码迷,mamicode.com
首页 > 系统相关 > 详细

HMAC

时间:2016-09-01 15:55:42      阅读:296      评论:0      收藏:0      [点我收藏+]

标签:

介绍文档及C实现

http://www.ietf.org/rfc/rfc2104.txt

 

   
   
   
   
   
   
  Network Working Group H. Krawczyk
  Request for Comments: 2104 IBM
  Category: Informational M. Bellare
  UCSD
  R. Canetti
  IBM
  February 1997
   
   
  HMAC: Keyed-Hashing for Message Authentication
   
  Status of This Memo
   
  This memo provides information for the Internet community. This memo
  does not specify an Internet standard of any kind. Distribution of
  this memo is unlimited.
   
  Abstract
   
  This document describes HMAC, a mechanism for message authentication
  using cryptographic hash functions. HMAC can be used with any
  iterative cryptographic hash function, e.g., MD5, SHA-1, in
  combination with a secret shared key. The cryptographic strength of
  HMAC depends on the properties of the underlying hash function.
   
  1. Introduction
   
  Providing a way to check the integrity of information transmitted
  over or stored in an unreliable medium is a prime necessity in the
  world of open computing and communications. Mechanisms that provide
  such integrity check based on a secret key are usually called
  "message authentication codes" (MAC). Typically, message
  authentication codes are used between two parties that share a secret
  key in order to validate information transmitted between these
  parties. In this document we present such a MAC mechanism based on
  cryptographic hash functions. This mechanism, called HMAC, is based
  on work by the authors [BCK1] where the construction is presented and
  cryptographically analyzed. We refer to that work for the details on
  the rationale and security analysis of HMAC, and its comparison to
  other keyed-hash methods.
   
   
   
   
   
   
   
   
   
   
   
  Krawczyk, et. al. Informational [Page 1]
 
  RFC 2104 HMAC February 1997
   
   
  HMAC can be used in combination with any iterated cryptographic hash
  function. MD5 and SHA-1 are examples of such hash functions. HMAC
  also uses a secret key for calculation and verification of the
  message authentication values. The main goals behind this
  construction are
   
  * To use, without modifications, available hash functions.
  In particular, hash functions that perform well in software,
  and for which code is freely and widely available.
   
  * To preserve the original performance of the hash function without
  incurring a significant degradation.
   
  * To use and handle keys in a simple way.
   
  * To have a well understood cryptographic analysis of the strength of
  the authentication mechanism based on reasonable assumptions on the
  underlying hash function.
   
  * To allow for easy replaceability of the underlying hash function in
  case that faster or more secure hash functions are found or
  required.
   
  This document specifies HMAC using a generic cryptographic hash
  function (denoted by H). Specific instantiations of HMAC need to
  define a particular hash function. Current candidates for such hash
  functions include SHA-1 [SHA], MD5 [MD5], RIPEMD-128/160 [RIPEMD].
  These different realizations of HMAC will be denoted by HMAC-SHA1,
  HMAC-MD5, HMAC-RIPEMD, etc.
   
  Note: To the date of writing of this document MD5 and SHA-1 are the
  most widely used cryptographic hash functions. MD5 has been recently
  shown to be vulnerable to collision search attacks [Dobb]. This
  attack and other currently known weaknesses of MD5 do not compromise
  the use of MD5 within HMAC as specified in this document (see
  [Dobb]); however, SHA-1 appears to be a cryptographically stronger
  function. To this date, MD5 can be considered for use in HMAC for
  applications where the superior performance of MD5 is critical. In
  any case, implementers and users need to be aware of possible
  cryptanalytic developments regarding any of these cryptographic hash
  functions, and the eventual need to replace the underlying hash
  function. (See section 6 for more information on the security of
  HMAC.)
   
   
   
   
   
   
   
   
  Krawczyk, et. al. Informational [Page 2]
 
  RFC 2104 HMAC February 1997
   
   
  2. Definition of HMAC
   
  The definition of HMAC requires a cryptographic hash function, which
  we denote by H, and a secret key K. We assume H to be a cryptographic
  hash function where data is hashed by iterating a basic compression
  function on blocks of data. We denote by B the byte-length of such
  blocks (B=64 for all the above mentioned examples of hash functions),
  and by L the byte-length of hash outputs (L=16 for MD5, L=20 for
  SHA-1). The authentication key K can be of any length up to B, the
  block length of the hash function. Applications that use keys longer
  than B bytes will first hash the key using H and then use the
  resultant L byte string as the actual key to HMAC. In any case the
  minimal recommended length for K is L bytes (as the hash output
  length). See section 3 for more information on keys.
   
  We define two fixed and different strings ipad and opad as follows
  (the ‘i‘ and ‘o‘ are mnemonics for inner and outer):
   
  ipad = the byte 0x36 repeated B times
  opad = the byte 0x5C repeated B times.
   
  To compute HMAC over the data `text‘ we perform
   
  H(K XOR opad, H(K XOR ipad, text))
   
  Namely,
   
  (1) append zeros to the end of K to create a B byte string
  (e.g., if K is of length 20 bytes and B=64, then K will be
  appended with 44 zero bytes 0x00)
  (2) XOR (bitwise exclusive-OR) the B byte string computed in step
  (1) with ipad
  (3) append the stream of data ‘text‘ to the B byte string resulting
  from step (2)
  (4) apply H to the stream generated in step (3)
  (5) XOR (bitwise exclusive-OR) the B byte string computed in
  step (1) with opad
  (6) append the H result from step (4) to the B byte string
  resulting from step (5)
  (7) apply H to the stream generated in step (6) and output
  the result
   
  For illustration purposes, sample code based on MD5 is provided as an
  appendix.
   
   
   
   
   
   
   
  Krawczyk, et. al. Informational [Page 3]
 
  RFC 2104 HMAC February 1997
   
   
  3. Keys
   
  The key for HMAC can be of any length (keys longer than B bytes are
  first hashed using H). However, less than L bytes is strongly
  discouraged as it would decrease the security strength of the
  function. Keys longer than L bytes are acceptable but the extra
  length would not significantly increase the function strength. (A
  longer key may be advisable if the randomness of the key is
  considered weak.)
   
  Keys need to be chosen at random (or using a cryptographically strong
  pseudo-random generator seeded with a random seed), and periodically
  refreshed. (Current attacks do not indicate a specific recommended
  frequency for key changes as these attacks are practically
  infeasible. However, periodic key refreshment is a fundamental
  security practice that helps against potential weaknesses of the
  function and keys, and limits the damage of an exposed key.)
   
  4. Implementation Note
   
  HMAC is defined in such a way that the underlying hash function H can
  be used with no modification to its code. In particular, it uses the
  function H with the pre-defined initial value IV (a fixed value
  specified by each iterative hash function to initialize its
  compression function). However, if desired, a performance
  improvement can be achieved at the cost of (possibly) modifying the
  code of H to support variable IVs.
   
  The idea is that the intermediate results of the compression function
  on the B-byte blocks (K XOR ipad) and (K XOR opad) can be precomputed
  only once at the time of generation of the key K, or before its first
  use. These intermediate results are stored and then used to
  initialize the IV of H each time that a message needs to be
  authenticated. This method saves, for each authenticated message,
  the application of the compression function of H on two B-byte blocks
  (i.e., on (K XOR ipad) and (K XOR opad)). Such a savings may be
  significant when authenticating short streams of data. We stress
  that the stored intermediate values need to be treated and protected
  the same as secret keys.
   
  Choosing to implement HMAC in the above way is a decision of the
  local implementation and has no effect on inter-operability.
   
   
   
   
   
   
   
   
   
  Krawczyk, et. al. Informational [Page 4]
 
  RFC 2104 HMAC February 1997
   
   
  5. Truncated output
   
  A well-known practice with message authentication codes is to
  truncate the output of the MAC and output only part of the bits
  (e.g., [MM, ANSI]). Preneel and van Oorschot [PV] show some
  analytical advantages of truncating the output of hash-based MAC
  functions. The results in this area are not absolute as for the
  overall security advantages of truncation. It has advantages (less
  information on the hash result available to an attacker) and
  disadvantages (less bits to predict for the attacker). Applications
  of HMAC can choose to truncate the output of HMAC by outputting the t
  leftmost bits of the HMAC computation for some parameter t (namely,
  the computation is carried in the normal way as defined in section 2
  above but the end result is truncated to t bits). We recommend that
  the output length t be not less than half the length of the hash
  output (to match the birthday attack bound) and not less than 80 bits
  (a suitable lower bound on the number of bits that need to be
  predicted by an attacker). We propose denoting a realization of HMAC
  that uses a hash function H with t bits of output as HMAC-H-t. For
  example, HMAC-SHA1-80 denotes HMAC computed using the SHA-1 function
  and with the output truncated to 80 bits. (If the parameter t is not
  specified, e.g. HMAC-MD5, then it is assumed that all the bits of the
  hash are output.)
   
  6. Security
   
  The security of the message authentication mechanism presented here
  depends on cryptographic properties of the hash function H: the
  resistance to collision finding (limited to the case where the
  initial value is secret and random, and where the output of the
  function is not explicitly available to the attacker), and the
  message authentication property of the compression function of H when
  applied to single blocks (in HMAC these blocks are partially unknown
  to an attacker as they contain the result of the inner H computation
  and, in particular, cannot be fully chosen by the attacker).
   
  These properties, and actually stronger ones, are commonly assumed
  for hash functions of the kind used with HMAC. In particular, a hash
  function for which the above properties do not hold would become
  unsuitable for most (probably, all) cryptographic applications,
  including alternative message authentication schemes based on such
  functions. (For a complete analysis and rationale of the HMAC
  function the reader is referred to [BCK1].)
   
   
   
   
   
   
   
   
  Krawczyk, et. al. Informational [Page 5]
 
  RFC 2104 HMAC February 1997
   
   
  Given the limited confidence gained so far as for the cryptographic
  strength of candidate hash functions, it is important to observe the
  following two properties of the HMAC construction and its secure use
  for message authentication:
   
  1. The construction is independent of the details of the particular
  hash function H in use and then the latter can be replaced by any
  other secure (iterative) cryptographic hash function.
   
  2. Message authentication, as opposed to encryption, has a
  "transient" effect. A published breaking of a message authentication
  scheme would lead to the replacement of that scheme, but would have
  no adversarial effect on information authenticated in the past. This
  is in sharp contrast with encryption, where information encrypted
  today may suffer from exposure in the future if, and when, the
  encryption algorithm is broken.
   
  The strongest attack known against HMAC is based on the frequency of
  collisions for the hash function H ("birthday attack") [PV,BCK2], and
  is totally impractical for minimally reasonable hash functions.
   
  As an example, if we consider a hash function like MD5 where the
  output length equals L=16 bytes (128 bits) the attacker needs to
  acquire the correct message authentication tags computed (with the
  _same_ secret key K!) on about 2**64 known plaintexts. This would
  require the processing of at least 2**64 blocks under H, an
  impossible task in any realistic scenario (for a block length of 64
  bytes this would take 250,000 years in a continuous 1Gbps link, and
  without changing the secret key K during all this time). This attack
  could become realistic only if serious flaws in the collision
  behavior of the function H are discovered (e.g. collisions found
  after 2**30 messages). Such a discovery would determine the immediate
  replacement of the function H (the effects of such failure would be
  far more severe for the traditional uses of H in the context of
  digital signatures, public key certificates, etc.).
   
  Note: this attack needs to be strongly contrasted with regular
  collision attacks on cryptographic hash functions where no secret key
  is involved and where 2**64 off-line parallelizable (!) operations
  suffice to find collisions. The latter attack is approaching
  feasibility [VW] while the birthday attack on HMAC is totally
  impractical. (In the above examples, if one uses a hash function
  with, say, 160 bit of output then 2**64 should be replaced by 2**80.)
   
   
   
   
   
   
   
   
  Krawczyk, et. al. Informational [Page 6]
 
  RFC 2104 HMAC February 1997
   
   
  A correct implementation of the above construction, the choice of
  random (or cryptographically pseudorandom) keys, a secure key
  exchange mechanism, frequent key refreshments, and good secrecy
  protection of keys are all essential ingredients for the security of
  the integrity verification mechanism provided by HMAC.
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
  Krawczyk, et. al. Informational [Page 7]
 
  RFC 2104 HMAC February 1997
   
   
  Appendix -- Sample Code
   
  For the sake of illustration we provide the following sample code for
  the implementation of HMAC-MD5 as well as some corresponding test
  vectors (the code is based on MD5 code as described in [MD5]).
   
  /*
  ** Function: hmac_md5
  */
   
  void
  hmac_md5(text, text_len, key, key_len, digest)
  unsigned char* text; /* pointer to data stream */
  int text_len; /* length of data stream */
  unsigned char* key; /* pointer to authentication key */
  int key_len; /* length of authentication key */
  caddr_t digest; /* caller digest to be filled in */
   
  {
  MD5_CTX context;
  unsigned char k_ipad[65]; /* inner padding -
  * key XORd with ipad
  */
  unsigned char k_opad[65]; /* outer padding -
  * key XORd with opad
  */
  unsigned char tk[16];
  int i;
  /* if key is longer than 64 bytes reset it to key=MD5(key) */
  if (key_len > 64) {
   
  MD5_CTX tctx;
   
  MD5Init(&tctx);
  MD5Update(&tctx, key, key_len);
  MD5Final(tk, &tctx);
   
  key = tk;
  key_len = 16;
  }
   
  /*
  * the HMAC_MD5 transform looks like:
  *
  * MD5(K XOR opad, MD5(K XOR ipad, text))
  *
  * where K is an n byte key
  * ipad is the byte 0x36 repeated 64 times
   
   
   
  Krawczyk, et. al. Informational [Page 8]
 
  RFC 2104 HMAC February 1997
   
   
  * opad is the byte 0x5c repeated 64 times
  * and text is the data being protected
  */
   
  /* start out by storing key in pads */
  bzero( k_ipad, sizeof k_ipad);
  bzero( k_opad, sizeof k_opad);
  bcopy( key, k_ipad, key_len);
  bcopy( key, k_opad, key_len);
   
  /* XOR key with ipad and opad values */
  for (i=0; i<64; i++) {
  k_ipad[i] ^= 0x36;
  k_opad[i] ^= 0x5c;
  }
  /*
  * perform inner MD5
  */
  MD5Init(&context); /* init context for 1st
  * pass */
  MD5Update(&context, k_ipad, 64) /* start with inner pad */
  MD5Update(&context, text, text_len); /* then text of datagram */
  MD5Final(digest, &context); /* finish up 1st pass */
  /*
  * perform outer MD5
  */
  MD5Init(&context); /* init context for 2nd
  * pass */
  MD5Update(&context, k_opad, 64); /* start with outer pad */
  MD5Update(&context, digest, 16); /* then results of 1st
  * hash */
  MD5Final(digest, &context); /* finish up 2nd pass */
  }
   
  Test Vectors (Trailing ‘\0‘ of a character string not included in test):
   
  key = 0x0b0b0b0b0b0b0b0b0b0b0b0b0b0b0b0b
  key_len = 16 bytes
  data = "Hi There"
  data_len = 8 bytes
  digest = 0x9294727a3638bb1c13f48ef8158bfc9d
   
  key = "Jefe"
  data = "what do ya want for nothing?"
  data_len = 28 bytes
  digest = 0x750c783e6ab0b503eaa86e310a5db738
   
  key = 0xAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
   
   
   
  Krawczyk, et. al. Informational [Page 9]
 
  RFC 2104 HMAC February 1997
   
   
  key_len 16 bytes
  data = 0xDDDDDDDDDDDDDDDDDDDD...
  ..DDDDDDDDDDDDDDDDDDDD...
  ..DDDDDDDDDDDDDDDDDDDD...
  ..DDDDDDDDDDDDDDDDDDDD...
  ..DDDDDDDDDDDDDDDDDDDD
  data_len = 50 bytes
  digest = 0x56be34521d144c88dbb8c733f0e8b3f6
   
  Acknowledgments
   
  Pau-Chen Cheng, Jeff Kraemer, and Michael Oehler, have provided
  useful comments on early drafts, and ran the first interoperability
  tests of this specification. Jeff and Pau-Chen kindly provided the
  sample code and test vectors that appear in the appendix. Burt
  Kaliski, Bart Preneel, Matt Robshaw, Adi Shamir, and Paul van
  Oorschot have provided useful comments and suggestions during the
  investigation of the HMAC construction.
   
  References
   
  [ANSI] ANSI X9.9, "American National Standard for Financial
  Institution Message Authentication (Wholesale)," American
  Bankers Association, 1981. Revised 1986.
   
  [Atk] Atkinson, R., "IP Authentication Header", RFC 1826, August
  1995.
   
  [BCK1] M. Bellare, R. Canetti, and H. Krawczyk,
  "Keyed Hash Functions and Message Authentication",
  Proceedings of Crypto‘96, LNCS 1109, pp. 1-15.
  (http://www.research.ibm.com/security/keyed-md5.html)
   
  [BCK2] M. Bellare, R. Canetti, and H. Krawczyk,
  "Pseudorandom Functions Revisited: The Cascade Construction",
  Proceedings of FOCS‘96.
   
  [Dobb] H. Dobbertin, "The Status of MD5 After a Recent Attack",
  RSA Labs‘ CryptoBytes, Vol. 2 No. 2, Summer 1996.
  http://www.rsa.com/rsalabs/pubs/cryptobytes.html
   
  [PV] B. Preneel and P. van Oorschot, "Building fast MACs from hash
  functions", Advances in Cryptology -- CRYPTO‘95 Proceedings,
  Lecture Notes in Computer Science, Springer-Verlag Vol.963,
  1995, pp. 1-14.
   
  [MD5] Rivest, R., "The MD5 Message-Digest Algorithm",
  RFC 1321, April 1992.
   
   
   
  Krawczyk, et. al. Informational [Page 10]
 
  RFC 2104 HMAC February 1997
   
   
  [MM] Meyer, S. and Matyas, S.M., Cryptography, New York Wiley,
  1982.
   
  [RIPEMD] H. Dobbertin, A. Bosselaers, and B. Preneel, "RIPEMD-160: A
  strengthened version of RIPEMD", Fast Software Encryption,
  LNCS Vol 1039, pp. 71-82.
  ftp://ftp.esat.kuleuven.ac.be/pub/COSIC/bosselae/ripemd/.
   
  [SHA] NIST, FIPS PUB 180-1: Secure Hash Standard, April 1995.
   
  [Tsu] G. Tsudik, "Message authentication with one-way hash
  functions", In Proceedings of Infocom‘92, May 1992.
  (Also in "Access Control and Policy Enforcement in
  Internetworks", Ph.D. Dissertation, Computer Science
  Department, University of Southern California, April 1991.)
   
  [VW] P. van Oorschot and M. Wiener, "Parallel Collision
  Search with Applications to Hash Functions and Discrete
  Logarithms", Proceedings of the 2nd ACM Conf. Computer and
  Communications Security, Fairfax, VA, November 1994.
   
  Authors‘ Addresses
   
  Hugo Krawczyk
  IBM T.J. Watson Research Center
  P.O.Box 704
  Yorktown Heights, NY 10598
   
  EMail: hugo@watson.ibm.com
   
  Mihir Bellare
  Dept of Computer Science and Engineering
  Mail Code 0114
  University of California at San Diego
  9500 Gilman Drive
  La Jolla, CA 92093
   
  EMail: mihir@cs.ucsd.edu
   
  Ran Canetti
  IBM T.J. Watson Research Center
  P.O.Box 704
  Yorktown Heights, NY 10598
   
  EMail: canetti@watson.ibm.com
   
   
   
   
   
   
  Krawczyk, et. al. Informational [Page 11]
 
   

HMAC

标签:

原文地址:http://www.cnblogs.com/guxuanqing/p/5829841.html

(0)
(0)
   
举报
评论 一句话评论(0
登录后才能评论!
© 2014 mamicode.com 版权所有  联系我们:gaon5@hotmail.com
迷上了代码!