This page was last updated Feb 27, 2015.
 What is OCB?
 Can you describe how OCB works?
 Are there different versions of OCB?
 What are some of OCB’s properties?
 How much faster is OCB than CCM or GCM?
 Is OCB good in hardware?
 What papers of mine deal with OCB?
 Is the notion of authenticatedencryption something new?
 What does the name stand for?
 What blockcipher should I use with OCB?
 Should one write “OCBAES” or “AESOCB”?
 Is OCB in standards?
 What did Jutla do?
 What did Gligor and Donescu do?
 What is this attack described by Ferguson?
 What is a nonce and why do you need one?
 Is there code available? Is it free?
 Are OCB test vectors available?
 Is it safe to use a new algorithm in a product / standard?
 Are there competing integrated authenticatedencryption proposals?
 Are there competing twopass authenticatedencryption proposals?
 Is OCB patented?
 Do I need a license?
 Who is the author of this webpage?
What is OCB?
OCB is a blockcipherbased mode of operation that
simultaneously provides
both privacy and authenticity for a usersupplied plaintext.
Such a method is called an authenticatedencryption scheme.
What makes OCB remarkable
is that it achieves authenticated encryption in almost the same
amount of time as the fastest conventional mode, CTR mode, achieves privacy alone.
Using OCB one buys privacy and authenticity about as cheaply (or more cheaply—CBC was once the norm) as
one used to pay for achieving privacy alone.
Despite this, OCB is simple and clean,
and easy to implement in either hardware or software.
OCB accomplishes its work without
bringing in the machinery of universal hashing,
a technique that does not seem to lend itself to implementations
that are simple and fast in both hardware and software.
Somewhat more precisely, OCB solves the problem of
noncebased authenticatedencryption with associateddata (AEAD).
The associateddata part of the name means that
when OCB encrypts a plaintext it can bind it to some other string,
called the
associated data, that is authenticated but not encrypted.
The noncebased part of the name means that OCB requires a nonce
to encrypt each message.
OCB does not require the nonce to be random; a counter, say, will work fine.
Unlike some modes, the plaintext provided to OCB can
be of any length, as can the associated data, and OCB
will encrypt the plaintext without padding it to some
convenientlength string, an approach that would yield a longer ciphertext.
The design of OCB was strongly influenced by Charanjit Jutla’s mode
IAPM; I consider the
papers I’ve written related to OCB to be
followon work to Jutla’s paper.
I took up the design of OCB largely in response to
NIST’s modesofoperation activities: NIST had figured out that not only
do we need a modern blockcipher (AES),
but one also needs, just as importantly, modern ways to use use the
construct well.
In the past, when one wanted a sharedkey mechanism providing
both privacy and authenticity,
the usual thing to do was to
separately encrypt and compute a MAC, using two different keys.
(The word MAC stands for Message Authentication Code.)
The encryption is what buys you privacy and the
MAC is what gets you authenticity.
The cost to get privacyandauthenticity in this way is
about the cost to encrypt (with a privacyonly scheme) plus the cost to MAC.
If one is encrypting and MACing in a conventional way,
like CTRmode encryption and the CBC MAC,
the cost for privacyandauthenticity is
going to be twice the cost for privacy alone, just counting blockcipher calls.
Thus people have often hoped for a simple and cheap way to get
message authenticity as an almost “incidental’ adjunct to message privacy.
Sure;
I’ll draw a pretty picture, too.
To make things concrete,
let’s assume you’re using a blockcipher of E=AES with 128bit keys.
The block length is n=128 bits.
Let’s assume a nonce of 96 bits.
Other values can be accommodated,
but 96 bits (12 bytes) is the recommended nonce length.
Let M be the message you want to encrypt.
Let A be the associated data (if there isn’t any, regard it as the empty string).
Let K be the OCB encryption key, which is just a key for the underlying blockcipher, AES.
Let N be the 96bit nonce—for example, a counter that gets incremented with
each message you encrypt.
Let’s first look at the setting when M is a positive multiple of 128 bits.
Then break M into 128bit blocks
M = M_{1} ... M_{m}.
Encryption then works as shown in the following picture.
Read each column toptobottom, and then go across lefttoright.
We’ll defer describing the Init and Inc functions for just a moment.
(These functions depend on K too, but we haven’t reflected that
in the notation.)
The value Checksum is the 128bit string
Checksum = M_{1} ⊕…⊕ M_{m}.
We’ll describe how to compute the value Auth in moment.
It’s determined by
A and K.
The number τ, the tag length of the scheme, is, like the blockcipher
E, a parameter of the mode.
It’s a number 0 ≤ τ ≤ 128.
Now the picture:
The ciphertext
is the 128m + τ bit string
CT = C_{1} C_{2} …
C_{m} T.
Let us explain how Init works.
(After that, we’ll explain how the function Inc works.)
We are given a 96bit nonce N.
We save the last six bits as the number Bottom (a value between 0 and 63)
and then we create a 128bit value Top by prepending to N the
the 32bit constant 0x00000001 and zeroing out the last six bits.
Let Ktop = E_{K} (Top) and
let Stretch be the 256bit string Stretch = Ktop  (Ktop⊕(Ktop<<8)).
By S << i we mean the left shift of the 128bit string S by
i positions (leftmost bits vanish, zero bits come in on the right).
The value denoted above as Init(N), the initial value for Δ, is
the first 128 bits of Stretch << Bottom.
Why all this peculiar stuff? Beyond the fact that one can prove that it does
what we want, the intent is very simple: if N
is counter, Ktop is going to change only once every 64 calls.
As a consequence, if we keep Ktop around as long as needed, then, 63 out of 64 times,
all that will be needed to compute Init(N) is a logical operation to extract
the lowest six bits of N and then a logical shift of the string Stretch
by this many bits. Just a few cycles. The other one time out of 64 we will
need a blockcipher call. The amortized cost to compute Init,
meaning the average cost over a sequence of successive calls, is extremely low,
and we don't need to implement anything like a GF(2^{128}) multiply.
Now let’s explain how the Inc_{i} function works.
First, for S a 128bit string, let
double(S) = S << 1 if the first bit of S is 0,
and let
double(S) = (S << 1)⊕135 otherwise (where 135 denotes
the 128bit string that encodes that decimal value).
For those of you for whom this means something,
double(S) is just the multiplication of S by the field point
“x” when using a particular representation for
the finite field with 2^{128} points.
Given all these Kderived Lvalues, we set
L_{∗}=E_{K}(0^{128}),
L_{$}=double(L_{∗}),
L[0]=double(L_{$}), and
L[j]=double(L[j1]) for all j ≥ 1.
Given the above, define
Inc_{i} (Δ) = Δ ⊕L[j] where
j = ntz(i) is the number of trailing zero bits in the binary representation
of i.
Likewise define
Inc_{∗}(Δ) = Δ⊕L_{∗}.
and
Inc_{$}(Δ) = Δ⊕L_{$}.
We call the increment function we have described “tablebased” because, to implement them,
it is natural to precompute a table of
128bit
L_{∗},
L_{$}, and
L[j].
Then, for each kind of increment, we just xor in the needed value from the table.
Decryption under OCB works in the expected way: given K, N, and
CT = C_{1} C_{2} …
C_{m} T,
recover M in the natural way and recompute the tag T* that “should”
be at the end of CT.
Regard M as the (authentic) underlying plaintext for CT if
T = T*.
Regard the ciphertext as invalid if T ≠ T*.
When M is not a positive multiple of 128 bit we process the final block
a little differently. This time we break the plaintext M into blocks
M = M_{1} ... M_{m} M_{∗} where
M_{i} = 128 and
0 ≤ M_{∗} < 128.
Computation works as in the following picture.
We redefine Checksum as
Checksum = M_{1} ⊕…⊕
M_{m} ⊕ M_{∗} 10* where the
M_{∗} 10*
notation means to append a single 1bit and then the minimum number of 0bits to get the string to be 128 bits.
Finally, I’ll describe how to compute Auth, the 128bit string one gets
by processing the associated data A.
As before, we distinguish two cases, when A is or is not a multiple of
128 bits.
The picture below shows the two settings.
The increment function Inc is just as before, the only thing that is different
is the initial value of Δ, which is Δ = Init = 0^{128}.
When A is the empty string, the value returned is Auth = 0^{128}.
There are three named version of OCB, corresponding to three points in
time over the algorithm’s 10year evolution.
If I need to be specific, I’ll say OCB1, OCB2, or OCB3.
I sometimes use the name “OCB” as a generic label for all of the different
versions. Other times I use OCB to mean the final algorithm, OCB3.
For example, when I described OCB in the section above, I was in fact describing OCB3.
I don’t like to change an algorithm I’ve published;
there should be a compelling reason to do so.
Let me describe what was significant with each algorithm in the chain.
 OCB1 — From the ACM CCS 2001 paper by Rogaway, Bellare, Black, and Krovetz.
 The scheme that my team first defined built on the ideas of Charanjit Jutla’s IAPM.
OCB1 brought to the table an ability to handle arbitrarylength plaintexts, doing so with
minimal expansion in the length of the ciphertext. To make this happen, message padding could not
be used. Also, IV requirements were minimized; the IV only had to be a nonce. The number
of blockciphers calls was reduced to m+2 for encrypting an mblock message.
We made do with a single underlying blockcipher key. The increment function was made simpler and
more efficient than what Jutla had described.
 OCB2 — From the Asiacrypt 2004 paper by Rogaway.
 Mode OCB1 had a defect that practitioners were quick to point out:
it had not been designed to natively handle associateddata (AD).
Associateddata refers to stuff, say a message header, that
needs to be authenticated but should not encrypted.
Thanks to Burt Kaliski, of RSA,
for first bringing to my attention the need for AD support.
With OCB2 I added in support for AD.
I also implemented a new way of doing the increment (Inc) function, using
doubling instead of the tablebased approach.
I had thought that doing increments by doubling would make for a faster scheme.
Finally, I recast things in the tweakableblockcipher setting
that had been suggested earlier by Liskov, Rivest, and Wagner.
 OCB3 — From the FSE 2011 paper by Krovetz and Rogaway.
 The final version of OCB goes back to a tablebased
approach for incrementing offsets (the Δvalues); experimental
findings made it clear that doubling was not the win that I had thought.
Beyond this, we (usually) shave off one blockcipher call and
(usually) eliminate the blockcipher latency that we used to suffer two or three times.
McGrew and Viega had complained about these two defects in a 2004 paper of theirs.
In doing this new work Krovetz and I wanted to demonstrate that the extra blockcipher
call and the extra bit of latency were
inessential aspects of the basic design.
We also wanted to see if these matters actually mattered.
We found that our changes did speed things up, although the improvement was
relatively small.
There were no security issues that motivated the evolution of OCB;
all versions have been proven secure, with similar bounds,
assuming the blockcipher has its expected property (it’s a “strong PRP”).
I note that, in refining OCB, I have resisted the temptation to design for “beyond birthdaybound”
security; while this can be done, the benefit, using currently known techniques, does not seem worth the
added complexity, assuming that one is using a 128bit blockcipher.
The changes from OCB1 to OCB2 to OCB3 aren’t really all that big, so
why have I bothered to continue to refine this mode?

One answer is that I regard authenticatedencryption
as one of the crucial “outputs” of the cryptographic community; it is one a handful
of interfaces and abstractions where a cryptographer can create
something that may actually get used and make a difference. Given the primitive’s importance,
it seems to me that, in some sense, my community has failed if we do not deliver,
and get into popular use, the best AE schemes that we can devise.

Beyond this, I kind of got obsessed.
I wanted to achieve AE just as cleanly and efficiently as could be.
Even if a scheme was practical and recordsetting fast,
it didn’t seem enough if it could be made faster or more elegant.
It was an accident—yet it seems a fitting fact—that OCB looks a lot like
OCD.
If you implement OCB, please use the final version, OCB3.
In time, I hope the name OCB, when understood to be a particular algorithm,
will be understood to mean OCB3.
OCB has been designed to simultaneously have lots of desirable properties.
Listed in no particular order, these properties include the following.
 OCB uses an amortized m + a + 1.016 blockcipher calls, where
m is the blocklength of the plaintext M and a is the blocklength of the
associated data A.
This assumes a counter nonce. (In the worst case, when the nonce is not a counter, the
1.016 constant becomes the number 2.)
 If the header A is fixed during a session then, after preprocessing,
there is effectively no cost to have A authenticated: the mode
will use m + 1.016 blockcipher calls regardless of A.
 OCB is fully parallelizable: its blockcipher calls
may be performed simultaneously. (There is a dependency of the first block on the nonceassociated
blockcipher call one time in 64.)
Thus OCB is suitable for encrypting messages in hardware at
the highest network speeds, and it helps avoid pipeline stalls when implemented
in software. This is especially important in the presence of new AES instructions on
Intel processors. The parallelizability of OCB makes it suitable for use with bitsliced
AES implementations, too.
 Assuming the underlying blockcipher is not susceptible to timing attack, natural implementations
of OCB would not be, either.
 OCB is online:
one need not know the length of A or M to proceed with encryption, and
one need not know the length of A or C to proceed with decryption.
 One can truncate the tag T to any number of bits and see the expected behavior.
For example, if one wants to use τ=32 bit tags, the adversary will
be able to forge each ciphertext with probability 1/2^{32}, but low querycomplexity
or low computingtime attacks will not exist—in contrast, say,
to GCM (see the
Ferguson attack).
 OCB can encrypt messages of any bit length.
Messages don’t have to be a multiple of the blocklength, and no
separate padding regime is ever used.
No bits are wasted in the ciphertext due to padding;
ciphertexts are of “minimal” length.
 OCB allows arbitrary associateddata A to be specified when
one encrypts or decrypts a string.
This string will be authenticated but not encrypted.
 Encryption and decryption depend on a nonce N, which must be
selected as a new value for each encryption.
The nonce need not be random or secret.
A counter will work fine. Nonces may have any number of bits less than 128.
The recommendation is to use a 96bit nonce.
 OCB uses a single key K for the underlying blockcipher, and all blockcipher
calls are keyed by K.
 The main computational work beyond the blockcipher calls consists of
three 128bit xors per block.
 Keysetup would typically involve just a single blockcipher call and some doubling operations.
 OCB can be implemented to run in very little memory: the main memory cost is that needed to hold
the AES subkeys.
 OCB avoids 128bit addition, which is endianbiased and can be expensive
in software or dedicated hardware.
In general, OCB is nearly endianneutral.
 OCB respects wordalignment in the message and the header.
 OCB enjoys provable security.
One must assume that the underlying blockcipher is secure in the
customary (strong PRP) sense. Security falls off
in s^{2}/2^{128}
where s is the total number of blocks one acts on.
A very strong form of privacy is achieved:
noncebased indistinguishability from random bits.
Likewise, a very strong form of authenticity is achieved:
noncebased authenticity of ciphertexts.
 OCB nicely supports incremental interfaces. The processing of each
128bit block of plaintext is independent of whether or not this is the final block of plaintext.
 The only parameters of OCB are the blockcipher E and the tag length τ. The message space
and nonce space are independent of these choices.
How much faster is OCB than CCM or GCM?
Usually much faster—like a factor of 26—but a precise answer
depends on a many factors. Please see the
performance page
for charts and tables across various platforms.
Is OCB good in hardware?
OCB was designed to perform well not only in software, but also in
hardware; the algorithm is efficiently realizable by
smart cards, FPGAs, and highspeed ASICs.
Pipeline delays have been minimized (especially in OCB3), and
short messages, like long ones, are optimized for.
Chip area, latency, and power consumption should be dominated
by the underlying blockcipher.
Here are a few papers that discuss hardware implementations of OCB:
What papers of mine deal with OCB?
OCB was developed over a sequence of academic papers published
between 2001 and 2011:
 The original OCB paper.
The proceedings version is in
ACM CCS 2001 and
the journal version is in
ACM TISSEC 2003.
 A paper about dealing with associated data. Appears in
ACM CCS 2002.
 A paper to develop the message authentication code, PMAC, that shaped the way AUTH
is computed in OCB. Appears in
Eurocrypt 2002.
 A paper about efficiently realizing tweakable blockciphers, and
about using tweakable blockcipher to improve OCB. Appears in
Asiacrypt 2004.
 Finally, a timing study about AE modes, along with refinements to get to the final
version of OCB. Appears at FSE 2011.
OCB is a wellknown mode. It represents the
stateoftheart for a blockcipherbased mode of operation for authenticated
encryption.
Is the notion of authenticatedencryption something new?
Yes and no.
The general idea to combine
privacy and authenticity is folklore, and lots of attempts can
be found in the security literature and in systems.
From a modern perspective,
the problem was routinely being solved incorrectly, but nobody in the
cryptographic community was paying much attention.
The first
definitions for the authenticatedencryption goal appeared in the
Asiacrypt 2000 paper BN00 of Bellare and Rogaway, and,
independently, in the
FSE 2000 paper KY00 of Katz and Yung.
Further investigation of the notion appears in the Asiacrypt 2000 paper
BN00 of Bellare and Namprempre.
Noncebased notions of authenticated encryption were first formalized in the ACM CCS 2001 paper
cited above, while
the role of associated data was first formalized in the ACM CCS 2002 paper cited above.
What does the name stand for?
It was originally intended to stand for offset codebook.
That’s a highly abbreviated
operational description of what is going on in the mode:
for most message blocks,
one offsets the block, applies the blockcipher, and then offsets the
result once again.
I prefer you call the mode “OCB” than what this acronym was meant to suggest.
What blockcipher should I use with OCB?
You can use OCB with any blockcipher that you like.
But the obvious choice is AES.
AES actually comes in three flavors: AES128, AES192, and AES256.
I like AES128.
If you use OCB with a blockcipher having a 64bit blocklength
instead of a 128bit blocklength,
you’d get a worse security bound and would need to change keys
well before you encrypt 2^{32} blocks (as with all wellknown
modes that use a 64bit blockcipher).
This is a fairly onerous requirement;
OCB is really only meant to be used with 128bit blockciphers.
I prefer OCBAES (or OCBAES128), going from highlevel protocol on down.
Just like HMACSHA1.
Is OCB in standards?
OCB3 is defined in RFC 7253,
an informational RFC sponsored by the CFRG.
Before this, OCB1 was an optional method in the IEEE
802.11 standard,
where it was called WRAP (Wireless Robust Authentication Protocol).
Finally, OCB2 is specified in
ISO/IEC 19772:2009. That standard also includes AE
schemes CCM, GCM, EAX, and two further schemes.
Unfortunately, the
table of contents
seems to be all that I can legally provide for an ISO standard.
While OCB1 was submitted to NIST, it was never adopted as a NIST recommendation.
William Burr and others have explained that
NIST was reluctant to standardize any authenticatedencryption
mode with associated IP.
What did Jutla do?
I am not the first to give an integrated
authenticatedencryption mode of operation.
Charanjit Jutla, from IBM Research,
was the first to publicly describe a correct blockcipherbased mode
of operation that combines privacy and authenticity at a
small increment to the cost of providing privacy alone.
Jutla’s scheme appears as
IACR2000/39.
He went on to publish the work in a
Eurocrypt 2001 paper.
A Journal of Cryptology
version appeared in 2008.
Jutla actually described two schemes: one is CBClike (IACBC)
and one is ECBlike (IAPM). The latter
is what OCB builds on.
OCB uses highlevel ideas from Jutla’s scheme
but adds in many additional tricks.
Virgil Gligor and Pompiliu Donescu
were working on authenticatedencryption schemes at the same time that Jutla was.
They described an authenticatedencryption mode, XCBC, in
[Gligor, Donescu; Aug 18, 2000], taken from
Gligor’s homepage.
The mode is not similar to OCB, but
it is similar to Jutla’s IACBC.
We do not know the precise history of
Gligor/Donescu and Jutla in devising their schemes;
Jutla’s scheme was publicly distributed first, but only by a little.
OCB was invented after both pieces of work made their debut.
One of the contributions of
[Gligor, Donescu] is the use
of mod2^{n} arithmetic in this context. In particular,
Jutla had suggested the use of
modp addition, and it was nonobvious that moving into
this weaker algebraic structure would be OK.
Later versions of [Gligor, Donescu; 18 August 2000] added in new schemes.
Following Jutla, [Gligor, Donescu; October 2000] mention
a parallelizable mode of operation similar to Jutla’s IAPM.
Following my own work,
[Gligor, Donescu; April 2001] updated their
modes to use a single key and
to deal with messages of arbitrary bit length.
A 2002 note by Niels Ferguson,
Collision attacks on OCB,
points out that, after about 2^{64} blocks of messages are queried
(300 million terabytes), all authenticity with OCB is forfeit.
This is just the usual birthday attack.
Ferguson finds this alarming.
He should not.
Security of almost every encryption (or authentication) scheme based
on a 128bit blockcipher
vanishes at a rate of σ^{2}/2^{128} where
σ is the number of blocks queried.
And security, once lost, is routinely &ldsqu;completely&rdsquo; lost—what
Ferguson seems to object to most.
There are a few blockcipherbased AE schemes that have been designed to achieve
security beyond the birthday bound;
for example, see
Authenticated
Encryption Mode for Beyond The Birthday Bound Security, by T. Iwata.
Our own belief is that, for a standardized mode of operation based on a 128bit blockcipher,
beyondbirthdaybound security would be worthwhile only if it can be achieved
very easily and very cheaply (say 1\% overhead compared to a more conventional bound).
That’s not what existing
solutions are offering.
Birthdaybound attacks (as well as good cryptographic hygine)
motivate rekeying well in advance of birthdaybound concerns.
In RFC 7253 we say
that a given a key should be used to encrypt at most 2^{48} blocks (about 280 terabytes).
In OCB, you must present a nonce each time you want to
encrypt a string.
When you decrypt, you need to present the same nonce.
The nonce doesn’t have to be random or secret or unpredictable.
It does have to be something new with each message you encrypt.
A counter value will work for a nonce, and that is what is recommended.
It doesn’t matter what the counter is initialized to.
In general, a nonce is a string that is used used at most one time
(either for sure, or almost for sure) within
some established context.
In OCB, this “context” is the “session” associated to the
underlying encryption key K.
For example, you may distribute the key K by a sessionkey
distribution protocol, and then start using it.
The nonces should now be unique within that session.
Generating new nonces is the user’s responsibility, as
is communicating them to the party that will decrypt.
A nonce is nothing new or strange for an encryption mode.
If there were no nonce there would be only one
ciphertext for each plaintext, key, and AD, and this means that the
scheme would necessarily
leak information (e.g., is the plaintext for
the message just received the same as the plaintext for the
previous message received?).
In CBC mode the counterpart of a nonce is called an IV (initialization vector)
and it needs to be adversarially unpredictable if you’re going to meet a
strong definition of security.
In OCB we have a weaker requirement on the nonce, and
so we refrain from calling it an IV.
What happens if you repeat the nonce?
You’re going to mess up authenticity for all future messages, and you’re
going to mess up privacy for the messages that use the repeated
nonce.
So don’t do this.
It is the user’s obligation to ensure that nonces don’t
repeat within a session.
In settings where this is infeasible, OCB should not be used.
We note that most conventional encryption modes fail, often quite miserably,
if you repeat their supplied IV/nonce.
Only recently (2006) did cryptographer’s develop a notion for an encryption
scheme being maximally “resilient” to noncereuse.
The notion is due to Tom Shrimpton and me and we also developed a scheme,
SIV, to solve this problem. But it’s a conventional
composed scheme: half the speed, or worse, of OCB.
SIV makes a nice alternative to a composed scheme like CCM,
EAX, offering comparable performance but a better
security guarantee.
Yes there’s code and yes it’s free.
Ted Krovetz wrote some nice code.
See the toplevel OCB page for a link.
Other code for OCB is showing up in various cryptographic libraries.
For example, OCB is now in OpenSSL
In correctly coding OCB, I think the biggest difficulty is
endian problems. The spec is subtly bigendian
in character, a consequence of it’s use of “standard”
mathematical conventions.
If your implementation doesn’t agree with the reference one,
chances are there’s some silly endianrelated error.
There are wellknown perils with implementing encryption algorithms,
such as needlessly opening the door to
timing attacks.
OCB test vectors can be found near the end of
RFC 7253.
Is it safe to use a new algorithm in a product / standard?
In the past, one had to wait years
before using a new cryptographic scheme;
one needed to give cryptanalysts
a fair chance to attack the thing. Assurance in a scheme’s correctness
sprang from the absence of damaging attacks by smart people,
so you needed to wait long enough that at least a few smart people would try,
and fail, to find a damaging attack.
But this entire approach has become largely outmoded,
for schemes that are not true primitives,
by the advent of provable security.
With a provablysecure scheme assurance does not stem from a failure
to find attacks; it comes from proofs,
with their associated bounds and definitions.
Our work includes a proof that OCBE
is secure as long as blockcipher E is secure (where E is an arbitrary block
cipher and “secure” for OCBE and secure for E
are welldefined notions).
Of course it is conceivable that OCBAES could fail because AES
has some huge, unknown problem.
But other issues aren’t really of concern.
Here we’re in a domain
where the underlying definition is simple and rock solid;
where there are no “random oracles” to worry about;
and where the underlying cryptographic assumptions are completely standard.
There are, although none are as refined as OCB.
See the KrovetzRogaway paper
The software performance of authenticatedencryption modes
for a chart pointing to a number of AE schemes. Going back to the original
integrated (aka, onepass) schemes from Jutla and
from GligorDonescu:
 Jutla (2000) propose two modes, IACBC and IAPM.
The latter is the parallelizable scheme and is therefore the one to compare to OCB.
Various methods are specified for offset
generation in IAPM, but none are competitive with OCB.
Jutla assumes that messages have a multiple of n bits; padding is used,
presumably, to deal with other lengths.
Once a padding regime is added, say obligatory 10*padding,
the length of ciphertexts will
be longer than OCB ciphertexts any time that the
message being encrypted is
a nonmultiple of 16 bytes.
IAPM uses twice the key material of OCB.
Associated data is not handled.
 Gligor and Donescu (2000) propose
four authenticatedencryption schemes, which the authors call
XCBC$XOR, XCBCXOR, XCBCSXOR, and XECBSXOR.
Only the last of these modes
is parallelizable.
The addition of a parallelizable scheme
came subsequent to the publication of Jutla’s work,
and XECBSXOR is similar to Jutla’s IAPM.
XECBSXOR uses mod 2^{n} addition instead of modp addition.
This difference might suggest an efficiency
improvement, but this is unclear, as
[Jutla] uses one modp addition and three 128bit xors while
[Gligor, Donescu] uses three 128bit additions and one 128bit xor.
One expects the use of an additive ring (integers mod 2^{n}) instead of
a field (GF(p) or GF(2^{n})) to worsen any possible security bound by
at least a log factor.
No proof has been given for this mode.
Following my own work, [Gligor, Donescu] includes techniques
so as to use a single key and to deal with messages of
arbitrary bit length. But the techniques are not pushed
as far as with OCB and there is ciphertextexpansion
for all messages which are a nonmultiple of the 16 bytes.
In my opinion, there remains no desirable, standardized, integratedscheme alternative
to OCB; no other integrated scheme has as yet gotten
any traction in products or standards.
But various composed schemes are reasonable alternatives to OCB,
particularly CCM and GCM.
And the CAESAR competition,
which has drawn an amazing 57 submissions, is likely to eventually
yield some wellvetted alternatives.
Are there competing composed
authenticatedencryption proposals?
The traditional approach for achieving authenticatedencryption was to use
generic composition.
Using separate keys, you should encrypt the plaintext and then MAC the
resulting ciphertext. You can use any encryption scheme and any MAC
that you like, and the composite scheme is guaranteed to do what it should do.
To offer the same basic “service” as OCB
you’ll need, in the end, to do something like this. For privacy,
use CBC mode with ciphertextstealing and an IV derived by enciphering
the nonce. For authenticity,
use the CBC MAC variant known as CMAC.
Use standard key separation to make all the needed keys.
Make sure the MAC is taken over the IV of the encryption scheme.
There are various pitfalls people run into when trying to do a homebrewed
combination of privacy and authenticity. Common errors include:
(1) a failure to properly perform key separation;
(2) a failure to use a MAC that is secure across different message lengths;
(3) omitting the IV from what is MACed;
(4) encrypting the MACed plaintext as opposed to the superior method
(at least with respect to general assumptions)
of MACing the computed ciphertext.
For all of these, I cannot advise people to rolltheirown
authenticatedencryption scheme.
Even standards get things wrong.
A Eurocrypt 2014
paper by Namprempre, Shrimpton, and me reexamines the genericcomposition approach and what
has been understood of it.
It turns out that even standards (like the ISO standard mentioned above) do not get things rights.
The problem, in part, is that the theory literature’s early starting points and constructive goals
do not match up with standardized starting points and the conventional AEAD goal.
To date, the most significant composed authenticated encryptions
schemes are
CCM and GCM.
Some other techniques include
SIV and
EAX.
All of these modes comprise IPfree means of performing authenticatedencryption with associateddata.

CCM is an authenticatedencryption mode designed by
Housley, Ferguson, and Whiting. It’s inspired by the generic
composition of counter mode encryption and the CBC MAC
(but CCM is not in fact an instance of generic composition).
CCM was developed as a lessefficient but IPfree alternative to OCB.
It is specified in
NIST Special Publication 80038C.
It’s kind of a cobbledup thing,
but the it got into 802.11i,
and then other standards. The bounds are good and no significant weaknesses are
known.
Most criticisms focus on efficiency concerns, including
the one I wrote with David Wagner, A Critique of CCM.

GCM is an authenticatedencryption mode designed by
McGrew and Viega.
It combines a privacyonly encryption mechanism (CTR mode)
with a CarterWegman MAC,
the latter based on multiplication in the finite field with 2^{128} points.
The initial purpose of GCM was to have a parallelizable, hardwareefficient, and
patentfree AEAD scheme.
Being parallelizable means that it is useful for
highspeed hardwarebased applications.
Having to implement multiplies in GF(2^{128}) can
be slow unless large tables are used, but it’s pretty fast
if you use such tables, and hardware chip area for GF(2^{128})
multiplication can be low.
New Intel chips support a “carryless multiply” instruction meant to
help the mode run fast.
GCM is specified in Draft NIST publication 80038D.
We also elaborate on SIV, a deterministic or misuseresistant authenticatedencryption
mode. It was designed by
Rogaway and Shrimpton to solve the “key wrap” problem.
When used as a noncebased AEAD scheme, SIV has the unusual property that
it doesn’t “break” if a nonce should get reused: all that
happens is that repetitions of this exact (nonce, AD, plaintext)
are revealed.
The mode can also be used without any nonce and, when used as such,
all it reveals are repetitions in (nonce, AD, plaintext) pairs.
SIV suffers from a latency problem (unavoidable for the
deterministic or misuseresistant AEAD goal): one must not only
make two passes over the input, but one cannot output any ciphertext
bits until all plaintexts bits are read.
SIV is described in a
Eurocrypt 2006 paper and an associated spec,
RFC 5297, by Dan Harkins.
The main advantage of OCB over CCM, GCM, and other composed schemes is software
speed. See the performance page
for data.
It is; I myself have received US patents
7,949,129, and
8,321,675 on OCB.
(Patents
7,046,802 and
7,200,227 were
allowed to lapse, and expired 2014.05.16 and 2015.04.02.)
That said, I have freely licensed any IP I own over
a very large space: there is
one license grant for opensource software;
one for nonmilitary software; and a third
done just for OpenSSL.
See the licensing page for more information.
If you want to use OCB in a manner not covered by the patent grants,
please contact me.
I license OCB under fair, reasonable, and nondiscriminatory terms, with
licensees paying a modest onetime fee.
There are further patents in the AE space. I would single out those of
Jutla (IBM)— 6,963,976,
7,093,126, and
8,107,620—and of
Gligor and Donescu (VDG)—6,973,187.
Do the claims of these patents read against OCB?
It is difficult to answer such a question,
and I am not in a position to do so. Read the claim language yourself.
I freely license OCB for most (but not all) settings.
Please see the separate licensing page
for further information.
I am,
Phillip Rogaway,
a professor in the
Department of Computer Science at the
University of California, Davis.
I have also been a regular visitor to universities in Thailand,
most often at the Department of Computer Science at
Chiang Mai University.
For 20 years I’ve worked to develop and popularize
a style of cryptography I call practiceoriented provable security.
This framework is a perfect fit
to the problem of designing an authenticatedencryption scheme.
One of the “outputs” of practiceoriented provable security
has been cryptographic schemes that are wellsuited standards.
In particular, I am coinventor on the schemes
known as CMAC, DHIES, EAX, OAEP, PSS, and XTS,
which are in standards from ANSI, IEEE, ISO, and NIST.
The kind of academic work I do can be seen by looking at
dblp
or
google Scholar.
Back to the
OCB homepage