements.

DETAILED DESCRIPTION

In some implementations, short-output hash functions can be used in cryptographic schemes. For example, short-output hash functions may be used in signature schemes (e.g·, Pintsov-Vanstone signatures, keyed Pintsov-Vanstone signatures, etc.), certificate schemes (e.g., Optimal Mail Certificate (OMC) schemes, Elliptic Curve Qu-Vanstone (ECQV) implicit certificate schemes, etc.), and other types of cryptographic schemes. In some instances, a cryptographic scheme uses one or more hash functions that produce an output having a bit-length equal to the security parameter of the cryptographic scheme. A cryptographic scheme may use a hash function that produces an output having a different bit-length. In some contexts, cryptographic schemes that can use short-output functions are preferable, for example, because digests may be transmitted with less bandwidth (e.g., half as much), because

CA 02792575 2012-09-28 short-output hash functions may be implemented by reusing a block cipher, or because of a combination of these and other advantages.

[0001] In some implementations, a hash function can be represented as //: {0,1}* -> {0,1}\ where {0,1}* represents a binary string of arbitrary length, and (0, l}ⁱ represents a binary string of length I. In some instances, I = K, where κ represents the security parameter of a cryptographic scheme. One example of a hash function is AES128-MMO, which maps {0,1}* to {0,1}¹²⁸. A hash function may take other forms or have different properties. Other example hash functions are the SHA-family of hash functions, MD5, and others.

The security parameter κ is a value, typically an integer, that indicates the level of security that a cryptographic scheme is designed to provide. For example, a cryptography system operating with a security parameter of κ = 128 is designed to provide 128-bit security; a cryptography system operating with a security parameter of κ = 256 is designed to provide 256-bit security; etc. The security level in a cryptographic scheme generally relates to the difficulty of the underlying cryptographic problem. In some example Elliptic Curve Cryptographic (ECC) schemes, the security parameter κ = t, where represents a prime finite field, p is an odd prime number relating to the size of the finite field, and log₂ p ~ 2t. In other words, the elliptic curve cryptographic scheme is designed to supply κ = t bits of security —· meaning that solving the logarithm problem on the associated elliptic curve is believed to take approximately 2^f operations.

In some cases, one or more hash functions can be used in a manner that provides resistance to certain types of attacks against a cryptographic scheme. For example, to resist some types of attacks, some signature schemes require a hash value produced by a function that is collision-resistant. A hash function //: [0,1}* -» {0,1}⁽can be collision-resistant, for example, when it is computationally intractable to find distinct inputs m and m' such that H(m) = Some cryptographic schemes do not require a collision-resistant hash function.

Some cryptographic schemes require a hash value produced by a function that is random-prefix preimage (rpp) resistant, random-prefix second-preimage (rpsp) resistant, or both. A function is rpp-resistant and rpsp-resistant when the rpp problem and the rpsp problem are difficult. The problem of finding a preimage of a hash

CA 02792575 2012-09-28 function generally means, for a given value y, find m, such that H(m) = y. In the random prefix preimage (rpp) problem, the adversary trying to solve the rpp problem must choose a y value, is then given a random value R, and must then find m, such that H(R\\nï) = y. In the random-prefix second-preimage (rpsp) problem, the adversary must choose an m, is then given a random R, and must subsequently find an m’, such that H(R\\m) = H(R\\m’).

The rpp and rpsp problems are believed to require κ work for hash functions producing zc-bit outputs. But a “herding attack” may be used to solve the rpp and rpsp problems in O(2^3k/⁴) time for some types of hash functions (e.g., those that have the Merkle-Damgârd (MD) structure, and possibly others). The Merkle-Damgârd (MD) construction is a commonly-used way to construct a hash function H: {0,1}* -+ {0,1}⁽from a compression function h: (0,1}^μ -> (0,1}⁽, for some fixed μ> I. For example, common hash functions such as the SHA family and MD5 are constructed with the MD construction. As such, for some cryptographic schemes where κ bits of security are desired, a hash function H: (0,1}* -+ {0,1}* that produces a zc-bit hash value and has the MD structure may be insufficient to meet security requirements, in some instances.

In some implementations, a hash-based function F: {0,1}* -» {0,1}^K that produces a zc-bit output value can be used to meet κ-bit security requirements. For example, a hash-based function F that does not have the MD structure can be used as a countermeasure against (e.g., to reduce exposure to) a herding attack or another type of attack. In some instances, a hash-based function F that produces a zc-bit hash value can be rpp-resistant, rpsp-resistant, or both. In some implementations, the example function 200 shown in FIG. 2 does not have the MD structure, produces a κ-bit output value, and can be implemented in a manner that is rpp-resistant, rpsp-resistant, or both.

In some cases, truncating the output of a hash function can remove the hash function’s MD structure. For example, a 2zc-bit hash value produced by a hash function having the MD structure can be truncated to a zc-bit value. This may destroy the structure instilled by the MD construction, prevent the herding attack from being effective, and remove the 0(2^3k/⁴) solution to the rpp and rpsp problems. In some scenarios, a hash function with 2zc-bit outputs is not available. For example, block cipher-based hash functions, such as AES-MMO, produce a block-length output value.

CA 02792575 2012-09-28

In some AES systems the block length is fixed at 128 bits. A hash function that produces a 2/c-bit output may not be available in such systems when κ > 64 (as is typical in some systems). Moreover, in some environments hash functions that produce

2x-bit digests may be less efficient than functions that produce to-bit outputs, and they may require additional code.

In some implementations, one or more short-output hash functions that have the MD-structure can be modified, combined, or used in another manner to produce a different short-output function that does not have the MD-structure. The short-output function that does not have the MD-structure can be used to improve the security of PV signatures, keyed PV signatures, implicit certificates, and other schemes. In some cases, a hash function that is based on a block cipher and produces a to-bit output can be modified to remove its MD-structure. In such cases, the modified hash function may resist the herding attack without requiring a hash function that produces a 2/c-bit output. For example, herding attacks on AES-128-MMO may be prevented, in some instances, by hashing the message m as ff(ki||ff(w||k2))> where the keys ky and k₂are public values.

In a first example, the hash function F(m) = H(K®Cy\\H(K©c₂\\m)) may be used, where K is a key value (which may be a public key value), Cy and c₂ are public constants, and H is a hash function having the MD structure. Here, the operator © represents the bitwise exclusive-or (“XOR”) operation. The order in which the bit-wise XOR operation and the concatenation operation are done does not matter. In an example, the message m can be of arbitrary length, while K, Cy, c₂ can be of 128-bits. At least one of K, Cy and c₂ is of non-zero length. In this example, both F: {0,1}* -» (0,1}^K and H: (0,1}* -> (0,1}^K produce a to-bit output; but even when H has the MD structure, F does not have the MD structure. As such, the hash function F can be used as a counter measure against certain types of attacks. This example uses the MAC construction HMAC. Because removing the MD structure may, in some contexts, be simpler than computing the secure MAC of a message, the function F in this example may be simplified, for example, to F(m) = H(H(0^K\\m)) to obtain the same or similar advantages. Here, 0^K represents a to-bit string of zeros. Other constant values could be used, such as, for example, a to-bit string of ones or a different to-bit constant value.

CA 02792575 2012-09-28

In a second example, prefix-free encoding may be used. Prefix-free encoding can be accomplished by converting the hash function input to a prefix-free format. In other words, each hash function input can be reformatted, e.g., by applying a reformatting function before applying the hash function, to map the input to a value in prefix-free set. A set of strings Σ can be considered prefix-free when, for all distinct x,y e Σ, x is not a prefix of y. For example, fixed-length inputs will be prefix-free. Using a prefix-free encoding in the hash function can remove the MD structure.

Prefix-free encoding may be used, for example, in cases where a hash function uses a block cipher and in other cases. A hash function that uses a block cipher typically divides an input into blocks of equal length and operates sequentially on the blocks. As such, hash functions that use a block cipher typically define a fixed blocklength as an operating parameter. An example of a hash function that uses a block cipher is the AES-MMO hash function, which uses an AES block cipher. In some instances, the AES-128-MMO hash function divides the input into 128-bit blocks and produces a 128-bit hash value as output. The AES-MMO hash function may operate, for example, based on sequentially applying an AES encryption function. Prefix-free encoding may be used in other cases as well.

Prefix-free encoding can be accomplished in a number of ways. For example, there are multiple different ways of converting hash function inputs to prefix-free formats. In some instances, before applying the hash function to the input, a value identifying the number of blocks in the input is prepended to the input. When a value is prepended to the input, the value is joined to the beginning of the input. Alternatively, in some instances, before hashing the input, a zero bit (“0”) is appended to each block of the input, except for the last block where a one bit (“1”) is appended. When a value is appended to a block, the value is joined to the end of the block. In this example, the input can first be divided into blocks that are one bit shorter than they normally would be, to leave space for the extra bit. As another alternative, before applying the hash function to the input, the input can be converted to a fixed-length input format. This can be accomplished, for example, by defining the input length as part of the system parameters in the description of the hash function. Additional or different techniques may be used.

CA 02792575 2012-09-28

In a third example, the hash function F(m) = G(//(m)) may be used, where G: {0,1}* -+ {0,1}* and H: {0,1}* -* {0,1}^K (or H: {0,1}* -+ {0,1}^K) are two different hash functions. As such, this example utilizes two distinct hash functions. Here, F: (0,1}* -+ {0,1}\ G: (0,1}* -+ {0,1}^K and H: (0,1}* -+ {0,1}^K each produce a x-bit output. This construction may be useful, for example, in contexts where two different hash functions are available. Here, F, G, and H produce a x-bit output; but even when H, G, or both have the MD structure, F does not have the MD structure.

[0002] Each of these three examples can be proven to be secure, for example, by showing that they are indistinguishable from a random oracle. The notion of the random oracle is captured with the following game between a challenger and attacker, for a fixed construction F (which may be any of the examples above, or another).

1. The challenger chooses p 6 {real, ideal} at random.

2. The attacker is allowed to adaptively make queries for messages M G {0,1}*, and receives responses in {0,1}^K. When p is real, the response is H(M); when p is ideal, the response is a random value from {0,1}^K. For simplicity, assume all queries are distinct.

3. The adversary outputs a guess p' £ {real, ideal}. The function H is said to be indistinguishable from a random oracle if no adversary can output p' — p with probability significantly different from 1/2.

A more technical definition can be used, for example, to model the underlying compression function h (used to construct F and H) as a random oracle or ideal block cipher. Each of the three examples provided above emulate a random oracle provided h is ideal.

FIG. 1 is a schematic diagram of an example communication system 100. The communication system 100 includes a certificate authority server 104, two terminals 102, 106, and a data network 108. The communication system 100 can include additional, fewer, or different components. For example, the communication system 100 may include additional storage devices, additional servers (including additional certificate authority servers), additional terminals, and other features not shown in the figure.

CA 02792575 2012-09-28

The certificate authority server 104 and the terminals 102, 106 can communicate with each other and with other components of the communication system 100 over the data network 108. In the example shown in FIG. 1, the terminal can send a certificate request 120 to the certificate authority server 104, and the certificate authority server 104 can respond by sending an implicit certificate 122 to the terminal 102. The terminal 102 can send a signed message 124 to the terminal 106, and the terminal 106 can verify the authenticity of the signed message 124 using the implicit certificate 122. The communication system 100 can support additional or different types of communication. In some implementations, the terminals 102, 106 can also exchange encrypted messages and other types of information with each other, with the certificate authority server 104, and with other components of the communication system 100.

The example certificate authority server 104 is a computing device that can perform operations of a certificate authority in a cryptography system. The certificate authority server 104 generally includes one or more data processing apparatus, data storage media, and data communication interfaces. The certificate authority server 104 is generally operable to receive, transmit, process, and store information associated with the cryptography system. Although FIG. 1 shows a single certificate authority server 104, a certificate authority can be implemented using multiple certificate authority servers 104, including server clusters, as well as additional or different types of computing devices other than servers.

The certificate authority server 104 shown in FIG. 1 can be implemented as a computing system that includes one or more processors 112, memory 110, and one or more input/output controllers 114. The memory 110 can include, for example, random access memory (RAM), storage devices (e.g., a writable read-only memory (ROM), etc.), a hard disk, or other types of storage media. A computing system can be preprogrammed or it can be programmed (and reprogrammed) by loading a program from another source (e.g., from a DVD, from another computer device through a data network, or in another manner). The input/output controller 114 can be coupled to input/output devices (e.g., a monitor, a keyboard, etc.) and to the data network 108. The input/output devices receive and transmit data in analog or digital form over

CA 02792575 2012-09-28 communication links such as a serial link, wireless link (e.g., infrared, radio frequency, etc.), parallel link, or another type of link.

The memory 110 can store instructions (e.g., computer code) associated with computer applications, programs and computer program modules, and other resources.

For example, the memory 110 can store instructions associated with the computer program modules of a cryptography system. The memory 110 can also store application data and data objects that can be interpreted by applications, programs, modules, or virtual machines running on the computing system. For example, the memory 110 can store the data objects that are obtained or processed by the computing system. The memory 110 can store additional information, for example, files and instruction associated with an operating system, device drivers, archival data, or other types of information.

The processors 112 can execute instructions to generate output data based on data inputs. For example, the processors 112 can run applications and programs by executing or interpreting the software, scripts, functions, executables, and other types of computer program modules. In some instances, a processors may perform one or more of the operations shown in FIGS. 2, 3, 4 and 5. The input data received by the processor 112 and the output data generated by the processor 112 can be stored in a computer-readable medium, such as the memory 110 or a storage device.

The data network 108 can include any type of data communication network.

For example, the data network 108 can include a wireless or wired network, a cellular network, a telecommunications network, an enterprise network, an application-specific public network, a Local Area Network (LAN), a Wide Area Network (WAN), a private network, a public network (such as the Internet), a WiFi network, a network that includes a satellite link, or another type of data communication network. The data network 108 can include a tiered structure defined by firewalls or similar features that implement various levels of security.

The example terminals 102, 106 are computing devices that can communicate over the data network 108 based on cryptographic communication schemes. The terminals 102, 106 each generally include one or more data processing apparatus, data storage media, and data communication interfaces. The terminals 102, 106 are generally operable to receive, transmit, process, and store information. Although FIG.

CA 02792575 2012-09-28 shows two terminals 102, 106, a communication system 100 may include any number of terminals. The communication system 100 can include groups or subgroups of terminals that can communicate with each other, but not necessarily with the terminals in other groups or subgroups. In some implementations, each group of terminals can access a dedicated certificate authority server and a database of implicit certificates that have been issued by the dedicated certificate authority server. The communication system 100 can include terminals of disparate types, having different types of hardware and software configurations, and in a variety of different locations. In some cases, multiple devices or subsystems can be identified together as a single terminal.

The example terminals 102, 106 shown in FIG. 1 can each be implemented as computing systems that include memory, one or more data processors, and one or more input/output controllers. A terminal can include user interface devices, for example, a monitor, touchscreen, mouse, or keyboard. The terminals 102, 106 can interface with the data network 108, for example, through a wired connection, a wireless connection, or any suitable combination. The memory of the terminal can store messages and information associated with the cryptography system. For example, a terminal may store public and private key data, digital certificate data, and other types of information. The memory of the terminal can store instructions (e.g., computer code) associated with computer applications, programs and computer program modules, and other resources. For example, the terminals can store instructions associated with the computer program modules of a cryptography system.

Terminals can include handheld devices such as smart phones, personal digital assistants (PDAs), portable media players, laptops, notebooks, tablets, and others.

Terminals can include work stations, mainframes, non-portable computing systems, devices installed in structures, vehicles, and other types of installations. Terminals can include embedded communication devices. For example, the terminals can include messaging devices that are embedded in smart energy meters of a smart energy system. Other types of terminals may also be used.

In one aspect of operation, the terminal 102 sends the certificate request 120 to the certificate authority server 104, and the certificate authority server 104 generates the implicit certificate 122 for the terminal 102. The implicit certificate 122 associates

CA 02792575 2012-09-28 a particular public key value with a particular entity (e.g., the terminal 102, a user associated with the terminal 102, a software module implemented at the terminal 102, or another type of entity). The terminal 102 receives the implicit certificate 122 from the certificate authority server 104. When the terminal 102 has a message to send to the terminal 106, the terminal 102 can generate a digital signature for the message based on the implicit certificate 122. The digital signature can be combined with the message to form the signed message 124, which the terminal 102 sends to the terminal 106. In some implementations, the digital signature and the message are sent separately. The terminal 106 receives the signed message 124, obtains the implicit certificate 122, and verifies the digital signature based on the implicit certificate 122. Implicit certificates can also be used in other types of schemes, for example, encryption schemes.

The implicit certificate 122 binds an entity (e.g., the terminal 102, a user associated with the terminal 102, a software module implemented at the terminal 102, or another type of entity) to a particular public key value that can be used to verify digital signatures generated by the terminal 102. The terminal 106 can obtain the implicit certificate 122 to verify that the digital signature was generated by the entity associated with the terminal 102, and not by an impostor. The terminal 106 can also verify that the implicit certificate 122 was generated by a trusted third party. In this manner, the implicit certificate 122 serves as confirmation by the trusted third party that the signed message 124 was signed by the entity associated with the terminal 102 and not by an impostor.

The example implicit certificate 122 includes an identification of the entity associated with the terminal 102. The example implicit certificate 122 also includes information that can be used to construct the entity’s public key. In some cases, using the implicit certificate 122 to verify a digital signature also confirms that the entity is in possession of the corresponding private key. The example implicit certificate 122 shown in FIG. 1 includes neither an explicit representation of the public key nor an explicit representation of the certificate authority’s digital signature. Thus, in some implementations, the implicit certificate 122 is more compact than some other types of digital certificates. In some cases, the implicit certificate 122 includes a digital signature of the certificate authority that allows entities, for example an entity

CA 02792575 2012-09-28 associated with the terminal 106, to verify that the implicit certificate 122 was generated by the trusted certificate authority. The certificate authority can, in some cases, require the entity to prove knowledge of the entity’s private key. In some cases, the implicit certificate 122 includes an explicit representation of the user’s public key.

Instead of explicitly representing the public key of the terminal 102, the example implicit certificate 122 in FIG. 1 includes public key reconstruction data that can be combined with other information (e.g., the certificate authority’s public key, etc.) to generate the public key of the entity associated with the terminal 102. The example implicit certificate 122 is constructed such that successful verification of a digital signature generated by the terminal 102 serves as confirmation that the terminal 102 is in possession of the private key. Thus, according to some implicit certificate schemes, binding of an entity to its public key and the entity’s knowledge of its private key can be verified in unison during key usage.

Implicit certificate schemes can be implemented based on different types of groups. For example, the ECQV implicit certificate scheme, as well as others, may be implemented using a group of points on an elliptic curve, a multiplicative group of a finite field, or other groups where the discrete logarithm problem may be hard.

In some instances, the terminals 102, 106 can communicate according to a cryptographic scheme that does not involve a certificate authority or digital certificates. For example, the terminals 102, 106 may communicate based on a signature scheme, an encryption scheme, or another type of cryptographic scheme that does not use public keys. As another example, the terminals 102, 106 may communicate based on a signature scheme, an encryption scheme, or another type of cryptographic scheme that uses public keys that are not certified by a certificate authority. Moreover, in some cases, some or all operations of the certificate authority can be performed by one or both of the terminals 102, 106. In some instances, for example in a peer-to-peer computing environment, server functionality can be distributed among client devices.

FIG. 2 is a schematic diagram of an example hash-based function 200. The example hash-based function 200 shown in FIG. 2 can be implemented using additional, fewer, or different operations, which can be performed in the order shown or in a different order. In some implementations, one or more of the operations can be

CA 02792575 2012-09-28 repeated or iterated, for example, until a terminating condition is reached. The hashbased function 200 can be implemented by an entity in the cryptography system, such as, for example, the certificate authority server 104 or the terminals 102, 106 shown in FIG. 1. The hash-based function 200 may be executed by a special-purpose processor, by a general-purpose processor, or by any suitable combination of these and other devices, as appropriate. In some instances, the hash-based function 200 is executed by a hash module. The hash module may include data processing apparatus (e.g., one or more a special-purpose processors, one or more a general-purpose processor, etc.), software, firmware, or any suitable combination of these and other resources that are configured to perform the operations shown in FIG. 2.

The hash-based function 200 can be applied to a variety of cryptographic schemes. In some cases, the hash-based function 200 can be used to implement a hash function in a signature scheme such as, for example, the Elliptic Curve PintsovVanstone (ECPV) signatures schemes described in: the ANSI X9.62-2005 standard (Public Key Cryptography for the Financial Services Industry: The Elliptic Curve Digital Signature Algorithm (ECDSA)), the IEEE Pl363a standard (Standard Specifications for Public Key Cryptography: Pintsov-Vanstone Signatures with Message Recovery); the ISO/IEC 9796-3:2006 standard (Information technology Security techniques - Digital signature schemes giving message recovery - Part 3: Discrete logarithm based mechanisms); the SuiteE standard (A Cryptographic Suite for Embedded Systems); and possibly others. In some cases, the hash-based function 200 can be used to implement a hash function in a keyed ECPV signature scheme. Some example keyed ECPV signature schemes allow the signer to encrypt the recoverable part of the signed message in such a way that only the intended recipient may recover it; other parties may still verify the signature on the public part of the message. As another example, the hash-based function 200 can be used to implement a hash function in a digital signature scheme such as, for example, the elliptic curve QuVanstone (ECQV) implicit certificate schemes described in: the SEC 4 standard (Elliptic Curve Qu-Vanstone Implicit Certificate Scheme (ECQV)), and the SuiteE standard.

In some instances, all entities in the cryptography system can access the hashbased function 200. For example, the entities may each have one or more data

CA 02792575 2012-09-28 processing apparatus operable to implement the function. As such, in some example signature schemes, the signer and verifier both have access to the hash-based function 200; in some implicit certificate schemes, the requester, the relying party, and the certificate authority all have access to the hash-based function 200. An identification of the hash-based function 200 may be distributed to the entities, for example, in the domain parameters of the cryptographic scheme, or the entities may agree in advance to use the function 200 in another manner.

At a high level, the example hash-based function 200 receives a function input 202 and produces a function output 220. The function input 202 can include a single input or it can include multiple inputs. The function input 202 can be formatted as any suitable data format (e.g., binary, numeric, text, etc., as appropriate). The function output 220 can include a single output or it can include multiple outputs. The function output 220 can be formatted according to any suitable data format (e.g, binary, numeric, text, etc., as appropriate). In some implementations, the hash-based function can be expressed F: {0,1}*-> {0,1}^Z, as a function that maps variable-length bit strings to fixed-length bit strings of length I. In some implementations, the function output 220 is a binary value of length K, where κ represents the security parameter of a cryptography system.

In some implementations, the hash-based function 200 is indistinguishable from a truly random function. For example, from an adversary’s perspective the hashbased function 200 may emulate a random oracle. In some implementations, the hashbased function 200 does not have the MD structure and can be used as a countermeasure against (e.g., to reduce exposure to) a herding attack or another type of attack. In some implementations, the hash-based function 200 is rpp-resistant, rpspresistant, or both. The hash-based function 200 may also be collision-resistant in some instances.

The function input 202 can include any suitable data, data objects, files, variables, or other information. The function input 202 may include, for example, an integer, an elliptic curve point, alphanumeric information, an identifier, or any other suitable type of information. The function input 202 can represent any suitable information. The function input 202 may represent, for example, key data (e.g., public key, private key, symmetric key, short-term key, long-term key, etc.), certificate data

CA 02792575 2012-09-28 (e.g., implicit certificate data, explicit certificate data, etc.), encoded message data, plaintext message data, signature data, etc.

In some example digital certificate schemes, the function input 202 includes a digital certificate. For example, the function input 202 can be the implicit certificate

Certy of the ECQV scheme represented in FIG, 3. The function input 202 may represent additional or different information in a digital certificate scheme, as appropriate. In some example digital signature schemes, the function input 202 includes message data. For example, the function input 202 can be C||V, the encoded message part C concatenated with the second message part V in the example ECPV signature scheme represented in FIG. 4. As another example, the function input 202 can be C_± 11C₂11V, the first encoded message part Q concatenated with the second encoded message part C₂ concatenated with the third message part V in the example keyed ECPV signature scheme represented in FIG. 5. The function input 202 may represent additional or different information in a digital signature scheme, as appropriate.

At 204, a first input is prepared based on the function input 202. The first input can be prepared based on additional or different data. Preparing the first input generally includes generating an input for the first hash function. In some instances, the first hash function is applied directly to the function input 202, and no manipulation or arithmetic is required to prepare the first input.

In some implementations, preparing the first input includes parsing the first input into multiple blocks of equal length. For example, when the first hash function includes an encryption function (e.g., AES in counter or cipher block chaining mode, etc.) or another type of operation that operates on blocks of data, the function input

202 can be parsed into blocks of the appropriate length.

In some implementations, preparing the first input includes concatenating a value with the function input 202. The concatenation may be expressed v||m, where m represents the function input 202 and v represents the value concatenated with the function input 202. In some implementations, a constant value (e.g. a bit string of zeros) is concatenated with the function input 202. In some instances, the length of the constant is equal to the length of the first hash function’s output. For example, if the first hash function produces a hash value 208 of length k, then preparing the first input

CA 02792575 2012-09-28 may include concatenating 0^K (a κ-bit string of zeros) to the function input 202 produce 0^K||m, where m represents the function input 202. In this example, the 0^K bit string is used; different bit strings may be used, as appropriate. In some implementations, the value v is obtained based on a public key value. For example, the value v may be obtained by computing the XOR value X©c₂, where K is a public key and c₂ is a public constant. The first input can be generated by concatenating the XOR value with the function input 202 to produce K®c₂1\m.

At 206, a first hash function is applied to the first input. The first hash function can be any suitable hash function. In some instances, the first hash function can be represented as //: (0,1}* -> (0, l}k In some instances, I = κ, where κ represents the security parameter of a cryptographic scheme. The first hash function may produce an output of any suitable length, including 128 bits or another length. The first hash function can have the Merkle-Damgârd (MD) structure. For example the first hash function //: {0,1}* -> [0,1}* can be formed from a compression function h: {0,1}^μ ->

{0, l}¹, for some fixed μ > I. A first hash function may take other forms or have different properties. In some implementations, the first hash function can be an AESMMO hash function, a SHA-family hash function, MD5, or a different hash function, as appropriate.

Applying the first hash function to the first input produces a first hash value

208. At 210, a second input is prepared based on the first hash value 208. The second input can be prepared based on additional or different data. Preparing the second input generally includes generating an input for the second hash function. In some instances, the second hash function is applied directly to the first hash value 208, and no manipulation or arithmetic is required to prepare the second input.

In some implementations, preparing the second input includes parsing the second input into multiple blocks of equal length, as described above. In some implementations, preparing the second input includes concatenating a value with the first hash value 208. The concatenation may be expressed w|\u, where u represents the first hash value 208 and w represents the value concatenated with the first hash value

208. In some implementations, the value w is obtained based on a public key value.

For example, the value w may be obtained by computing the XOR value K®c₂, where

CA 02792575 2012-09-28

F is a public key and c_r is a public constant. The second input can be generated by concatenating the XOR value with the first hash value 208 to produce F©c₁||u.

At 212, a second hash function is applied to the second input. The second hash function can be any suitable hash function. The second hash function can be the same hash function as the first hash function, which was applied at 206. For example, the second hash function can be applied by executing the same software, logic, scripts, modules, functions, or processes that were executed in applying the first hash function.

In some implementations, the second hash function is not the same hash function that was applied to the first input at 206. In other words, the first hash function and the second hash function can be two different hash functions. As a particular example, the first hash function could be MD5 and the second hash function could be SHA-1. Other combinations of hash functions could be used.

In some instances, the second hash function can be represented as G : {0,1}* -> {0,1}^Z. In some instances, I = zc, where zc represents the security parameter of a cryptographic scheme. The first and second hash functions may produce outputs of the same length, or they may produce outputs of different length. The second hash function can have the Merkle-Damgârd (MD) structure. For example the second hash function G: (0,1}* -+ (0,1}⁽ can be formed from a compression function g: {0,1}^μ -+ {0,1}^Z, for some fixed μ > I. The second hash function may take other forms or have different properties. In some implementations, the second hash function can be an AES-MMO hash function, a SHA-family hash function, MD5, or a different hash function, as appropriate.

Applying the second hash function to the second input produces the second hash value 214. At 216, the function output 220 is prepared based on the second hash value 214. The function output 220 can be prepared based on additional or different data. Preparing the function output 220 generally includes generating an output of the appropriate size and format. In some instances, the second hash value 214 is directly provided as the function output 220, and no manipulation or arithmetic is required to prepare the function output 220. In some instances, the hash value 214 is converted to an integer or another data format at 216.

CA 02792575 2012-09-28

The example hash-based function 200 shown in FIG. 2 includes two hash functions, but generally a hash-based function can include any suitable number of hash functions, as appropriate. As such, the example hash-based function 200 can be extended to include three or more hash functions.

In some aspects, the example hash-based function 200 shown in FIG. 2 can be represented as two hash functions operating on the function input 202. For example, when the first and second hash functions are the same hash function H, the hash-based function 200 can be represented as HÇKQcy | |//(F©c₂1 |m)), as H(H(0^K||m)), or in another suitable manner (where m represents the function input 202). As another example, when the first and second hash functions are different hash functions H and G respectively, the hash-based function 200 can be represented H(G(rri)) or in another suitable manner (where m represents the function input 202). In addition to these three examples, other variations may also be used to achieve one or more of the advantages contemplated here.

The function output 220 can include any suitable data, data objects, files, variables, or other information. In some examples, the function output 220 is a fixedlength binary value. The function output 220 can be used for any suitable purpose, for example, in cryptographic operations. In some instances, the function output 220 may be combined with key data (e.g., public key, private key, symmetric key, short-term key, long-term key, etc.), certificate data (e.g., implicit certificate data, explicit certificate data, etc.), encoded message data, plaintext message data, signature data, etc. As another example, the function output 220 may be transmitted to, or included in a message transmitted to, a correspondent in a cryptography system.

In some example digital certificate schemes, the function output 220 is combined with one or more values to produce a private key reconstruction value. For example, the function output 220 can be the hash value e in the ECQV scheme represented in FIG. 3. A certificate authority may combine the function output 220 with a private value k and a certificate authority’s private key d_CA to produce the private key contribution value r = ek + d_CA (modn). A certificate requester or a correspondent may combine the function output 220 with a public key reconstruction value P_u and a certificate authority’s public key Q_CA to produce the requester’s public key Qu = eP_u + Qca- A certificate requester may combine the function output 220

CA 02792575 2012-09-28 with a private value ky and a private key contribution value r to produce the requester’s private key dy = eky + r (mod ri). The function output 220 can be used in other ways and for additional or different purposes in a digital certificate scheme, as appropriate.

In some example digital signature schemes, the function output 220 is combined with one or more values to produce a digital signature component. For example, the function output 220 can be the hash value e in the ECPV scheme represented in FIG. 4 or in the keyed ECPV scheme represented in FIG. 5. A signer may combine the function output 220 with the integer z and the signer’s longer term private key d_A to produce the signature component s = z — d_Ae (modn). A verifier may combine the function output 220 with the signer’s long-term public key G_A, the base-point generator G, and the signature component s to produce the elliptic curve point Q' = sG + eG_A. The function output 220 can be used in other ways and for additional or different purposes in a digital signature scheme, as appropriate.

FIGS. 3, 4, and 5 show examples of cryptographic schemes where the hashbased function 200 may be used. For example, in some instances, the function F shown in FIGS. 3, 4, and 5 can be implemented using the hash-based function 200 shown in FIG. 2. The hash-based function 200 can also be used in other contexts and for other purposes.

The examples shown in FIGS. 3, 4, and 5 are ECC-based cryptographic schemes. In particular, FIG. 3 shows example operations in an elliptic curve QuVanstone (ECQV) implicit certificate scheme; FIG. 4 shows example operations in an elliptic curve Pintsov-Vanstone (ECPV) signature scheme; and FIG. 5 shows example operations in a keyed variant of the ECPV signature scheme. The hash-based function

200 can be used in these example ECC-based cryptographic schemes as well as other types of ECC-based cryptographic schemes. In some implementations, the hash-based function 200 may also be used in other suitable types of cryptographic schemes.

In an ECC-based cryptographic scheme, information is encoded in elliptic curve points of an elliptic curve group. An elliptic curve group can be described in terms of a solution to an equation over a finite field, for example,, a prime finite field or a characteristic-two finite field. Each point in the elliptic curve group is a pair of field elements corresponding to a solution to an elliptic curve equation. The elliptic curve

CA 02792575 2012-09-28 group also includes an identity element. As a particular example, let F_p represent a prime finite field where p is an odd prime number, and let a, b G F_p satisfy 4. a³ +

27. b² 4= 0 (mod p). The elliptic curve group £(F_p) over F_p, which is defined by the parameters a,b G F_p, includes the set of points P = (x, y) for x, y G F_p that represent a solution to the equation y² = x³ + a.x + b (mod p), together with a point at infinity 0 that is the identity element of the elliptic curve group £(F_p).

In an ECC-based scheme, elliptic curve domain parameters over F_pcan be identified by a sextuple T = (p, a, b, G, n, h). The integer p specifies the finite field F_p. Field elements a, b G F_p specify an elliptic curve F(F_p) over F_p as discussed above.

The elliptic curve point G = (x_G, y_G) on E(F_p) is a base point generator. The integer n specifies the order of the base point generator G, having the property nG = 0. The cofactor h is equal to #F(F_p)/n, which is the number of points on the elliptic curve F(F_p) divided by the order of the base point generator G. Elliptic curve domain parameters may alternatively be identified over other types of finite fields. For example, elliptic curve domain parameters over the characteristic two field F₂™ can be identified by a septuple T = (m,f(x),a,b,G,n,h), where m is an integer specifying the finite field F₂m and /(x) is an irreducible binary polynomial of degree m specifying the representation of F₂m . In some implementations, the elliptic curve domain parameters can be generated, validated, and utilized by a terminal associated with an entity or by a certificate authority.

In an ECC-based scheme, an elliptic curve key pair (d, Q) can be generated based on valid elliptic curve domain parameters, for example, T = (p, a, b, G, n, h) or T = (m,f(x),a, b, G,n, h). The key pair may be generated by selecting a random integer d in the interval [l,n — 1], computing Q = dG, and outputting the key pair (d,Q). The random integer d may be selected or obtained by a random number generator. In some implementations, elliptic curve key pairs can be generated, validated, and processed by a terminal associated with an entity or by a certificate authority.

Elliptic curve key pairs can be validated using multiple different types of techniques. Validating an elliptic curve key pair provides assurance that the public key satisfies the arithmetic requirements of the cryptography system, for example, to

CA 02792575 2012-09-28 prevent malicious insertion of an invalid public key to enable attacks or to detect inadvertent coding or transmission errors. For example, a public key can be validated by checking that Q A 0, checking that nQ Φ 0, and checking that the public key Q satisfies the elliptic curve equation specified by the elliptic curve domain parameters

T = (p,a,b,G,n,fi) or T = (m,f(x),a,b,G,n,h), for example, based on the coordinates (xq, y_ç) of the elliptic curve point specified by the public key Q.

FIG. 3 is a signaling and flow diagram showing operations in an example implicit certificate scheme 300. The example implicit certificate scheme 300 includes operations performed by a requester 302, a certificate authority 304, and a correspondent 306. The requester 302 and the correspondent 306 can be, or they can be associated with, entities in a cryptography system. For example, the requester 302 can be the terminal 102 in FIG. 1, and the correspondent 306 can be the terminal 106 in FIG. 1. Similarly, the certificate authority 304 can be, or it can be associated with a certificate authority entity. For example, the certificate authority 304 can be the certificate authority server 104 in FIG. 1. The entities represented in FIG. 3 can be implemented by additional or different types of hardware, software, systems, devices, or combinations of these, as appropriate.

At a high level, the requester 302 sends a request to the certificate authority 304 and receives an implicit certificate in response. The implicit certificate allows the correspondent 306 to reconstruct a certified public key of the requester 302. In an example shown, the certificate authority 304 can obtain the certificate authority’s key pair (d_CA, Q_Ca)> and the requester 302 and the correspondent 306 can receive authentic copies of the certificate authority’s public key and domain parameters.

The example operations and communications shown in FIG. 3 are described with respect to the ECQV implicit certificate scheme. The operations and communications shown in FIG. 3 can be adapted or modified for additional or different types of implicit certificate schemes. In the example shown in FIG. 3, the requester 302, the certificate authority 304, and the correspondent 306 are each in possession of elliptic curve domain parameters (e.g., T = (p, a, b, G,n, hf) and a random number generator, and they have agreed on a function F that produces a hash value, a certificate encoding scheme, valid data types, and other parameters necessary to carry

CA 02792575 2012-09-28 out the operations shown. In addition, in some implementations the requester 302 has been assigned a unique identifier U.

At 310, the requester 302 generates a certificate request. For example, the requester 302 can generate the request by selecting a value k_u G_R [l,n— 1] and computing Ry = kyG. At 312, certificate request data are sent from the requester 302 to the certificate authority 304. For example, the certificate request data can include the requester’s identifier U and the elliptic curve point R_υ in the appropriate data formats. The data may be transmitted in any suitable manner, using any suitable hardware. In some implementations, the request data are transmitted over a data network, over one or more wired or wireless links, etc. When the certificate authority 304 receives the request, the certificate authority 304 can verify the identity of the requester 302, perform validity checks, and determine that an implicit certificate will be issued.

At 314, the certificate authority 304 generates an implicit certificate. The certificate authority 304 encodes the public key reconstruction data for the requester 302 in the implicit certificate. The public key reconstruction data can include elliptic curve point Pu. The elliptic curve point Pu can be generated, for example, by computing Py = Ry + kG. Here, k G_R [l,n — 1] represents an integer, and G represents the base point generator specified in the domain parameters. In some implementations, k is a random number in the interval [1, n — 1] generated using the random number generator at the certificate authority. Additional or different techniques can be used to generate the public key reconstruction data.

In the example shown in FIG. 3, the certificate authority 304 generates the implicit certificate Certy by calling a certificate encoding routine

Certy = Encode(Py,U,*), generates the hash value e by computing e = F(Certy), and generates the private key contribution data r = ek + d_CA mod n. The implicit certificate data can be encoded in Certy using any suitable encoding scheme. For example, a fixed-length field scheme, a minimal ANS.l encoding scheme, an X.509compliant ANS.l encoding scheme, or another suitable scheme can be used.

In the example shown in FIG. 3, the certificate authority 304 obtains the hash value e by evaluating a function F(Certy). The hash value e can be generated using any appropriate hash function or hash-based function F, and the function F can be

CA 02792575 2012-09-28 evaluated based on additional or different input, as appropriate. In some implementations, the function F is the hash-based function 200 shown in FIG. 2.

In some implementations, after generating the implicit certificate and the private key contribution data, the certificate authority 304 generates the requester’s public key Qy. The certificate authority 304 may generate the requester’s public key, for example, by computing Qy = ePy + Q_CA. Here, Q_CA = d_CAG is the certificate authority’s public key that corresponds to the certificate authority’s private key d_CA. The certificate authority 304 may generate the requester’s public key Qy, for example, to verify that the public key Qy corresponds to a point on the elliptic curve specified by the domain parameters, to verify that the public key Qy is not the point at infinity O, to verify that the public key Qy has not already been assigned to another entity, or for another suitable purpose.

At 316a, certificate data are sent from the certificate authority 304 to the requester 302. For example, the certificate data can include the values r and Certy in the appropriate data formats. The requester 302 receives the certificate data from the certificate authority 304. At 316b, implicit certificate is sent from to the correspondent 306. For example, the Certy and additional information may be sent to the correspondent 306 in the appropriate data formats. The correspondent 306 may receive the implicit certificate from the certificate authority 304, the requester 302, or another source. The data may be transmitted in any suitable manner, using any suitable hardware. In some implementations, the certificate data are transmitted over a data network, over one or more wired or wireless links, etc.

At 318, the requester 302 uses the certificate data to generate the requester’s elliptic curve key pair (dy, Qy). The requester 302 can generate the elliptic curve key pair (dy, Qy) by computing the hash value e = F(Certy), computing the private key value dy = eky + r (mod ri), and computing the public key value Qy = ePy + Q_CA. Here, the requester 302 generates the hash value e using the same function F that the certificate authority 304 used to generate the hash value e when generating the certificate data at 314.

The requester 302 may use the hash value e to generate the requester’s public key Qy or the requester’s private key dy or both for any suitable purpose. For example, the requester 302 may verify that the public key Qy corresponds to a point on

CA 02792575 2012-09-28 the elliptic curve specified by the domain parameters, to verify that the public key Qy is not the point at infinity 0, or to perform other verifications. As another example, the requester may verify that the private key dy corresponds to an integer in the appropriate range (which may be specified by the domain parameters), to verify that the private key dy is not zero, or perform other verifications. In some instances, the requester 302 may use the private key dy to generate a digital signature, to encrypt a message, or for another type of cryptographic operation. The requester 302 may explicitly compute the requester’s private key dy or the requester’s public key Qy or both, or the requester 302 may incorporate the equation for the public key Qy = ePy + ¹⁰ Qca ^or the equation for the private key dy = eky + r (modn) or both into another cryptographic operation, as appropriate. As such, the requester 302 may use the requester’s private key dy or the requester’s public key Qy or both without explicitly computing them.

At 320, the correspondent 306 uses the implicit certificate to generate the requester’s public key Qy. The correspondent 306 can generate the requester’s public key by computing the hash value e = F(Certy) and computing the public key value Qy = ePy + Q_ca. Here, the correspondent 306 generates the hash value e using the same function F that the certificate authority 304 used to generate the hash value e when generating the certificate data at 314.

The correspondent 306 may use the hash value e to generate the requester’s public key Qy for any suitable purpose. For example, the correspondent 306 may verify that the public key Qy corresponds to a point on the elliptic curve specified by the domain parameters, to verify that the public key Qy is not the point at infinity O, or to perform other verifications. In some instances, the correspondent 306 may use the public key Qy to verify a digital signature, to decrypt a message, or for another type of cryptographic operation. The correspondent 306 may explicitly compute the pubic key Qy, or the correspondent 306 may incorporate the equation for the public key Qy = ePy + Q_ca into another cryptographic operation, as appropriate. As such, the correspondent 306 may use the requester’s public key Qy or both without explicitly computing it.

At 322, the requester 302 and the correspondent 306 can carry out cryptographic correspondence. For example, the cryptographic correspondence may

CA 02792575 2012-09-28 involve data encryption, data authentication, or a combination of these and other operations. The cryptographic correspondence may be based on the requester’s elliptic curve key pair (dy, Qy) and possibly additional key data. For example, the correspondent 306 may encrypt a message using the requester’s public key Qy, and the requester 302 may decrypt the encrypted message using the requester’s private key dy. As another example, the requester 302 may sign a message based on the requester’s private key dy, and the correspondent 306 may authenticate the signed message using the requester’s public key Qy.

FIG. 4 is a signaling and flow diagram showing operations in an example 10 digital signature scheme 400. The example digital signature scheme 400 includes operations performed by a signer 402 and a verifier 406. The signer 402 and the verifier 406 can be, or they can be associated with, entities in a cryptography system. For example, the signer 402 can be the terminal 102 in FIG. 1, and the verifier 406 can be the terminal 106 in FIG. 1. The entities represented in FIG. 4 can be implemented by additional or different types of hardware, software, systems, devices, or combinations of these, as appropriate.

At a high level, the signer 402 generates a signature on a message and sends a signed version of the message to the verifier 406. The signature allows the verifier 406 to recover the message and verify its authenticity. The example operations and communications shown in FIG. 4 are described with respect to the ECPV signature scheme. The operations and communications shown in FIG. 4 can be adapted or modified for additional or different types of signature schemes. In the example shown in FIG. 4, the signer 402 and the verifier 406 are each in possession of elliptic curve domain parameters (e.g., T = (p,a,b,G,n,h)) and a random number generator, and they have agreed on a function F that produces a hash value, an encoding function E, a key derivation function (KDF), valid data types, and other parameters necessary to carry out the operations shown. In some implementations, the encoding function F is a symmetric encryption function, and the decoding function F^_1 is the corresponding decryption function. The key derivation function can be any suitable function that produces a key appropriate for use in the encoding function E. Some example key derivation functions are provided in NIST SP 800-108 (National Institute of Standards and Technology, Special Publication 800-108).

CA 02792575 2012-09-28

In an example shown in FIG. 4, the signer 402 has a long-term key pair (d_A, G_a) that includes the signer’s long-term private key d_A and the signer’s long-term public key G_A = d_AG. The signer 402 may obtain the long-term key pair (d_A,G_A) based on interactions with a certificate authority, or the signer 402 may generate the long-term key pair (d_A, G_A) independent of a certificate authority. The verifier 406 may receive the signer’s long-term public key G_A from the signer 402 or from another source. The verifier 406 may obtain the signer’s long-term public key G_A from a digital certificate such as, for example, an implicit certificate issued by a certificate authority. In some implementations, there is no digital certificate associated with the signer’s long-term public key G_A.

At 410, the signer 402 obtains a message M. The message may be, or may include, any suitable data or information to be communicated to the verifier 406. For example, the message can include any suitable type of electronic document, data file, data object, or other form of information. In some cases, the message is an e-mail message, an electronic document, or an electronic data file that can be edited and rendered by appropriate software applications. In some cases, the message is a data message or a combination of data messages used in signaling applications among hardware components.

In the example shown in FIG. 4, the message M = A| |F includes two message parts: a first message part N and a second message part V. In some instances, the signer 402 identifies the two message parts, for example, based on the size of the message, based on the content of the message, based on a parsing algorithm, or based on a combination of these and any other suitable criteria. In some instances, the message includes a different number of message parts.

At 412, the signer 402 obtains an encoded message part C. In the example shown in FIG. 4, the signer 402 obtains the encoded message part C by computing an elliptic curve point Q = zG, deriving a key K = KDF(Q) based on a key derivation function (KDF), and computing the encoded message part C = E(M,K) based on an encoding function E. The integer z and the corresponding elliptic curve point Q can be considered a short-term (or “ephemeral”) key pair (z, Q). The signer’s short-term key pair (z, Q) may be obtained by any suitable technique. In some implementations, z is a random number in the interval [1, n - 1] generated using a random number generator

CA 02792575 2012-09-28 at the signer 402, and the corresponding elliptic curve point Q is generated by combining the integer z with the base point generator G. The key K can be derived from the elliptic curve point Q using any suitable key derivation function. For example, the key K can be derived by applying a key derivation function to a bit string derived from an x-coordinate of the point Q, or the key K can be derived by any other suitable technique. The encoded message part C can be computed from the message M and the key K using any suitable encoding function. For example, the encoding function can be, or it can include, an encryption function or another type of encoding function.

At 414, the signer 402 obtains a hash value e. In the example shown in FIG. 4, the signer 402 obtains the hash value e by evaluating a function F(C\ | F) based on the encoded message part C concatenated with the second message part V. The hash value e can be generated using any appropriate hash function or hash-based function F, and the function F can be evaluated based on additional or different inputs, as appropriate.

In some implementations, the function F is the hash-based function 200 shown in FIG. 2.

At 416, the signer 402 generates a signature part s. In the example shown in FIG. 4, the signer 402 obtains the signature part s by computing s = z — d_Ae (mod ri). (This equation can be modified, provided that there is a corresponding change to the operations performed by the verifier. As an example, the signature part s can be obtained at 416 by computing s = z-F d_Ae (modri), and the operations performed by the verifier (e.g., to recover the message parts at 422) can be modified accordingly. Additional or different modifications may be made, as appropriate.) Here, z is the integer that the signer 402 used at 412 to produce the elliptic curve point Q, d_A is the signer’s private key value, e is the hash value, and n is an integer representing the order of the base point generator G. The signature part s may be generated by any suitable technique.

At 418, the signer 402 sends the signed message to the verifier 406. In the example shown in FIG. 4, sending the signed message to the verifier 406 includes sending the signature part s, the encoded message part C, and the second message part V to the verifier 406. The data may be transmitted in any suitable manner, using any suitable hardware. In some implementations, the signed message is transmitted over a

CA 02792575 2012-09-28 data network, over one or more wired or wireless links, etc. The signature part s, the encoded message part C, and the second message part V may be sent together or they may be sent separately, or they may be sent in any suitable combination. The data may be converted to an appropriate data format for transmission to the verifier 406. The verifier 406 receives the signed message from the signer 402.

At 420, the verifier 406 obtains the hash value e. In the example shown in FIG. 4, the verifier 406 generates the hash value e using the same function F(C 11F) that the signer 402 used to generate the hash value e, which was used in generating the signature part s at 416. The verifier 406 obtains the hash value e based on the encoded message part C and the second message part F, which were both received from the signer 402. The verifier 406 may obtain the hash value e based on additional or different information, as appropriate.

At 422, the verifier 406 recovers a message M'. In the example shown in FIG. 4, the message M^f is recovered by computing an elliptic curve point Q' = sG + eG_A, deriving a key K' = KDF(Q') based on the key derivation function KDF, and computing the message M' = E~¹(C, K') based on the decoding function F^-1. The key K' is derived by evaluating the key derivation function KDF on the elliptic curve point Q', and the message M' is computed by using the key K' to evaluate the decoding function F^_1 on the encoded message part C. Here, the verifier 406 uses the same key derivation function KDF that the signer 402 used to derive the key K; and the decoding function F^_1 used by the verifier 406 is the inverse of the encoding function E that was used by the signer 402 to calculate the encoded message part C. For example, if the signer 402 uses an encryption function to produce the encoded message part C, then the verifier 406 uses a corresponding decryption function to produce the message

M'.

At 424, the verifier 406 verifies the recovered message M¹. Verifying the recovered message M' may indicate that the digital signature is authentic. The recovered message M' can be verified by any suitable technique. In some implementations, if the decoding operation F^_1 returns an error message or an invalid value, then the recovered message M' is not verified, and the digital signature may be considered inauthentic. In some implementations, if the decoding operation F^_1returns value that is valid or is not an error message, then the recovered message M' is

CA 02792575 2012-09-28 verified, and the digital signature may be considered authentic. In some implementations, if the digital signature is authentic, then Q' = Q, K' = K, and

M' = M, and therefore the recovered message M' matches the sender’s initial message

M = N\\V.

FIG. 5 is a signaling and flow diagram showing operations in another example digital signature scheme 500. The example digital signature scheme 500 includes operations performed by a signer 502 and a verifier 506. The signer 502 and the verifier 506 can be, or they can be associated with, entities in a cryptography system. For example, the signer 502 can be the terminal 102 in FIG. 1, and the verifier 506 can be the terminal 106 in FIG. 1. The entities represented in FIG. 5 can be implemented by additional or different types of hardware, software, systems, devices, or combinations of these, as appropriate.

At a high level, the signer 502 generates a signature on a message and sends a signed version of the message to the verifier 506. The signature allows the verifier 406 to recover the message and verify its authenticity. The signature may also include an encrypted portion that provides confidentiality for at least a portion of the message. The example operations and communications shown in FIG. 5 are described with respect to the keyed variant of the ECPV signature scheme. The operations and communications shown in FIG. 5 can be adapted or modified for additional or different types of signature schemes. In the example shown in FIG. 5, the signer 502 and the verifier 506 are each in possession of elliptic curve domain parameters (e.g., T = (p, a, b, G, n, h)) and a random number generator, and they have agreed on a function F that produces a hash value, an encoding function E, a key derivation function (KDF), valid data types, and other parameters necessary to carry out the operations shown. In some implementations, the encoding function F is a symmetric encryption function, and the decoding function F^_1 is the corresponding decryption function. The key derivation function can be any suitable function that produces a key appropriate for use in the encoding function E.

In an example shown in FIG. 5, the signer 502 has a long-term key pair (d_A, G_A) that includes the signer’s long-term private key d_A and the signer’s long-term public key G_A = d_AG. The signer 502 may obtain the long-term key pair (d_A,G_A) based on interactions with a certificate authority, or the signer 502 may generate the

CA 02792575 2012-09-28 long-term key pair (d_A, G_A) independent of a certificate authority. The verifier 506 may receive the signer’s long-term public key G_A from the signer 502 or from another source. The verifier 506 may obtain the signer’s long-term public key G_A from a digital certificate such as, for example, an implicit certificate issued by a certificate authority.

In some implementations, there is no digital certificate associated with the signer’s long-term public key G_A.

In an example shown in FIG. 5, the verifier 506 has a long-term key pair (d_B,G_B) that includes the verifier’s long-term private key d_B and the verifier’s longterm public key G_B = d_BG. The verifier 506 may obtain the long-term key pair (d_B, G_b) based on interactions with a certificate authority, or the verifier 506 may generate the long-term key pair (d_B, G_b) independent of a certificate authority. The signer 502 may receive the verifier’s long-term public key G_B from the verifier 506 or from another source. The signer 502 may obtain the verifier’s long-term public key G_Bfrom a digital certificate such as, for example, an implicit certificate issued by a certificate authority. In some implementations, there is no digital certificate associated with the verifier’s long-term public key G_B.

At 510, the signer 502 obtains a message M. The message may be, or may include, any suitable data or information to be communicated to the verifier 506. For example, the message can include any suitable type of electronic document, data file, data object, or other form of information. In some cases, the message is an e-mail message, an electronic document, or an electronic data file that can be edited and rendered by appropriate software applications. In some cases, the message is a data message or a combination of data messages used in signaling applications among hardware components.

In the example shown in FIG. 5, the message M = includes three message parts: a first message part a second message part N₂, and a third message part V. In some instances, the signer 502 identifies the three message parts, for example, based on the size of the message, based on the content of the message, based on a parsing algorithm, or based on a combination of these and any other suitable criteria. In some instances, the message includes a different number of message parts. The signer 502 can incorporate a suitable degree of redundancy in the first message part N-l or the second message part N₂ or both. In some implementations, the

CA 02792575 2012-09-28 redundancy in the first message part and the redundancy in the second message part N₂ is sufficient, for example, for the verifier 506 to identify the redundancy in the corresponding message parts Nf and N₂', for example, when Nf = N_± and JV₂' = N₂.

At 512, the signer 502 obtains encoded messages parts Q and C₂. In the example shown in FIG. 5, the signer 502 obtains the encoded message parts and C₂by computing a first elliptic curve point Q = zG, deriving a first key K₂ = KDF(Q) based on a key derivation function (KDF), computing a second elliptic curve point Q_b = zG_b, deriving a second key K₂ = KDF(Qg) based on the key derivation function, and computing the encoded message parts Q = E(N_lt Kfl), C₂ — E(N₂,K₂) based on an encoding function E. The integer z and the corresponding elliptic curve point Q can be considered a short-term (or “ephemeral”) key pair (z, Q). The signer’s short-term key pair (z,Q) may be obtained by any suitable technique. In some implementations, z is a random number in the interval [1, n — 1] generated using a random number generator at the signer 402, and the corresponding elliptic curve point

Q is generated by combining the integer z with the base point generator G. The elliptic curve point Qb is generated based on the short-term private key combined with the verifier’s public key G_B. The keys and K₂ can be derived from the elliptic curve point Q using any suitable key derivation function. The encoded message parts C_± and C₂ can be computed from the message parts N_x and N₂ and the keys K_x and K₂ using any suitable encoding function. For example, the encoding function can be, or it can include, an encryption function or another type of encoding function.

At 514, the signer 502 obtains a hash value e. In the example shown in FIG. 5, the signer 502 obtains the hash value e by evaluating a function F(C₁||C₂||F) based on the first encoded message part Q concatenated with the second encoded message part

C₂ concatenated with the third message part V. The hash value e can be generated using any appropriate hash function or hash-based function F, and the function F can be evaluated based on additional or different inputs, as appropriate. In some implementations, the function F is the hash-based function 200 shown in FIG. 2.

At 516, the signer 502 generates a signature part s. In the example shown in

FIG. 5, the signer 502 obtains the signature part s by computing s = z + d_Ae (mod ri). (This equation can be modified, provided that there is a corresponding change to the operations performed by the verifier. As an example, the signature part s can be

CA 02792575 2012-09-28 obtained at 516 by computing s = z — d_Ae (modn), and the operations performed by the verifier (e.g., to recover the message parts at 522) can be modified accordingly. Additional or different modifications may be made, as appropriate.) Here, z is the integer that the signer 502 used at 512 to produce the elliptic curve points Q and Q_B, d_A is the signer’s private key value, e is the hash value, and n is an integer representing the order of the base point generator G. The signature part s may be generated by any suitable technique.

At 518, the signer 502 sends the signed message to the verifier 506. In the example shown in FIG. 5, sending the signed message to the verifier 506 includes sending the signature part s, the encoded message parts Q and C₂, and the third message part V to the verifier 506. The data may be transmitted in any suitable manner, using any suitable hardware. In some implementations, the signed message is transmitted over a data network, over one or more wired or wireless links, etc. The signature part s, the encoded message parts Q and C₂, and the third message part V may be sent together or they may be sent separately, or they may be sent in any suitable combination. The data may be converted to an appropriate data format for transmission to the verifier 506. The verifier 506 receives the signed message from the signer 502.

At 520, the verifier 506 obtains the hash value e. In the example shown in FIG.

5, the verifier 506 generates the hash value e using the same function F(C₁1|C₂11E) that the signer 502 used to generate the hash value e, which was used in generating the signature part s at 516. The verifier 506 obtains the hash value e based on the encoded message parts C_r and C₂ and the third message part E, which were all received from the signer 502. The verifier 506 may obtain the hash value e based on additional or different information, as appropriate.

At 522, the verifier 506 recovers two message parts Nf and N₂'. In the example shown in FIG. 5, the message parts Nf and N₂' are recovered by computing a first elliptic curve point Q’ = sG — eG_A, computing a second elliptic curve point Qf = d_B[sG - eG_A], deriving a first key K/ = KDF(Q') based on the key derivation function KDF, deriving a second key K₂ =KDF(Q/) based on the key derivation function KDF, computing the first message part Nf = based on the decoding function £^_1, and computing the second message part N2' = E~¹(C2,K₂')

CA 02792575 2012-09-28 based on the decoding function F^_1. The keys Kf and K2' are derived by evaluating the key derivation function KDF on the elliptic curve points Q' and Q'B respectively, and the message parts Nf and N2' are each computed by using the keys Kf and K2' to evaluate the decoding function F'¹ on the respective encoded message parts Cy and C2. Here, the verifier 506 uses the same key derivation function KDF that the signer 502 used to derive the keys Ky and K₂, and the decoding function F^-1 used by the verifier 506 is the inverse of the encoding function E that was used by the signer 502 to calculate the encoded message parts Q and C₂. For example, if the signer 502 uses an encryption function to produce the encoded message parts Cy and C₂, then the verifier 506 uses a corresponding decryption function to produce the message parts Nf and N₂'.

At 524, the verifier 506 verities the recovered message parts Ny' and N₂'. Verifying the recovered message parts Ny' and N₂' may indicate that the digital signature is authentic. The recovered message parts Nf and N₂' can be verified by any suitable technique. In some implementations, if Ny' is recognized as a valid plaintext, then the recovered message part Ny' is verified, and the digital signature may be considered authentic. The recovered message part Ny' may be recognized as a valid plaintext, for example, if it contains the expected redundant information. In some implementations, if N₂' is recognized as a valid plaintext, then the recovered message part N₂' is verified and confidential. The recovered message part N₂' may be recognized as a valid plaintext, for example, if decryption operation is successful, if it contains the expected redundant information, or both. In some implementations, Q' = Q, Qb' = Qb, = Ky, K₂' = K₂, Ny' = Ny, and N₂' = N₂, and therefore a recovered message M' = V/UAtyll!/ matches the sender’s initial message M =

Subject matter and operations described in this specification can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Some of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions, encoded on non-transitory computer storage medium for execution by, or to control the operation of, data

CA 02792575 2012-09-28 processing apparatus. Alternatively or in addition, the program instructions can be encoded for transmission to suitable receiver apparatus for execution by a data processing apparatus. A computer storage medium can be, or be included in, a computer-readable storage device, a computer-readable storage substrate, a random or serial access memory array or device, or a combination of one or more of them. Moreover, while a computer storage medium is not a propagated signal, a computer storage medium can be a source or destination of computer program instructions encoded in an artificially-generated propagated signal. The computer storage medium can also be, or be included in, one or more separate physical components or media (e.g,, multiple cards, disks, or other storage devices).

Operations described in this specification can be implemented as operations performed by a data processing apparatus on data stored on one or more computerreadable storage devices or received from other sources. The term “data processing apparatus” encompasses all kinds of apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, a system on a chip, or multiple ones, or combinations, of the foregoing. The apparatus can include special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit). The apparatus can also include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, a cross-platform runtime environment, a virtual machine, or a combination of one or more of them.

A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, declarative or procedural languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub-programs, or portions of code). A computer program can be deployed to

CA 02792575 2012-09-28 be executed on one computing device or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

Processes and logic flows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform actions by operating on input data and generating output data. Processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit).

Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computing device. Generally, a processor will receive instructions and data from a read-only memory or a random access memory or both. The essential elements of a computing device are a processor for performing actions in accordance with instructions and one or more memory devices for storing instructions and data. Generally, a computing device will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more storage devices for storing data. However, a computing device need not have such devices. Moreover, a computer can be embedded in another device, e.g., a mobile telephone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a Global Positioning System (GPS) receiver, or a portable storage device (e.g., a universal serial bus (USB) flash drive), to name just a few. Devices suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

To provide for interaction with a user, subject matter described in this specification can be implemented on a computer having a display device, e.g., an LCD (liquid crystal display) screen for displaying information to the user and a keyboard and a pointing device, e.g., touch screen, stylus, mouse, etc. by which the user can

CA 02792575 2012-09-28 provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input. In addition, a computing device can interact with a user by sending documents to and receiving documents from a device that is used by the user.

Some of the subject matter described in this specification can be implemented in a computing system that includes a back-end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front-end component, e.g., a client computing device having a graphical user interface, or any combination of one or more such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a data network.

The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a data network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. In some implementations, a server transmits data to a client device. Data generated at the client device can be received from the client device at the server.

While this specification contains many specific implementation details, these should not be construed as limitations on the scope of what may be claimed, but rather as descriptions of features specific to particular implementations. Certain features that are described in this specification in the context of separate implementations can also be implemented in combination in a single implementation. Conversely, various features that are described in the context of a single implementation can also be implemented in multiple implementations separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination.

CA 02792575 2012-09-28

Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of various system components in the implementations described above should not be understood as requiring such separation in all implementations, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

In a general aspect of what is described here, a hash value is generated. The hash value may be used, for example, in a cryptographic scheme.

In some aspects of what is described here, a first hash value is obtained by applying a first hash function to a first input. The first input is based on an implicit certificate associated with an entity. A second hash value is obtained by applying a second hash function to a second input. The second input is based on the first hash value. A public key associated with the entity or a private key associated with the entity (or both) is generated based on the second hash value. The public key or the private key (or both) can be used, for example, in cryptographic operations.

These and other aspects may include one or more of the following features.

The first input is obtained by concatenating a value with the implicit certificate. The first hash function is configured to produce a κ-bit output, and the concatenating includes concatenating a κ-bit constant (e.g., a κ-bit string of zeros) with the implicit certificate. The concatenating includes obtaining a first bit string based on a public key and a first public constant and concatenating the first bit string with the implicit certificate. The second input is obtained by obtaining a second bit string based on the key and a second public constant and concatenating the second bit string with the first hash value. The first hash function and the second hash function are the same hash function.

Additionally or alternatively, these and other aspects may include one or more of the following features. The first hash function and the second hash function are two different hash functions. The second input is the first hash value. The first input is the implicit certificate. The first hash function is configured to produce a κ-bit output, and

CA 02792575 2012-09-28 the second hash function is configured to produce a κ-bit output. The public and private keys are configured according to a κ-bit level of security.

Additionally or alternatively, these and other aspects may include one or more of the following features. The public key associated with the entity is an elliptic curve point Qy. Generating the public key associated with the entity includes extracting from the implicit certificate an elliptic curve point Py that represents a public key reconstruction value, accessing an elliptic curve point Q_CA that represents a public key of a certificate authority that issued the implicit certificate, and computing the elliptic curve point Qy = ePy + Q_CA, where e is the second hash value. The public key is used to encrypt a message to the entity. The public key is used to authenticate a message from the entity.

Additionally or alternatively, these and other aspects may include one or more of the following features. The private key associated with the entity is an integer dy. Generating the private key associated with the entity includes accessing an integer ky that was used to generate a certificate request Ry = kyG (where G represents a base point generator), accessing an integer r that represents private key contribution data provided by the certificate authority that issued the implicit certificate, and computing the integer dy = eky + r (mod n), where e is the second hash value. The private key is used to sign a message from the entity. The private key is used to decrypt a message to the entity.

Additionally or alternatively, these and other aspects may include one or more of the following features. The second hash value can be obtained based directly or indirectly on the first hash value. The public key or private key (or both) can be generated based directly or indirectly on the second hash value. In some instances, the public key or private key (or both) can be generated based indirectly on the second hash value by: obtaining a third hash value based on the second hash value, and generating the public key or private key (or both) based directly on the third hash value. The third hash value can be obtained by evaluating a third hash function, and the third hash function may be evaluated based on any suitable combination of inputs.

In some instances, the technique can be extended to include three, four or a larger number of hash functions applied in sequence.

CA 02792575 2012-09-28

Additionally or alternatively, these and other aspects may include one or more of the following features. The operations can be performed by any suitable entity in a cryptography system. The operations can be performed by data processing apparatus associated with the certificate authority that issued the implicit certificate, by data processing apparatus associated with the entity, or by data processing apparatus associated with a correspondent, other than the entity, that validates the public key (or uses the public key) based on the implicit certificate.

In some aspects of what is described here, a certificate authority generates an implicit certificate based on a request from an entity. A first hash value is obtained by applying a first hash function to a first input. The first input is based on the implicit certificate. A second hash value is obtained by applying a second hash function to a second input. The second input is based on the first hash value. The certificate authority server generates private key contribution data for the entity based on the second hash value. The certificate authority sends the private key contribution data and the implicit certificate to the entity.

These and other aspects may include one or more of the following features. The private key contribution data is an integer r. Generating the private key contribution data includes accessing an integer k that was used to generate a public key reconstruction value Py = Ry + kG (where Ry is an elliptic curve point provided in the certificate request from the entity, and F is a base point generator), accessing an integer d_CA that represents a private key of the certificate authority, and computing the integer r = ek + d_CA (mod n), where e is the second hash value.

In some aspects of what is described here, an entity generates a signature based on a message. A first hash value is obtained by applying a first hash function to a first input. The first input is based on the message to be signed. A second hash value is obtained by applying a second hash function to a second input. The second input is based on the first hash value. A signature is generated based on the second hash value and a private key of the entity. The signature is sent to a recipient.

In some aspects of what is described here, a message recipient verifies a digital signature. The signature and an associated message are received from an entity. A first hash value is obtained by applying a first hash function to a first input. The first input is based on the message. A second hash value is obtained by applying a second hash

CA 02792575 2012-09-28 function to a second input. The second input is based on the first hash value. The signature is verified based on the second hash value and a public key of the entity.

These and other aspects may include one or more of the following features. The first input is generated based on C||k, an encoded message part concatenated with a second message part. The first input is generated based on Cy 11C₂11V, a first encoded message part Cy concatenated with a second encoded message part C₂ concatenated with a third message part V. The signature is an integer 5 generated based on the second hash value e, the entity’s long-term private key d_A and the entity’s short-term private key z. The signature is verified based on identifying redundancy in a recovered message part.

Additionally or alternatively, these and other aspects may include one or more of the following features. The first input is obtained by concatenating a value with the message. The first hash function is configured to produce a κ-bit output, and the concatenating includes concatenating a κ-bit constant (e.g., κ-bit string of zeros or another κ-bit constant) with the message. The concatenating includes obtaining a first bit string based on a public key and a first public constant and concatenating the first bit string with the message. The first input can be the message itself.

Thus, particular implementations of the subject matter have been described. Other implementations are within the scope of the following claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In certain implementations, multitasking and parallel processing may be advantageous.