Certified Pseudonym
Colligated
with Master Secret Key
!

Vijay Pasupathinathan Macquarie University, Sydney.
joint work with
Josef Pieprzyk, Macquarie University, Sydney.
and, Huaxiong Wang, NTU, Singapore.
Outline
๏

Introduction

๏ Why a new system?
๏ Contribution
๏

How to achieve?

๏ Anonymous Certification systems
๏

Proposed Protocol

๏ Assumptions
๏ Protocol settings
๏ Security
๏

Applications, Advantages and open problem
What is a Pseudonym?
!
!

A mechanism to hide a user’s identity by providing anonymity,
while being still suitable to authenticate the holder of the
pseudonym in a communication system. (Chaum, 1985).
How are they achieved?
๏ Chaum and Evertse (1986) developed a pseudonym
system and proposed an RSA based implementation
while relying on a trusted centre who must sign all
credentials.
๏ Chen (1995) extended the scheme and presented its
discrete-logarithm version that relies on a trusted centre.

๏ However, these schemes have a common
weakness. Although the identity of the user is
hidden, the credentials (such as certificates of
his/her public key) or pseudonyms can be
easily shared (unauthorised transfer) with other
users.
How are they achieved? part 2
๏ Based on security of preserving a high-value (master) secret key,
Canettie et al. (2000) and Lysayanskaya et al.(1999)
independently proposed non-transferable pseudonym systems.
๏ Security is also based on the idea that “to force a user to reveal
the master secret key if they choose to share their credentials”.
๏ The problem is during the registration phase, users are required
to disclose their true identity (master public key) to a CA.
๏ Makes them prone to collusion between a CA and a Verifier.
What do we want!
๏ Pseudonym system based on a single trusted master
secret-public key pair.
๏ Pseudonyms should be independent of the master
public key. (Anonymity)
๏ Ability to generate multiple pseudonyms easily from a
single trusted secret-key. (Colligation)
๏ Verifiable using certificates that were issued against
pseudonyms.
Desired System
Pseudonym 1

Pseudonym 2

User
Trusted 

SK0/PK0

SK0

PK 1

PK 2

.
.
.
Pseudonym i

Cert<PK1>

PK i

.
.
.
Pseudonym n

Certifier

PK n

Cert<PK2>

.....

Cert<PKn>
Desired System
Pseudonym 1

PK 1

Certifier
Cert<PK1>

User
Trusted 

SK0/PK0

SK0

SK0{M}

PK 1

?

Cert<PK1>

Verify Message
using PK1

Think as group signatures looking through a mirror!

Verifier
Proposed Protocol
๏ Make use of an ACS (to certify pseudonyms)
๏ Make use of squaring (to provide colligation)
๏ There exits an underlying link between all pseudonyms
and the root secret key.
Anonymous Certification System
๏ Anonymous certification system (ACS) represents the
certification process of a public key by a certifier who
does not know the public key.
๏ This could essentially be a blind signature on the public
key of the user.
๏ That is, it provides anonymity to the receiver.
๏ Whereas, group signature schemes as employed by
provide anonymity to the source.
Anonymous Certification System
๏ Consists of four (4) entities: a user, verifier, certifier and a
trustee (tracer).
๏ The protocol suites include:
๏ a certification protocol, where an user interacts with the certifier to
obtain a certified pseudonym, i.e., the pseudonym is blindly
signed.
๏ An identification protocol, where verifier interacts with the user to
authenticate the user's credential and provide services.
๏ A trace protocol, where the trustee participates and is invoked to
trace the real identity associated with the user's pseudonym.
Security Assumptions
๏ Factoring: The probability that any probabilistic
polynomial time algorithm, can factor a composite
formed from two primes is negligible.
๏ Square Root: the probability that a probabilistic
polynomial time algorithm can output b such that b2 ≡ a
mod N, where a ∈ QRN, is negligible.
๏ Square Decisional Diffie-Hellmann: Distinguish between
distributions of the form (g, ga , ga2) from (g, ga , gr),
where r is random and uniformly chosen. We assume
that there is no probabilistic polynomial-time algorithm
that can solve a random instance of the SDDH problem.
The U master public-secret key-pair is generated
as in Section 2.1.1. U then obtains a certificate on the
master public key PKU 0 from a certification authority
C , which represents the U ’s true identity.
The public key of the certification authority is
obabilistic
SKC and the trustee is PK = gSKT , where
PKC = g
T
1
on input NAn user and SK are the corresponding secret keysthe
SKC generates new identities using for
๏
T
two prime
the certification authority
generation process and the trustee respectively.
a quadratic
put b, such
3.2 Identity Generation
๏ Which takes the inputs,
e probabil-

gorithm A ,
negligible
maller than

al Diffiee-Hellman
ws. Disthe form
andom and
N − 1}. We
olynomiala random
probability

onstruction

Identity Generation
following key

U generates secret key.
๏ Nj, g, master new identities using the following key
generation process, which takes the inputs, N j , g, a
counter value i i (indicating the total number new
๏ a counter value(indicating the total number of of new identities being
identities being generated), identity level l (number
generated),
of identities generated previously) and the master
๏ identity level U(number of identities generated previously).
secret key SKl 0 .
I-Generation(g,i,l,SKU 0 )
2
SKU

For j = l,. . . ,i do PKU j = g
Return(PKU l ,. . . ,PKU j )

j
0

mod N j EndFor

During the first run the value of identity level l
r user
aining
raphic
s.

ter public key is certified by the manufacturer,
and the following describes the certification of the
pseudonyms.
The user, U , generates pseudonyms of the form
Certifier
User
๏ A modified Certification scheme
(PKU 1 , . . . , PKU l ) using the identity generation pror∈ Z
based on blind signature
x = PK cess described in Section 3.2. The user then identifies
g
−−−−−−
−−−−−→
scheme by (Pointcheval, 2000)
β, γ, s ∈ Z
himself/herself (using, s)the master public key) to the
(X, Y ) = EncElg
(P K
PK
α = x ·engages
g
certifier andIHI(PK ∥(X,· Y )∥α) in a ๏ Signature scheme now includes
certify protocol to obtain a
δ=
=δ−
the . The public key of i
certificate e on γa pseudonym PKU i mastervalue of PKUthe
←−−−−−
−−−−−−
y = r − eSK
user which is used by the
is −−−−−−−−−−−→
never revealed to the certifier. We shall express this
certifier to form the
=
phase as x = yg + β P K
ρ

Certification

User

m

R

N0
r

U0

a1 =

g w ; a2

x

Verifier

k, w ∈R ZNi
= (PKT · PKU0 )w
k
h = I I(g 2 )
H

h,(a1 ,a2 ),(X,Y )

−−−−−−
−−−−−→

R

nts the
er who
ially a
public
he re-

N0

β−SKU

P KT
−γ
C

0

c1 ∈R ZNi
c2 = I I(X, Y, a1 , a2 )
H

U0

c1 ,c2

←−−−−−
−−−−−−

i

z1 = 2k − c1 · SK2 0
U

Ui

z1 ,z2 ,CERTC ⟨P KU ⟩

z2 = w − s · c2 − − − − − −i
−−−−−→

e

Verify CERTC ⟨P KUi ⟩
and obtain (α, δ)

C

y

?

?

y+SKU

δ ′ = I I(PKUi ∥(X, Y )∥α)
H
?

a1 = g z2 X c2 ; a2 = PKz2 Y c2
T

e
C

0

commitment and is later verified
by the , C ,
Figure 1: (PKU , CERTC ⟨PKU ⟩) ← Certi f y(Uuser.CERTC ⟨PKU ⟩)
Modified Blind Certification Protocol of
Figure 2: Identification Protocol
i
i
0
(Pointcheval, 2000) - The signature on PK is (α, δ, ρ) and
?

c
h = I I(g z P KUi )
H

Ui

d threea receiver can verify using the relation α = g PKC
between the verifier V and the trustee T . To trigger
i.e. “ U (e.g. applications in the certify protocol with C protocol parengages based on the protocol V has to provide proof of usfier V ,but certain applications
TPM) require the new identities to beU ⟩ to obtain a certificate on PKU ,
ing CERTC ⟨PK protected even ticipation by U . We shall express this phase as
0
otocolfrom the certifier. So, we propose a modification to (PKU ) ← Trace(V , T , PKU , CERTC ⟨PKU ⟩,i⟨PROOFU ⟩)
CERTC ⟨PKU i ⟩”.
the certification scheme based on a blind signature
U in-scheme using a composite modulus by Pointcheval
2000). The blind
scheme now
.e. the(Pointcheval,master public keysignatureuser which is
includes the
the
2.1.3 to form theofcommitment and is
Protocol Identify
proto-used by the certifier
?

ρ

δ

i

0

Verifier

i

Trustee

σ = SIGNV ⟨c, z, h⟩

σ,α,δ,ρ,PKU ,PKC
i

i
ol
ne
oee
y

U0 , s)

=
h
d on
even
s
on to
ature
rheval
now
e
ch is
nd is
us

ol of
) and

i

0

CERTC ⟨PKU i ⟩”.

Identification Protocol
2.1.3 Protocol Identify
๏ Based offered by a
A user U who wishes to avail serviceson Pointcheval
verifier V , engages in a identification protocol to conoptimised identification
vince that he/she possess the necessary (Pointcheval,
scheme credentials.
We shall express this phase as 2000)
User

Verifier

k, w ∈R ZNi
a1 = g w ; a2 = (PKT · PKU0 )w
k
h = I I(g 2 )
H

h,(a1 ,a2 ),(X,Y )

−−−−−−
−−−−−→

c1 ∈R ZNi
c2 = I I(X, Y, a1 , a2 )
H

c1 ,c2

←−−−−−
−−−−−−

i

z1 = 2k − c1 · SK2 0
U

z1 ,z2 ,CERTC ⟨P KU ⟩

z2 = w − s · c2 − − − − − −i
−−−−−→

Verify CERTC ⟨P KUi ⟩
and obtain (α, δ)

๏ Now also includes the DLEQ logg C = log ⟩, Y
⟨PROOFU i ⟩ ← Identi f y(U , V , PKU i , CERTX ⟨PKU iPKTPKT )
?

δ ′ = I I(PKUi ∥(X, Y )∥α)
H
?

a1 = g z2 X c2 ; a2 =
?
c
h = I I(g z P KUi )
H

PKz2 Y c2
T

Figure 2: Identification Protocol

i.e. “ U engages in an identification protocol with
a verifier V using the psuedonymn PKU i and
(PKCERT ⟨PK , ⟩ and ⟩, ⟨PROOF ⟩) contains the encryption of
) ← Trace(V , T , PK CERT ⟨PK which
C
Ui
the identity under the public key PKT ”.
between the verifier V and the trustee T . To trigger
the protocol V has to provide proof of protocol participation by U . We shall express this phase as
U0

Ui

Verifier

C

Ui

Trustee

σ = SIGNV ⟨c, z, h⟩
σ,α,δ,ρ,PKU ,PKC

2.1.4 Protocol Trace
−−−−−−
− − − −i − →
CERTC ⟨P KUi ⟩

VERIFY

⟨σ⟩

Ui
ρ) and

sed on
d even
ion to
nature
cheval
me now
hich is
and is

nature
ccomFigure
ol trivl’s pa-

based
cheme
cheme
DL-EQ
ses his
with a
verifier
in the
ted by

Figure 2: Identification Protocol

⟩ the trustee T . To trigger
CERTC ⟨PKU iand and which contains the encryption of
between the verifier V
the
provide proof
theprotocol V .has to under thisof protocol par- key PKT ”.
identityshall express the public
ticipation by U We
phase as

Tracing Protocol

(PKU 0 ) ← Trace(V , T , PKU i , CERTC ⟨PKU i ⟩, ⟨PROOFU i ⟩)

2.1.4 Protocol Trace
Verifier

Trustee

σ = SIGNV ⟨c, z, h⟩

๏ Invoked by a verifier after a user
has misused a pseudonym.

A verifier who needs to trace the identity of the user
๏ Verifier provides proof of a
contacts the trustee T by providing with the transcript
users participation.
from an identification protocol ⟨PROOFU i ⟩. We shall
๏ Trustee can reveal a user’s
express this phase as
σ,α,δ,ρ,PKU ,PKC

−−−−−−
− − − −i − →
CERTC ⟨P KUi ⟩

VERIFYPKV ⟨σ⟩
?

h = I I(g z PKc i )
H
U
?

α = g ρ PKδ
C
Verify CERTC ⟨P KUi ⟩
Obtain (X, Y ) from ⟨PROOFUi ⟩
P KU0 = DecElgSK (X, Y )
T

Figure 3: Tracing Protocol

master public key.

(PKU 0 ) ← Trace(V , T , PKU i , CERTC ⟨PKU i ⟩, ⟨PROOFU i ⟩)
4 SECURITY

i.e. “ V engages in the tracing protocol with T using
4.1 Adversary Goals
the values PKU i , CERTC ⟨PKU i ⟩ and proof of identity
We assume an active adversary A , who is capable of
use ⟨PROOFU i ⟩ messages in the commu- master identity PKU 0 ”.
eavesdropping and injecting to obtain the
nication medium. We also assume that an adversary
may be also be a legitimate (but dishonest) participant
in a protocol, i.e. either the certifier or the verifier or
both may be dishonest.
As in (Damgard, 1988; Lysyanskaya et al., 1999),
Security
๏ The proposal is secure against (as identified by Damgard,
1988; Lysyanskaya,1999):
๏ Pseudonym forgery: where an adversary tries to forge a
pseudonym for some user.
๏ Identity compromise: An adversary in association with
other participants tries to obtain information regarding the
user's master public-secret key-pair
๏ Pseudonym linking and colligation: An adversary tries to
obtain information that links a pair of pseudonyms to the
same user or to a user's master public key.
Application to TPM
๏ We are considering a TPM setting because of tamper resistant
protection offered to the master secret key, but the protocols
can be applied to other structures like directory based services
(e.g. active directory, LDAP)
๏ The endorsement (EK) in a TPM will be of the form (PK0 ,SK0)
๏ A user who wishes to obtain services from an application software on a
machine generates a pseudonym of the form (PKi ,SKi )
๏ At the end of the protocol run the application software is provided a
guarantee on the identity of the user and the associated TPM, but the
system still protects the identity of both the TPM and the user
associated with it.
Advantages
๏ Compared to other pseudonym schemes, our scheme
has an efficient identification protocol.
๏ Computations may be performed on the module itself,
whereas the DAA scheme requires computation to be
distributed among the TPM and the host computer.
๏ there are no new secret key to be generated for each
pseudonyms, only counter values of the pseudonym
๏ no appreciable increase in storage requirement even
when the number of pseudonyms required are high
๏ ideally suited for storage constraint devices
What’s Missing? Future Work?
๏ Needs a strong composite modulus. (May be 4096 bits)
๏ Prime modulus method ruled out, as SDDH is trivial.

๏ Every generated pseudonym needs to fall with the same
group as the master secret key.
๏ Identity Transfer
๏ Pseudonym chains cannot be formed. (NOT YET!)
๏ That is, using PK1 to generate new pseudonyms, but still
verifiable using SK0.
Thank You
vijay@cprotocol.com

More Related Content

PDF
Asssignment2
PPTX
Tools and Techniques for Understanding Threading Behavior in Android*
PPTX
Yash Agarwal
PDF
Voce Tem Orgulho Do Seu Codigo
PDF
Advanced Dynamic Analysis for Leak Detection (Apple Internship 2008)
DOC
Tai lieu ky thuat lap trinh
PDF
Java Basics - Part2
PPTX
Accelerating Habanero-Java Program with OpenCL Generation
Asssignment2
Tools and Techniques for Understanding Threading Behavior in Android*
Yash Agarwal
Voce Tem Orgulho Do Seu Codigo
Advanced Dynamic Analysis for Leak Detection (Apple Internship 2008)
Tai lieu ky thuat lap trinh
Java Basics - Part2
Accelerating Habanero-Java Program with OpenCL Generation

What's hot (20)

PPT
Tutorial s crypto api session keys
PDF
Eric Lafortune - ProGuard: Optimizer and obfuscator in the Android SDK
PPTX
Writing Good Tests
PDF
Writing SOLID C++ [gbgcpp meetup @ Zenseact]
PPTX
How Data Flow analysis works in a static code analyzer
RTF
cs8project
PDF
Eric Lafortune - ProGuard and DexGuard for optimization and protection
PPTX
Technology, Process, and Strategy
PPT
Inheritance and-polymorphism
PDF
Blockchain: Developer Perspective
PPT
شرح مقرر البرمجة 2 لغة جافا - الوحدة الرابعة
PDF
Blockchain - a formal introduction
PDF
The art of reverse engineering flash exploits
PDF
PHP Benelux 2012: Magic behind the numbers. Software metrics in practice
PDF
A survey on Fully Homomorphic Encryption
PPTX
TypeScript - All you ever wanted to know - Tech Talk by Epic Labs
PPTX
Games, AI, and Research - Part 2 Training (FightingICE AI Programming)
PDF
Tools and Techniques for Understanding Threading Behavior in Android
PPTX
Qt Memory Management & Signal and Slots
PPTX
Java весна 2013 лекция 2
Tutorial s crypto api session keys
Eric Lafortune - ProGuard: Optimizer and obfuscator in the Android SDK
Writing Good Tests
Writing SOLID C++ [gbgcpp meetup @ Zenseact]
How Data Flow analysis works in a static code analyzer
cs8project
Eric Lafortune - ProGuard and DexGuard for optimization and protection
Technology, Process, and Strategy
Inheritance and-polymorphism
Blockchain: Developer Perspective
شرح مقرر البرمجة 2 لغة جافا - الوحدة الرابعة
Blockchain - a formal introduction
The art of reverse engineering flash exploits
PHP Benelux 2012: Magic behind the numbers. Software metrics in practice
A survey on Fully Homomorphic Encryption
TypeScript - All you ever wanted to know - Tech Talk by Epic Labs
Games, AI, and Research - Part 2 Training (FightingICE AI Programming)
Tools and Techniques for Understanding Threading Behavior in Android
Qt Memory Management & Signal and Slots
Java весна 2013 лекция 2
Ad

Similar to Certified Pseudonym Colligated with Master Secret Key (20)

PDF
Computer security module 3
PPT
User authentication crytography in cse engineering
PPTX
Blockchain privacy approaches in hyperledger indy
PPT
ch13 ABCD.ppt
PDF
Vtu network security(10 ec832) unit 3 notes.
PDF
Authentication in Different Scenarios
PDF
Simple Overview of PKI and Digital signature by Tarek_Gaber
PPT
Unit - 3.ppt
PPTX
Public-Key Cryptography.pptx
PPT
authentication u5.ppt
PDF
Certificate less key management scheme in
PDF
CERTIFICATE LESS KEY MANAGEMENT SCHEME IN MANET USING THRESHOLD CRYPTOGRAPHY
PPTX
Network security
PDF
Anonymous credentials with range proofs, verifiable encryption, ZKSNARKs, Cir...
PDF
Em24873876
PDF
Untraceable electronic mail, return addresses and digital pseudonyms - David ...
PDF
Analysis and improvement of pairing free certificate-less two-party authentic...
PDF
Computationally Efficient ID-Based Blind Signature Scheme in E-Voting
PDF
Authentication in Different Scenarios
PPT
ok_mary_pki1234public_key_encryption.ppt
Computer security module 3
User authentication crytography in cse engineering
Blockchain privacy approaches in hyperledger indy
ch13 ABCD.ppt
Vtu network security(10 ec832) unit 3 notes.
Authentication in Different Scenarios
Simple Overview of PKI and Digital signature by Tarek_Gaber
Unit - 3.ppt
Public-Key Cryptography.pptx
authentication u5.ppt
Certificate less key management scheme in
CERTIFICATE LESS KEY MANAGEMENT SCHEME IN MANET USING THRESHOLD CRYPTOGRAPHY
Network security
Anonymous credentials with range proofs, verifiable encryption, ZKSNARKs, Cir...
Em24873876
Untraceable electronic mail, return addresses and digital pseudonyms - David ...
Analysis and improvement of pairing free certificate-less two-party authentic...
Computationally Efficient ID-Based Blind Signature Scheme in E-Voting
Authentication in Different Scenarios
ok_mary_pki1234public_key_encryption.ppt
Ad

Recently uploaded (20)

PPT
Geologic Time for studying geology for geologist
PDF
Zenith AI: Advanced Artificial Intelligence
PDF
A comparative study of natural language inference in Swahili using monolingua...
PDF
Produktkatalog für HOBO Datenlogger, Wetterstationen, Sensoren, Software und ...
PDF
Taming the Chaos: How to Turn Unstructured Data into Decisions
PDF
Developing a website for English-speaking practice to English as a foreign la...
PDF
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
PPT
Module 1.ppt Iot fundamentals and Architecture
PDF
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
PDF
A Late Bloomer's Guide to GenAI: Ethics, Bias, and Effective Prompting - Boha...
PPTX
Configure Apache Mutual Authentication
PPTX
Microsoft Excel 365/2024 Beginner's training
PDF
The influence of sentiment analysis in enhancing early warning system model f...
PDF
UiPath Agentic Automation session 1: RPA to Agents
PDF
Getting started with AI Agents and Multi-Agent Systems
PDF
sbt 2.0: go big (Scala Days 2025 edition)
PDF
CloudStack 4.21: First Look Webinar slides
PDF
Consumable AI The What, Why & How for Small Teams.pdf
PDF
sustainability-14-14877-v2.pddhzftheheeeee
PDF
Five Habits of High-Impact Board Members
Geologic Time for studying geology for geologist
Zenith AI: Advanced Artificial Intelligence
A comparative study of natural language inference in Swahili using monolingua...
Produktkatalog für HOBO Datenlogger, Wetterstationen, Sensoren, Software und ...
Taming the Chaos: How to Turn Unstructured Data into Decisions
Developing a website for English-speaking practice to English as a foreign la...
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
Module 1.ppt Iot fundamentals and Architecture
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
A Late Bloomer's Guide to GenAI: Ethics, Bias, and Effective Prompting - Boha...
Configure Apache Mutual Authentication
Microsoft Excel 365/2024 Beginner's training
The influence of sentiment analysis in enhancing early warning system model f...
UiPath Agentic Automation session 1: RPA to Agents
Getting started with AI Agents and Multi-Agent Systems
sbt 2.0: go big (Scala Days 2025 edition)
CloudStack 4.21: First Look Webinar slides
Consumable AI The What, Why & How for Small Teams.pdf
sustainability-14-14877-v2.pddhzftheheeeee
Five Habits of High-Impact Board Members

Certified Pseudonym Colligated with Master Secret Key

  • 1. Certified Pseudonym Colligated with Master Secret Key ! Vijay Pasupathinathan Macquarie University, Sydney. joint work with Josef Pieprzyk, Macquarie University, Sydney. and, Huaxiong Wang, NTU, Singapore.
  • 2. Outline ๏ Introduction ๏ Why a new system? ๏ Contribution ๏ How to achieve? ๏ Anonymous Certification systems ๏ Proposed Protocol ๏ Assumptions ๏ Protocol settings ๏ Security ๏ Applications, Advantages and open problem
  • 3. What is a Pseudonym? ! ! A mechanism to hide a user’s identity by providing anonymity, while being still suitable to authenticate the holder of the pseudonym in a communication system. (Chaum, 1985).
  • 4. How are they achieved? ๏ Chaum and Evertse (1986) developed a pseudonym system and proposed an RSA based implementation while relying on a trusted centre who must sign all credentials. ๏ Chen (1995) extended the scheme and presented its discrete-logarithm version that relies on a trusted centre. ๏ However, these schemes have a common weakness. Although the identity of the user is hidden, the credentials (such as certificates of his/her public key) or pseudonyms can be easily shared (unauthorised transfer) with other users.
  • 5. How are they achieved? part 2 ๏ Based on security of preserving a high-value (master) secret key, Canettie et al. (2000) and Lysayanskaya et al.(1999) independently proposed non-transferable pseudonym systems. ๏ Security is also based on the idea that “to force a user to reveal the master secret key if they choose to share their credentials”. ๏ The problem is during the registration phase, users are required to disclose their true identity (master public key) to a CA. ๏ Makes them prone to collusion between a CA and a Verifier.
  • 6. What do we want! ๏ Pseudonym system based on a single trusted master secret-public key pair. ๏ Pseudonyms should be independent of the master public key. (Anonymity) ๏ Ability to generate multiple pseudonyms easily from a single trusted secret-key. (Colligation) ๏ Verifiable using certificates that were issued against pseudonyms.
  • 7. Desired System Pseudonym 1 Pseudonym 2 User Trusted SK0/PK0 SK0 PK 1 PK 2 . . . Pseudonym i Cert<PK1> PK i . . . Pseudonym n Certifier PK n Cert<PK2> ..... Cert<PKn>
  • 8. Desired System Pseudonym 1 PK 1 Certifier Cert<PK1> User Trusted SK0/PK0 SK0 SK0{M} PK 1 ? Cert<PK1> Verify Message using PK1 Think as group signatures looking through a mirror! Verifier
  • 9. Proposed Protocol ๏ Make use of an ACS (to certify pseudonyms) ๏ Make use of squaring (to provide colligation) ๏ There exits an underlying link between all pseudonyms and the root secret key.
  • 10. Anonymous Certification System ๏ Anonymous certification system (ACS) represents the certification process of a public key by a certifier who does not know the public key. ๏ This could essentially be a blind signature on the public key of the user. ๏ That is, it provides anonymity to the receiver. ๏ Whereas, group signature schemes as employed by provide anonymity to the source.
  • 11. Anonymous Certification System ๏ Consists of four (4) entities: a user, verifier, certifier and a trustee (tracer). ๏ The protocol suites include: ๏ a certification protocol, where an user interacts with the certifier to obtain a certified pseudonym, i.e., the pseudonym is blindly signed. ๏ An identification protocol, where verifier interacts with the user to authenticate the user's credential and provide services. ๏ A trace protocol, where the trustee participates and is invoked to trace the real identity associated with the user's pseudonym.
  • 12. Security Assumptions ๏ Factoring: The probability that any probabilistic polynomial time algorithm, can factor a composite formed from two primes is negligible. ๏ Square Root: the probability that a probabilistic polynomial time algorithm can output b such that b2 ≡ a mod N, where a ∈ QRN, is negligible. ๏ Square Decisional Diffie-Hellmann: Distinguish between distributions of the form (g, ga , ga2) from (g, ga , gr), where r is random and uniformly chosen. We assume that there is no probabilistic polynomial-time algorithm that can solve a random instance of the SDDH problem.
  • 13. The U master public-secret key-pair is generated as in Section 2.1.1. U then obtains a certificate on the master public key PKU 0 from a certification authority C , which represents the U ’s true identity. The public key of the certification authority is obabilistic SKC and the trustee is PK = gSKT , where PKC = g T 1 on input NAn user and SK are the corresponding secret keysthe SKC generates new identities using for ๏ T two prime the certification authority generation process and the trustee respectively. a quadratic put b, such 3.2 Identity Generation ๏ Which takes the inputs, e probabil- gorithm A , negligible maller than al Diffiee-Hellman ws. Disthe form andom and N − 1}. We olynomiala random probability onstruction Identity Generation following key U generates secret key. ๏ Nj, g, master new identities using the following key generation process, which takes the inputs, N j , g, a counter value i i (indicating the total number new ๏ a counter value(indicating the total number of of new identities being identities being generated), identity level l (number generated), of identities generated previously) and the master ๏ identity level U(number of identities generated previously). secret key SKl 0 . I-Generation(g,i,l,SKU 0 ) 2 SKU For j = l,. . . ,i do PKU j = g Return(PKU l ,. . . ,PKU j ) j 0 mod N j EndFor During the first run the value of identity level l
  • 14. r user aining raphic s. ter public key is certified by the manufacturer, and the following describes the certification of the pseudonyms. The user, U , generates pseudonyms of the form Certifier User ๏ A modified Certification scheme (PKU 1 , . . . , PKU l ) using the identity generation pror∈ Z based on blind signature x = PK cess described in Section 3.2. The user then identifies g −−−−−− −−−−−→ scheme by (Pointcheval, 2000) β, γ, s ∈ Z himself/herself (using, s)the master public key) to the (X, Y ) = EncElg (P K PK α = x ·engages g certifier andIHI(PK ∥(X,· Y )∥α) in a ๏ Signature scheme now includes certify protocol to obtain a δ= =δ− the . The public key of i certificate e on γa pseudonym PKU i mastervalue of PKUthe ←−−−−− −−−−−− y = r − eSK user which is used by the is −−−−−−−−−−−→ never revealed to the certifier. We shall express this certifier to form the = phase as x = yg + β P K ρ Certification User m R N0 r U0 a1 = g w ; a2 x Verifier k, w ∈R ZNi = (PKT · PKU0 )w k h = I I(g 2 ) H h,(a1 ,a2 ),(X,Y ) −−−−−− −−−−−→ R nts the er who ially a public he re- N0 β−SKU P KT −γ C 0 c1 ∈R ZNi c2 = I I(X, Y, a1 , a2 ) H U0 c1 ,c2 ←−−−−− −−−−−− i z1 = 2k − c1 · SK2 0 U Ui z1 ,z2 ,CERTC ⟨P KU ⟩ z2 = w − s · c2 − − − − − −i −−−−−→ e Verify CERTC ⟨P KUi ⟩ and obtain (α, δ) C y ? ? y+SKU δ ′ = I I(PKUi ∥(X, Y )∥α) H ? a1 = g z2 X c2 ; a2 = PKz2 Y c2 T e C 0 commitment and is later verified by the , C , Figure 1: (PKU , CERTC ⟨PKU ⟩) ← Certi f y(Uuser.CERTC ⟨PKU ⟩) Modified Blind Certification Protocol of Figure 2: Identification Protocol i i 0 (Pointcheval, 2000) - The signature on PK is (α, δ, ρ) and ? c h = I I(g z P KUi ) H Ui d threea receiver can verify using the relation α = g PKC between the verifier V and the trustee T . To trigger i.e. “ U (e.g. applications in the certify protocol with C protocol parengages based on the protocol V has to provide proof of usfier V ,but certain applications TPM) require the new identities to beU ⟩ to obtain a certificate on PKU , ing CERTC ⟨PK protected even ticipation by U . We shall express this phase as 0 otocolfrom the certifier. So, we propose a modification to (PKU ) ← Trace(V , T , PKU , CERTC ⟨PKU ⟩,i⟨PROOFU ⟩) CERTC ⟨PKU i ⟩”. the certification scheme based on a blind signature U in-scheme using a composite modulus by Pointcheval 2000). The blind scheme now .e. the(Pointcheval,master public keysignatureuser which is includes the the 2.1.3 to form theofcommitment and is Protocol Identify proto-used by the certifier ? ρ δ i 0 Verifier i Trustee σ = SIGNV ⟨c, z, h⟩ σ,α,δ,ρ,PKU ,PKC i i
  • 15. ol ne oee y U0 , s) = h d on even s on to ature rheval now e ch is nd is us ol of ) and i 0 CERTC ⟨PKU i ⟩”. Identification Protocol 2.1.3 Protocol Identify ๏ Based offered by a A user U who wishes to avail serviceson Pointcheval verifier V , engages in a identification protocol to conoptimised identification vince that he/she possess the necessary (Pointcheval, scheme credentials. We shall express this phase as 2000) User Verifier k, w ∈R ZNi a1 = g w ; a2 = (PKT · PKU0 )w k h = I I(g 2 ) H h,(a1 ,a2 ),(X,Y ) −−−−−− −−−−−→ c1 ∈R ZNi c2 = I I(X, Y, a1 , a2 ) H c1 ,c2 ←−−−−− −−−−−− i z1 = 2k − c1 · SK2 0 U z1 ,z2 ,CERTC ⟨P KU ⟩ z2 = w − s · c2 − − − − − −i −−−−−→ Verify CERTC ⟨P KUi ⟩ and obtain (α, δ) ๏ Now also includes the DLEQ logg C = log ⟩, Y ⟨PROOFU i ⟩ ← Identi f y(U , V , PKU i , CERTX ⟨PKU iPKTPKT ) ? δ ′ = I I(PKUi ∥(X, Y )∥α) H ? a1 = g z2 X c2 ; a2 = ? c h = I I(g z P KUi ) H PKz2 Y c2 T Figure 2: Identification Protocol i.e. “ U engages in an identification protocol with a verifier V using the psuedonymn PKU i and (PKCERT ⟨PK , ⟩ and ⟩, ⟨PROOF ⟩) contains the encryption of ) ← Trace(V , T , PK CERT ⟨PK which C Ui the identity under the public key PKT ”. between the verifier V and the trustee T . To trigger the protocol V has to provide proof of protocol participation by U . We shall express this phase as U0 Ui Verifier C Ui Trustee σ = SIGNV ⟨c, z, h⟩ σ,α,δ,ρ,PKU ,PKC 2.1.4 Protocol Trace −−−−−− − − − −i − → CERTC ⟨P KUi ⟩ VERIFY ⟨σ⟩ Ui
  • 16. ρ) and sed on d even ion to nature cheval me now hich is and is nature ccomFigure ol trivl’s pa- based cheme cheme DL-EQ ses his with a verifier in the ted by Figure 2: Identification Protocol ⟩ the trustee T . To trigger CERTC ⟨PKU iand and which contains the encryption of between the verifier V the provide proof theprotocol V .has to under thisof protocol par- key PKT ”. identityshall express the public ticipation by U We phase as Tracing Protocol (PKU 0 ) ← Trace(V , T , PKU i , CERTC ⟨PKU i ⟩, ⟨PROOFU i ⟩) 2.1.4 Protocol Trace Verifier Trustee σ = SIGNV ⟨c, z, h⟩ ๏ Invoked by a verifier after a user has misused a pseudonym. A verifier who needs to trace the identity of the user ๏ Verifier provides proof of a contacts the trustee T by providing with the transcript users participation. from an identification protocol ⟨PROOFU i ⟩. We shall ๏ Trustee can reveal a user’s express this phase as σ,α,δ,ρ,PKU ,PKC −−−−−− − − − −i − → CERTC ⟨P KUi ⟩ VERIFYPKV ⟨σ⟩ ? h = I I(g z PKc i ) H U ? α = g ρ PKδ C Verify CERTC ⟨P KUi ⟩ Obtain (X, Y ) from ⟨PROOFUi ⟩ P KU0 = DecElgSK (X, Y ) T Figure 3: Tracing Protocol master public key. (PKU 0 ) ← Trace(V , T , PKU i , CERTC ⟨PKU i ⟩, ⟨PROOFU i ⟩) 4 SECURITY i.e. “ V engages in the tracing protocol with T using 4.1 Adversary Goals the values PKU i , CERTC ⟨PKU i ⟩ and proof of identity We assume an active adversary A , who is capable of use ⟨PROOFU i ⟩ messages in the commu- master identity PKU 0 ”. eavesdropping and injecting to obtain the nication medium. We also assume that an adversary may be also be a legitimate (but dishonest) participant in a protocol, i.e. either the certifier or the verifier or both may be dishonest. As in (Damgard, 1988; Lysyanskaya et al., 1999),
  • 17. Security ๏ The proposal is secure against (as identified by Damgard, 1988; Lysyanskaya,1999): ๏ Pseudonym forgery: where an adversary tries to forge a pseudonym for some user. ๏ Identity compromise: An adversary in association with other participants tries to obtain information regarding the user's master public-secret key-pair ๏ Pseudonym linking and colligation: An adversary tries to obtain information that links a pair of pseudonyms to the same user or to a user's master public key.
  • 18. Application to TPM ๏ We are considering a TPM setting because of tamper resistant protection offered to the master secret key, but the protocols can be applied to other structures like directory based services (e.g. active directory, LDAP) ๏ The endorsement (EK) in a TPM will be of the form (PK0 ,SK0) ๏ A user who wishes to obtain services from an application software on a machine generates a pseudonym of the form (PKi ,SKi ) ๏ At the end of the protocol run the application software is provided a guarantee on the identity of the user and the associated TPM, but the system still protects the identity of both the TPM and the user associated with it.
  • 19. Advantages ๏ Compared to other pseudonym schemes, our scheme has an efficient identification protocol. ๏ Computations may be performed on the module itself, whereas the DAA scheme requires computation to be distributed among the TPM and the host computer. ๏ there are no new secret key to be generated for each pseudonyms, only counter values of the pseudonym ๏ no appreciable increase in storage requirement even when the number of pseudonyms required are high ๏ ideally suited for storage constraint devices
  • 20. What’s Missing? Future Work? ๏ Needs a strong composite modulus. (May be 4096 bits) ๏ Prime modulus method ruled out, as SDDH is trivial. ๏ Every generated pseudonym needs to fall with the same group as the master secret key. ๏ Identity Transfer ๏ Pseudonym chains cannot be formed. (NOT YET!) ๏ That is, using PK1 to generate new pseudonyms, but still verifiable using SK0.