Network Working Group                                          D. McGrew
Internet-Draft                                       Cisco Systems, Inc.
Expires: January 2, 2007                                       July 2006


                        Authenticated Encryption
                      draft-mcgrew-auth-enc-00.txt

Status of this Memo

   By submitting this Internet-Draft, each author represents that any
   applicable patent or other IPR claims of which he or she is aware
   have been or will be disclosed, and any of which he or she becomes
   aware will be disclosed, in accordance with Section 6 of BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-
   Drafts.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   The list of current Internet-Drafts can be accessed at
   http://www.ietf.org/ietf/1id-abstracts.txt.

   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html.

   This Internet-Draft will expire on January 2, 2007.

Copyright Notice

   Copyright (C) The Internet Society (2006).

Abstract

   This draft defines algorithms for authenticated encryption with
   additional authenticated data (AEAD), and defines a uniform interface
   and a registry for such algorithms.  The interface and registry can
   be used as an application independent set of cryptoalgorithm suites.
   This approach provides advantages in efficiency and security, and
   promotes the reuse of crypto implementations.






McGrew                   Expires January 2, 2007                [Page 1]


Internet-Draft          Authenticated Encryption               July 2006


Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
     1.1   Conventions Used In This Document  . . . . . . . . . . . .  4
   2.  AEAD Interface . . . . . . . . . . . . . . . . . . . . . . . .  5
     2.1   Authenticated Encryption . . . . . . . . . . . . . . . . .  5
     2.2   Authenticated Decryption . . . . . . . . . . . . . . . . .  7
     2.3   Data Formatting  . . . . . . . . . . . . . . . . . . . . .  7
   3.  Requirements on AEAD algorithms  . . . . . . . . . . . . . . .  8
     3.1   Example IV Formation . . . . . . . . . . . . . . . . . . .  8
   4.  Requirements on the use of AEAD algorithms . . . . . . . . . .  9
   5.  AEAD Algorithms  . . . . . . . . . . . . . . . . . . . . . . . 10
     5.1   AEAD_AES_128_GCM . . . . . . . . . . . . . . . . . . . . . 10
       5.1.1   AEAD_AES_256_GCM . . . . . . . . . . . . . . . . . . . 10
     5.2   AEAD_AES_128_CCM . . . . . . . . . . . . . . . . . . . . . 10
       5.2.1   AEAD_AES_256_CCM . . . . . . . . . . . . . . . . . . . 10
     5.3   AEAD_AES_128_HMAC_SHA1 . . . . . . . . . . . . . . . . . . 11
       5.3.1   Test Cases . . . . . . . . . . . . . . . . . . . . . . 12
       5.3.2   AEAD_AES_256_HMAC_SHA1 . . . . . . . . . . . . . . . . 12
   6.  IANA Considerations  . . . . . . . . . . . . . . . . . . . . . 14
   7.  Example Usage  . . . . . . . . . . . . . . . . . . . . . . . . 15
   8.  Open Questions . . . . . . . . . . . . . . . . . . . . . . . . 17
   9.  Security Considerations  . . . . . . . . . . . . . . . . . . . 19
   10.   Acknowledgments  . . . . . . . . . . . . . . . . . . . . . . 20
   11.   References . . . . . . . . . . . . . . . . . . . . . . . . . 21
     11.1  Normative References . . . . . . . . . . . . . . . . . . . 21
     11.2  Informative References . . . . . . . . . . . . . . . . . . 21
       Author's Address . . . . . . . . . . . . . . . . . . . . . . . 22
       Intellectual Property and Copyright Statements . . . . . . . . 23






















McGrew                   Expires January 2, 2007                [Page 2]


Internet-Draft          Authenticated Encryption               July 2006


1.  Introduction

   Many cryptographic protocols provide both confidentiality and message
   authentication.  Often an encryption method and a message
   authentication code (MAC) are used to provide those security
   services, with each algorithm using an independent key.  More
   recently, the idea of providing both security services using a single
   cryptoalgorithm has become accepted.  In this concept, the cipher and
   MAC are replaced by an Authenticated Encryption with Associated Data
   (AEAD) algorithm.  Many crypto algorithms that implement AEAD
   algorithms have been defined, including block cipher modes of
   operation and dedicated algorithms.  Several of these algorithms have
   been adopted and proven useful in practice.  In this document we
   define an AEAD algorithm as an abstraction, by specifying an
   interface to an AEAD and defining an IANA registry for AEAD
   algorithms.  We populate this registry with six AEAD algorithms: AES
   in Galois/Counter Mode [GCM], AES in Counter and CBC MAC mode [CCM],
   and an algorithm that composes AES CBC and HMAC-SHA1, with key sizes
   of 128 and 256.  This approach enables applications that need
   cryptographic security services to more easily adopt those services.

   The approach benefits the application designer by allowing them to
   focus on the important issues of security services, canonicalization,
   and data marshaling, and relieving them of the need to design crypto
   mechanisms that meet their security goals.  Importantly, the security
   of an AEAD algorithm can be analyzed independent from its use in a
   particular application.  This property frees the user of the AEAD of
   the need to consider security aspects such as the relative order of
   authentication and encryption and the security of the particular
   combination of cipher and MAC, such as the potential loss of
   confidentiality through the MAC.  The application designer that uses
   the AEAD interface need not select a particular AEAD algorithm during
   the design stage.  Additionally, the interface to the AEAD is
   relatively simple, since it requires only a single key as input and
   it requires only a single identifier to indicate the algorithm in use
   in a particular case.

   The AEAD approach benefits the implementer of the crypto algorithms
   by making available optimizations that are otherwise not possible to
   reduce the amount of computation, the implementation cost, and/or the
   storage requirements.  The simpler interface makes testing easier;
   this is a considerable benefit for a crypto algorithm implementation.
   By providing a uniform interface to access cryptographic services,
   the AEAD approach allows a single crypto implementation to easily
   support multiple applications.  For example, a hardware module that
   supports the AEAD interface can easily provide crypto acceleration to
   any application using that interface, even to applications that had
   not been designed when the module was built.



McGrew                   Expires January 2, 2007                [Page 3]


Internet-Draft          Authenticated Encryption               July 2006


   The AEAD specification does not address security protocol issues such
   as anti-replay services or access control decisions that are made on
   authenticated data.  Instead, the specification aims to abstract the
   cryptography away from those issues.  The interface, and the guidance
   about how to use it, are consistent with the recommendations from
   [EEM04].

   In the following, we define the AEAD interface (Section 2), and then
   outline the requirements that each AEAD algorithm must meet
   (Section 3) and provide guidance on the use of AEAD algorithms
   (Section 4).  Then we define six AEAD algorithms (Section 5), and
   establish an IANA registry for AEAD algorithms (Section 6).  Lastly,
   we describe an example usage of the interface (Section 7) and discuss
   some open questions (Section 8).

1.1  Conventions Used In This Document

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in [RFC2119].































McGrew                   Expires January 2, 2007                [Page 4]


Internet-Draft          Authenticated Encryption               July 2006


2.  AEAD Interface

   An AEAD algorithm has two operations, authenticated encryption and
   authenticated decryption.  The inputs and outputs of these algorithms
   are defined in terms of bit strings.  However, an implementation MAY
   choose to accept only inputs whose bit-lengths are multiples of 8;
   that is, it may accept only octet strings.

   Each AEAD algorithm uses an Initialization Vector, or IV, to
   initialize the algorithm for processing a particular message.  The IV
   is generated by the authenticated encryption algorithm, and it is
   provided as an input to the authenticated decryption algorithm.  The
   IV is an output of the AEAD algorithm, rather than an input to it, so
   that it is under the control of the crypto module implementing that
   algorithm.

      Rationale.  Proper IV generation is a crucial for security, and
      the requirements on how IVs are generated are different for
      different algorithms.  Giving the AEAD algorithm the
      responsibility of generating its own IV hides that dependency from
      the user, better reflects the interface between a crypto module
      and an application, and promotes the testability of
      implementations.

   Each AEAD algorithm MUST specify the way in which its IVs are
   generated, to allow for external testing of the IV generation
   component of the encryption algorithm.

   An AEAD algorithm is called deterministic if the outputs of the
   encryption algorithm are a deterministic function of its inputs.
   Otherwise, the AEAD algorithm is called random.

   An AEAD algorithm MAY be stateful, that is, it may require that state
   is maintained between each invocation of the authenticated encryption
   operation, and/or between each invocation of the authenticated
   decryption operation.

2.1  Authenticated Encryption

   The authenticated encryption operation has three inputs, each of
   which is a bit string:

      a secret key K,

      a plaintext P, which is to be encrypted,

      additional authenticated data (AAD), which is denoted as A. This
      data is authenticated, but not encrypted.



McGrew                   Expires January 2, 2007                [Page 5]


Internet-Draft          Authenticated Encryption               July 2006


   A deterministic AEAD encryption algorithm MUST accept an additional
   input, and that value MUST be included verbatim in the IV.  This
   input is called the IV contribution.

   There are three outputs:

      an initialization vector IV,

      a ciphertext C,

      an authentication tag T.

   For each AEAD algorithm, the length of the key is fixed.  The lengths
   of the other inputs and outputs may vary, and may include zero.  The
   length of the ciphertext is equal to, or greater than, the length of
   the plaintext.  The length of the tag is denoted as t.

   Both confidentiality and message authentication is provided on the
   plaintext.  The strength of the authentication of P, IV and A is
   attenuated by the length t of the authentication tag.  When the
   length of P is zero, the AEAD algorithm acts as a message
   authentication code on the input A.

   The additional authenticated data A is used to protect information
   that needs to be authenticated, but which must be left unencrypted.
   When using an AEAD to secure a network protocol, this input could
   include addresses, ports, sequence numbers, protocol version numbers,
   and other fields that indicate how the plaintext should be handled,
   forwarded, or processed.  In many situations, it is desirable to
   authenticate these fields, though they must be left in the clear to
   allow the network or system to function properly.  When this data is
   included in the AAD, authentication is provided without copying the
   data into the ciphertext.

   The IV is authenticated internally to the algorithm, and it is not
   necessary to include it in the AAD input.  The IV MAY be included in
   the AAD if it convenient to the application.

   The IV MAY be transported along with the plaintext.  The entire IV
   need not be transmitted; it is sufficient to provide the receiver
   with enough information to allow the IV to be reconstructed.  Because
   the authenticated decryption process detects incorrect IV values, no
   security failure results when a receiver incorrectly reconstructs an
   IV.







McGrew                   Expires January 2, 2007                [Page 6]


Internet-Draft          Authenticated Encryption               July 2006


2.2  Authenticated Decryption

   The authenticated decryption operation has five inputs: K, IV , C ,
   A, and T, as defined above.  It has only a single output, either the
   plaintext value P or a special symbol FAIL that indicates that the
   inputs are not authentic.  A ciphertext C , initialization vector IV
   , additional authenticated data A and tag T are authentic for key K
   when IV, C, and T are generated by the encrypt operation with inputs
   K, A and P, for some plaintext P. The authenticated decrypt operation
   will, with high probability, return FAIL whenever its inputs were not
   created by the encrypt operation with the identical key (assuming
   that the AEAD algorithm is secure).

2.3  Data Formatting

   This document does not specify any particular encoding for the AEAD
   inputs and outputs, since the encoding does not affect the security
   services provided by an AEAD algorithm.

   An application using an AEAD SHOULD use the following rules for
   ordering the AEAD outputs (e.g. within a packet) , in order to
   facilitate efficient and simple implementations of AEAD algorithms:

      The Authentication Tag SHOULD follow all other fields.

      The Ciphertext SHOULD follow all fields other than the
      Authentication Tag.
























McGrew                   Expires January 2, 2007                [Page 7]


Internet-Draft          Authenticated Encryption               July 2006


3.  Requirements on AEAD algorithms

   As described above, a deterministic AEAD algorithm MUST accept an IV
   contribution input to its encryption operation.  This input MUST be
   included, verbatim, in the IV that is generated by that operation.  A
   deterministic algorithm MAY use the IV formation method defined in
   Section 3.1, or any other method that meets this requirement.

3.1  Example IV Formation

   One method to construct IVs that include a contribution, and which
   conforms to many different methods of IV construction, is as follows.
   The leftmost bits of the IV consist of the contribution.  For any
   fixed key, the length of the IV contribution is fixed.  The rightmost
   bits of the IV consist of an unsigned integer in network byte order.
   The length of the integer part is chosen so that the total length of
   the IV is the desired value.  The integer part of the IV is equal to
   one for the first IV, and increments by one for each successive IV
   that is generated.  The integer part is never equal to the all-zero
   value.

      Rationale.  This method is used by both GCM ESP [RFC4106] and CCM
      ESP [RFC4309], in which the IV contribution contains the Salt
      value and the lowest eight octets of the IV are explicitly carried
      in the ESP packet.  In GCM ESP, the IV contribution must be at
      least four octets long, so that it can contain the Salt value.  In
      CCM ESP, the IV contribution must be at least three octets long
      for the same reason.  This IV generation method is also used by
      several counter mode variants including CTR ESP.






















McGrew                   Expires January 2, 2007                [Page 8]


Internet-Draft          Authenticated Encryption               July 2006


4.  Requirements on the use of AEAD algorithms

   This section provides advice that must be followed in order to use an
   AEAD algorithm securely.

   If the AAD input is constructed out of multiple data elements, then
   it is essential that it be unambiguously parseable into its
   constituent elements, without the use of any unauthenticated data in
   the parsing process.  This requirement ensures that an attacker
   cannot fool a receiver into accepting a bogus set of data elements as
   authentic by altering a set of data elements that were used to
   construct an AAD input in an authenticated encryption operation in
   such a way that the data elements are different, but the AAD input is
   unchanged.  This requirement is trivially met if the AAD is composed
   of fixed-width elements.  If the AAD contains a variable-length
   string, for example, this requirement can be met by also including
   the length of the string in the AAD.

   Similarly, if the plaintext is constructed out of multiple data
   elements, then it is essential that it be unambiguously parseable
   into its constituent elements, without using any unauthenticated data
   in the parsing process.  Note that data that included in the AAD may
   be used when parsing the plaintext, though of course since the AAD is
   not encrypted there is a potential loss of confidentiality when
   information about the plaintext is included in the AAD.


























McGrew                   Expires January 2, 2007                [Page 9]


Internet-Draft          Authenticated Encryption               July 2006


5.  AEAD Algorithms

   This section defines six AEAD algorithms; two are based on AES GCM,
   two are based on AES CCM, and two are based on a composition of AES
   CBC and HMAC SHA1.  Each pair includes an algorithm with a key size
   of 128 bits and one with a key size of 256 bits.

5.1  AEAD_AES_128_GCM

   This algorithm is deterministic and stateful.  An IV contribution
   with a length of between zero and eight octets is accepted, and the
   IV is constructed as described in Section 3.1.  The IV is 12 octets
   in length, and the secret key is 128 bits long.

   The AEAD_AES_128_GCM authenticated encryption algorithm works as
   specified in [GCM], by providing the key, IV, and plaintext to that
   mode of operation.  Test cases are provided in that reference.

5.1.1  AEAD_AES_256_GCM

   This algorithm is identical to AEAD_AES_128_GCM, but with the
   following differences:

      the secret key is 256 bits long, instead of 128 bits in length,
      and

      AES-256 GCM is used instead of AES-128 GCM.


5.2  AEAD_AES_128_CCM

   This algorithm is deterministic and stateful.  An IV contribution
   with a length of between zero and seven octets is accepted, and the
   IV is constructed as described in Section 3.1.  The IV is eleven
   octets long, and the secret key is 128 bits in length.

   The AEAD_AES_128_CCM authenticated encryption algorithm works as
   specified in [CCM], by providing the key, IV, and plaintext to that
   mode of operation.  Test cases are provided in that reference.

5.2.1  AEAD_AES_256_CCM

   This algorithm is identical to AEAD_AES_128_CCM, but with the
   following differences:

      the secret key is 256 bits long, instead of 128 bits in length,
      and




McGrew                   Expires January 2, 2007               [Page 10]


Internet-Draft          Authenticated Encryption               July 2006


      AES-256 CCM is used instead of AES-128 CCM.


5.3  AEAD_AES_128_HMAC_SHA1

   This algorithm random and stateless.  It is based on the "generic
   composition" of CBC encryption with HMAC authentication, with the the
   encrypt-then-MAC method [AE].  It uses the HMAC message
   authentication code [RFC2104] with the SHA-1 hash function [SHA1] to
   provide message authentication.  Test cases for HMAC_SHA1 are
   provided in [RFC2202].  For encryption, it uses AES-128 in the cipher
   block chaining (CBC) mode of operation as defined in Section 6.2 of
   [MODES], with the padding method defined by Appendix A of the same
   reference.  The input key is 128 bits long, and the IV is generated
   uniformly at random, and is also 128 bits long.

   The authenticated encryption algorithm is as follows, or uses an
   equivalent set of steps:

   1.  Generate the secondary keys MAC_KEY and ENC_KEY from the input
       key K as follows:

          MAC_KEY = HMAC_SHA1(K, "MAC");

          ENC_KEY = leftmost(HMAC_SHA1(K, "ENC"), 128);

       where the function leftmost(X, m) accepts a bitstring X and a
       non-negative integer m and returns the bitstring containing the
       leftmost m bits of X. MAC_KEY is 160 bits long, and ENC_KEY is
       128 bits long.

   2.  Generate a 128-bit IV uniformly at random.  (A pseudorandom
       process MAY be used if its strength is equivalent to AES-128.)

   3.  Pad the plaintext by appending a single '1' bit to that data and
       then appending to the resulting string as few '0' bits as are
       necessary to make the number of bits in the plaintext into a
       multiple of 128.  Note that padding MUST be added to the data; if
       the number of octets in the payload data is a multiple of 16,
       then 16 octets of padding will be added.

   4.  Encrypt the payload using AES-128 in CBC mode, using the ENC_KEY
       and the random IV.  Form the ciphertext by prepending the IV to
       the CBC ciphertext outputs.

   5.  Compute the authentication tag by applying HMAC_SHA1 to the AAD,
       the IV, and the ciphertext, using the MAC_KEY.




McGrew                   Expires January 2, 2007               [Page 11]


Internet-Draft          Authenticated Encryption               July 2006


   6.  Return the ciphertext and the authentication tag.

   The authenticated decryption algorithm is as follows, or uses an
   equivalent set of steps:

   1.  Generate the secondary keys MAC_KEY and ENC_KEY from the input
       key K as follows:

          MAC_KEY = HMAC_SHA1(K, "MAC");

          ENC_KEY = leftmost(HMAC_SHA1(K, "ENC"), 128);

   2.  Compute the MAC value by applying HMAC_SHA1 to the clear data,
       the IV, and the ciphertext, using the MAC_KEY.  Compare this
       value to the authentication tag.  If they match, then continue
       with the processing.  Otherwise, discard the data and return
       FAIL.

   3.  Decrypt the payload using AES-128 in CBC mode, with the ENC_KEY,
       using the first 128 bits of the ciphertext as the random IV.

   4.  Remove padding by deleting the final '1' bit and all of the
       following '0' bits.  The remaining data forms the payload data.

   5.  Return the plaintext.

   The length of the ciphertext can be inferred from that of the
   plaintext.  The number L of octets in the ciphertext is given by

      L = 16 * ( floor(M / 16) + 2)

   where M denotes the number of octets in the payload, and the function
   floor() rounds its argument down to the nearest integer.  This fact
   is needed by the encoding function, since the length of the
   ciphertext, rather than the length of the payload, must be
   authenticated.

5.3.1  Test Cases

   A future version of this document will include test cases for
   AEAD_AES_128_HMAC_SHA1.

5.3.2  AEAD_AES_256_HMAC_SHA1

   This algorithm is identical to AEAD_AES_128_HMAC_SHA1, but with the
   following differences:





McGrew                   Expires January 2, 2007               [Page 12]


Internet-Draft          Authenticated Encryption               July 2006


      the secret key is 256 bits long, instead of 128 bits in length,

      AES-256 CBC is used instead of AES-128 CBC, and

      if a pseudorandom process is used to generate the IVs, that
      process must have a cryptographic strength equivalent to that of
      AES-256.












































McGrew                   Expires January 2, 2007               [Page 13]


Internet-Draft          Authenticated Encryption               July 2006


6.  IANA Considerations

   IANA will define the "AEAD Registry" described below.  Additions and
   changes to the AEAD Registry are by expert review.  Each entry in the
   registry contains the following elements:

      a short name, such as "AEAD_AES_128_GCM", that starts with the
      string "AEAD",

      a positive number, and

      a reference to a specification that completely defines an AEAD
      algorithm and provides test cases that can be used to verify the
      correctness of an implementation.

   Requests to add an entry to the registry MUST include the name and
   the reference.  The number is assigned by IANA.  These number
   assignments SHOULD use the smallest available positive number.

   IANA will add the following six entries to the AEAD Registry:

  +----------------------------+---------------+--------------------+
  | Name                       |   Reference   | Numeric Identifier |
  +----------------------------+---------------+--------------------+
  | AEAD_AES_128_GCM           |  Section 5.1  |          1         |
  |                            |               |                    |
  | AEAD_AES_256_GCM           | Section 5.1.1 |          2         |
  |                            |               |                    |
  | AEAD_AES_128_CCM           |  Section 5.2  |          3         |
  |                            |               |                    |
  | AEAD_AES_256_CCM           | Section 5.2.1 |          4         |
  |                            |               |                    |
  | AEAD_AES_128_CBC_HMAC_SHA1 |  Section 5.3  |          5         |
  |                            |               |                    |
  | AEAD_AES_256_CBC_HMAC_SHA1 | Section 5.3.2 |          6         |
  +----------------------------+---------------+--------------------+















McGrew                   Expires January 2, 2007               [Page 14]


Internet-Draft          Authenticated Encryption               July 2006


7.  Example Usage

   ESP support is easy to achieve, and can interoperate with existing
   ESP implementations of ESP GCM [RFC4106] and ESP CCM [RFC4309].  The
   ESP sender-side processing is illustrated in Figure 1, which shows
   how the fields of an ESP packet correspond to the authenticated
   encryption inputs and outputs.  Note that the only the rightmost
   eight octets of the IV are explicitly carried in the packet.  ESP
   receiver-side processing is shown in Figure 2.

   The conventional algorithms used in "generic composition" within ESP
   include HMAC-SHA1 and AES in CBC mode.  Unfortunately, the CBC
   padding method used in ESP is not suitable for generic use, since it
   interposes the padding and the pad length inside of the plaintext.
   Thus the AES_128_CBC_HMAC_SHA1 AEAD algorithm is very close to, but
   slightly different from, the equivalent generic composition method
   for ESP.

                                           K  IV Contribution
                                           |   |
   +-----------------------+ -+            v   v
   |          SPI          |  |          +-------+
   +-----------------------+  +--> AAD ->|       |
   |    Sequence Number    |  |          | a     |
   +-----------------------+ -+          | u e   |
   |          IV           | <---- IV ---| t n a |
   |                       |             | h c l |
   +-----------------------+ -+          | e r g |
   |                       |  |          | n y o |
   ~       ESP Payload     ~  |          | t p r |
   |                       |  +--> P --->| i t i |
   +-----------+-----+-----+  |          | c i t |
   | Padding   |  PL | NH  |  |          | a o m |
   +-----------+-----+-----+ -+          | t n   |
   |       Integrity       |             | e     |
   ~         Check         ~ <---- T ----| d     |
   |         Value         |             +-------+
   +-----------------------+

                 Figure 1: ESP AEAD encryption processing.











McGrew                   Expires January 2, 2007               [Page 15]


Internet-Draft          Authenticated Encryption               July 2006


                                             K
                                             |
   +-----------------------+ -+              V
   |          SPI          |  |          +-------+
   +-----------------------+  +--> AAD ->|       |
   |    Sequence Number    |  |          | a     |
   +-----------------------+ -+          | u d   |
   |          IV           | ----- IV -->| t e a |
   |                       |             | h c l |
   +-----------------------+             | e r g |
   |                       | <-+         | n y o |
   ~       ESP Payload     ~   |         | t p r |
   |                       |   +-- P ----| i t i |
   +-----------+-----+-----+   |         | c i t |
   | Padding   |  PL | NH  | <-+         | a o m |
   +-----------+-----+-----+             | t n   |
   |       Integrity       |             | e     |
   ~         Check         ~ ----- T --->| d     |
   |         Value         |             +-------+
   +-----------------------+

                 Figure 2: ESP AEAD decryption processing.





























McGrew                   Expires January 2, 2007               [Page 16]


Internet-Draft          Authenticated Encryption               July 2006


8.  Open Questions

   The additional authenticated data input is well suited to
   authenticating headers.  Some cryptographic protocols have trailers
   that should be authenticated.  For example, in the Secure RTP
   protocol the authenticated data consists of the RTP header, the
   ciphertext containing the encrypted payload, and some additional
   data, in that order.  It is impossible for an AEAD to accommodate
   both the authenticated header and authenticated trailer without
   adding an additional input for the trailer.  Because none of the
   specified AEAD algorithms can handle both a trailer and a footer,
   this specification does not include a trailer in its interface.  We
   expect that protocols like SRTP will need to define different
   processing rules that include all of the authenticated-only data into
   a single AAD input in order to make use of this specification.
   However, new rules would need to be defined in order to use either
   GCM or CCM or any other AEAD transforms, so this need is not
   especially onerous.

   The TLS protocol as currently defined assumes that authentication
   will precede encryption.  Thus, in order to accommodate this
   specification, new processing rules would need to be written that
   make no assumptions about the relative ordering of the cryptographic
   services.  However, as above, these new rules would need to be
   defined anyway in order to use any AEAD algorithm.

   The AEAD algorithms selected reflect those that have been already
   adopted by standards.  Both random and deterministic algorithms have
   been provided.  It is an open question as to what other AEAD
   algorithms should be added.  Many variations on basic algorithms are
   possible, each with its own advantages.  While it is desirable to
   admit any algorithms that are found to be useful in practice, it is
   also desirable to limit the total number of registered algorithms.
   The current specification requires that a registered algorithm
   provide a complete specification and a set of validation data; it is
   hoped that these prerequisites set the admission criteria
   appropriately.

   Some users may view an IANA assignment as a recommendation or an
   endorsement of a particular AEAD algorithm.  Other users may desire
   to register an AEAD algorithm in order to allow for experimental or
   specialized use.  Because of these conflicting perspectives, it may
   be desirable to allocate a second IANA registry for experimental use.

   It may be desirable to replace HMAC-SHA1 with AES CMAC [CMAC] in the
   generic composition algorithm, or to introduce an additional
   algorithm that does so.




McGrew                   Expires January 2, 2007               [Page 17]


Internet-Draft          Authenticated Encryption               July 2006


   Directly testing a randomized AEAD encryption algorithm using a test
   cases with fixed inputs and outputs is not possible, since the
   encryption process is non-deterministic.  However, it is easy to test
   a randomized AEAD algorithm against fixed test cases.  The
   authenticated decryption algorithm is deterministic, and it can be
   directly tested.  The authenticated encryption algorithm can be
   tested by encrypting a plaintext, decrypting the resulting
   ciphertext, and comparing the original plaintext to the post-
   decryption plaintext.

   This specification is incomplete regarding the subject of
   authentication tag lengths.  A future version will need to provide
   clarification.  One alternative is to have each AEAD algorithm use a
   fixed tag length, though if the specification is inflexible, we are
   faced with the need to choose that length carefully.  If backwards
   compatibility is desirable, a length of 12 octets would be best; if
   security is considered paramount, then the longest is the best.  One
   way to add flexibility would be to have each AEAD return a fixed-
   length authentication tag, but then to define a standard mechanism by
   which that tag can be truncated when it is desirable to do so.

   The authentication tag could be included in the ciphertext, which
   would simplify the interface slightly, but would remove some
   flexibility.  Most existing security protocols include a separate
   integrity check field to carry a message authentication code, so a
   separate output for that tag was included in the AEAD interface.

   Some of the terminology in this specification is unwieldy, and could
   perhaps be improved.  For example, "AEAD algorithm" could be replaced
   with "crypto transform", which would be meaningful to a broader
   community, but is less precise.




















McGrew                   Expires January 2, 2007               [Page 18]


Internet-Draft          Authenticated Encryption               July 2006


9.  Security Considerations

   A future version of this document will define the security services
   that must be provided by an AEAD algorithm.















































McGrew                   Expires January 2, 2007               [Page 19]


Internet-Draft          Authenticated Encryption               July 2006


10.  Acknowledgments

   Eric Rescorla and Yoshi Kohno provided valuable comments on early
   versions of this document.















































McGrew                   Expires January 2, 2007               [Page 20]


Internet-Draft          Authenticated Encryption               July 2006


11.  References

11.1  Normative References

   [CCM]      "NIST Special Publication 800-38C: The CCM Mode for
              Authentication and Confidentiality",
               http://csrc.nist.gov/publications/nistpubs/SP800-38C.pdf.

   [GCM]      McGrew, D. and J. Viega, "The Galois/Counter Mode of
              Operation (GCM)", Submission to NIST. http://
              csrc.nist.gov/CryptoToolkit/modes/proposedmodes/gcm/
              gcm-spec.pdf, January 2004.

   [MODES]    "NIST Special Publication 800-38", Reccomendation for
              Block Cipher Modes of Operation http://csrc.nist.gov/
              publications/nistpubs/800-38a/sp800-38a.pdf.

   [RFC2104]  Krawczyk, H., Bellare, M., and R. Canetti, "HMAC: Keyed-
              Hashing for Message Authentication", RFC 2104,
              February 1997.

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119, March 1997.

   [RFC2202]  Cheng, P. and R. Glenn, "Test Cases for HMAC-MD5 and HMAC-
              SHA-1", RFC 2202, September 1997.

   [RFC4106]  Viega, J. and D. McGrew, "The Use of Galois/Counter Mode
              (GCM) in IPsec Encapsulating Security Payload (ESP)",
              RFC 4106, June 2005.

   [RFC4309]  Housley, R., "Using Advanced Encryption Standard (AES) CCM
              Mode with IPsec Encapsulating Security Payload (ESP)",
              RFC 4309, December 2005.

   [SHA1]     "FIPS 180-1: Secure Hash Standard,", Federal Information
              Processing Standard
              (FIPS) http://www.itl.nist.gov/fipspubs/fip180-1.htm.

11.2  Informative References

   [AE]     "Authenticated encryption: Relations among notions and
            analysis of the generic composition paradigm", Proceedings
            of ASIACRYPT 2000, Springer-Verlag, LNCS 1976, pp.
            531-545 http://www.

   [CMAC]   "NIST Special Publication 800-38B",  http://csrc.nist.gov/
            CryptoToolkit/modes/800-38_Series_Publications/



McGrew                   Expires January 2, 2007               [Page 21]


Internet-Draft          Authenticated Encryption               July 2006


            SP800-38B.pdf.

   [EEM04]  "Breaking and provably repairing the SSH authenticated
            encryption scheme: A case study of the Encode-then-Encrypt-
            and-MAC paradigm", ACM Transactions on Information and
            System Security, http://www-cse.ucsd.edu/users/tkohno/
            papers/TISSEC04/.


Author's Address

   David A. McGrew
   Cisco Systems, Inc.
   510 McCarthy Blvd.
   Milpitas, CA  95035
   US

   Phone: (408) 525 8651
   Email: mcgrew@cisco.com
   URI:   http://www.mindspring.com/~dmcgrew/dam.htm































McGrew                   Expires January 2, 2007               [Page 22]


Internet-Draft          Authenticated Encryption               July 2006


Intellectual Property Statement

   The IETF takes no position regarding the validity or scope of any
   Intellectual Property Rights or other rights that might be claimed to
   pertain to the implementation or use of the technology described in
   this document or the extent to which any license under such rights
   might or might not be available; nor does it represent that it has
   made any independent effort to identify any such rights.  Information
   on the procedures with respect to rights in RFC documents can be
   found in BCP 78 and BCP 79.

   Copies of IPR disclosures made to the IETF Secretariat and any
   assurances of licenses to be made available, or the result of an
   attempt made to obtain a general license or permission for the use of
   such proprietary rights by implementers or users of this
   specification can be obtained from the IETF on-line IPR repository at
   http://www.ietf.org/ipr.

   The IETF invites any interested party to bring to its attention any
   copyrights, patents or patent applications, or other proprietary
   rights that may cover technology that may be required to implement
   this standard.  Please address the information to the IETF at
   ietf-ipr@ietf.org.


Disclaimer of Validity

   This document and the information contained herein are provided on an
   "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
   OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
   ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED,
   INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
   INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
   WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.


Copyright Statement

   Copyright (C) The Internet Society (2006).  This document is subject
   to the rights, licenses and restrictions contained in BCP 78, and
   except as set forth therein, the authors retain all their rights.


Acknowledgment

   Funding for the RFC Editor function is currently provided by the
   Internet Society.




McGrew                   Expires January 2, 2007               [Page 23]