![](3D"master72_image004.gif")
On Privacy and Anonymi=
ty in Knowledge
Externalization
Yuen-Yan Chan and Chi-Hong Leung
<=
/span>
The Chinese University of Hong Kong
rosannachan@cuhk.ed=
u.hk , leung_=
chi_hong@yahoo.com.hk
Secure Knowledge Management (SKM 2004)=
24 September 2004
Buffalo USA=
![](3D"master71_image005.jpg")
![](3D"master71_image006.jpg")
Outline
|Knowledge
Externalization
zApproaches for Knowled=
ge
Externalization
zPrivacy Issues in Know=
ledge
Externalization
|Solution to Anonymity&=
#13;
zNetwork-Layer Solution=
s
zUse of Alias
zZero-knowledge Proofs&=
#13;
|Our Protocol
|Security Analysis
=
|Efficiency
Knowledge Externalization
|Knowledge
management includes the capability to collect, archive, mana=
ge,
evaluate, and distribute knowledge across an
organization
|Tacit kno=
wledge – “internal knowledge=
221;, persona=
l beliefs,
perspective and value
|Explicit =
knowledge
– exists in form of rea=
dable documents,
records
|Knowledge
Externalization: the transformation of tacit knowledge to exp=
licit
knowledge
Approaches for Knowled=
ge Externali=
zation
|Knowledge externalization involves the harnessin=
g of
tacit knowledge to explicit knowledge.
|Typical ways of achiev=
ing it
includes:
zpostings on an electro=
nic
discussion board
zthe uploading of one’s own records
zgenerating reports on =
one’s experience
zrecording of meetings,
interviews and phone calls
|usually involve the sharing and exposing of o=
ne’s internal =
knowing
Privacy Issues in Know=
ledge Externali=
zation
|Technical=
ly,
achieving knowledge externalization requires logging-o=
n to a
system with authorized identity
|However, =
to log-on
to knowledge management systems using the true identit=
y is
often subjected to the following threats:
zerroneous=
and
obtrusive requests for information
zbotherati=
on by
potential information seekers
ztraceabil=
ity of
the information providers (for example, when an employee =
is
speaking out against the management)
|Therefore=
, it is
desirable that the source of information is hidden=
but author=
ized at the s=
ame
time
|Solution:=
Anonymous
Authentication
Solutions to Anonymity
|Unconditi=
onal
anonymity, or complete anonymity (in which no authentic=
ation
is required and the users are freely login to a system), hinders =
knowledge=
exchange
as the source of information cannot be traced
<=
/span>
|Also, such
anonymity may be abused as one is not responsible for his or=
her
actions
|Therefore=
, what we
are interested is a more challenging solution: =
condit=
ional
anonymity (hereinafter refer to as “anonymity=
”).
|Three main
approaches to provide anonymity: &n=
bsp;
network-layer approach, use of alias, and zero-knowledge=
proofs
approach
Solution to Anonymity =
– Network Layer =
Approach<=
/span>
|Achieve anonymity in th=
e network
layer
|A user’s
action is hidden within the actions of many others
|Examples:
zC=
rowds
(Reiter and Rubin, 1997)
zM=
IX
(Chaum, 1981)
zO=
nion
Routing (Syverson, Goldschlag, and Reed, 1997)
Solution to Anonymity - Alias
|Use alias=
, or a
pseudonym, to substitute for the true identity of an entity&=
#13;
|Usually, a
pre-authentication phrase prior to the authentication is
required
zin
pre-authentication phrase, the entity proves its identity to a server us=
ing
conventional authentication methods
zafterward=
, the
server randomly generates a number which is uncorrela=
ted to
the true identity of the entity and digitally signs on it
|Examples:=
zPseudonym Sy=
stems
proposed by Lysyanskaya et al. (Lysyanskaya, Rivest, Sah=
ai, and
Wolf, 1999)
zT<=
span
lang=3DEN-US style=3D'font-size:71%;mso-ansi-language:EN-US'>emporary ide=
ntity
(TID) to achieve anonymity in wireless communication systems (Go=
and
Kim, 2001)
Solution to Anonymity =
– Zero Knowledge=
Proofs
|Is an interactive proof with a prover P and a=
verifier =
V
|P convinces V of the knowledge of a secret, <=
/span>without r=
evealing
any information about the secret or how to go about pro=
ving
this secret
|Examples:
<=
/span>
zElectronic C=
ash
proposes by David Chaum (Chaum, Fiat, and Naor, 1988)
zNon-transfer=
able
electronic voting pass proposed by Chan et al. (Chan, Wong a=
nd
Chan, 2000)
zOur
Protocol
Our Protocol - Preliminary
|A =
– User
|B =
– Knowledge management
server
|C - Centr=
al
repository
|
|(d, e) – be the private-public=
key
pair of B
|n =3D p•q where p and q are tw=
o large
prime numbers. d•e =3D 1 mod
(p-1)(q-1)
|u =
– the identity of A
=
;
|a, b, c a=
nd r – some random integers&=
#13;
|k =
– security parameter=
3;
|f() and g=
() – two-argument one-way
collision free hash functions
|R =
– a set, and R’ be mutually exclusive=
to R’.
Our Protocol – Pre-authentication
|In this phrase, A obtains an anonymo=
us pass
fro=
m B and
deposits the identity revocation value to C
|We employ the blind sig=
nature
technique (Chaum,
1983) in the generation of the anonymous pass
<=
span
class=3DBB style=3D'position:absolute;left:-4.71%;top:.39em'>|A pass signed by B is
produced
=
|Please refer to paper s=
ection
4.2 for details
Our Protocol - Authentication
|When A logins to the knowledge management server B, A and
B perform =
authentication
|To do this, A presents the pass that it obtained from the
Pre-Authentication phase
|B undergo random challenges on the p=
ass to check A’s
knowledge on the pass
=
|Please refer to paper s=
ection
4.3 for details
Our Protocol - Revocation
|In case A misbehaves, the revocation phase can be executed to revo=
ke the
identity of A.
<=
span
class=3DBB style=3D'position:absolute;left:-4.71%;top:.39em'>|This phase is performed=
by B and C.
Security Analysis
|Authentic=
ity
zB authent=
icates A
based on the pass verification
|Anonymity=
zA<=
span
lang=3DEN-US style=3D'font-family:Arial;mso-hansi-font-family:Arial;font-=
size:
71%;mso-fareast-language:ZH-TW'>’s identity is protected
throughout the process
|Masquerade
Prevention
zDuring the
pre-authentication phrase, A is not anonymous and B should ma=
ke sure A’s identity before sign=
ing on
the pass
| Chance of Successful Cheating by =
A
z1/(2k/2), which decreases exponentially with the
value of the security parameter k
|Stolen Pa=
ss
zSuppose t=
he pass
is stolen by eavesdropper E during pre-authentication phrase,=
this
will not bring any lost to A because E does not know the valu=
es of ai, bi, ci,and ai XOR
u.
Efficiency
|Most operations involved in our prot=
ocol are hashes and random n=
umber
generation
|Hashes are light in terms of computational power (O’Mahony,
Peirce, and Tweari, 1997).
Conclusion
|We have s=
tudied
the privacy issues involved in knowledge
externalization
|In order =
to
protect the privacy of the knowledge source, we have propos=
ed a
solution for conditional anonymous authentication
zin which =
one can
login to a knowledge management system anonymously, wh=
ile the
identity will be revoked conditionally
|We have a=
lso
reviewed the methods for providing anonymity in general&#=
13;
|Security =
and
efficiency analysis of our proposed protocol is also provi=
ded
Reference
|Y. Chan, =
J. C.
Wong, and A. C. Chan (2000) Anonymous Electronic Voting System with
Non-Transferable Voting Passes. In proceedings=
of SEC 2000,
pp.321-330.
|D. Chaum.=
(1981)
Untraceable electronic mail, return addresses, and digital pseudonyms,
Communications of the ACM.
|D. Chaum.=
(1983)
Blind Signature for Untraceable Payment. Advances in Cryptology – Proceedings of CRYPTO=
’82, pp.199-20=
3.
|D. Chaum,=
A. Fiat
and M. Naor. (1988) Untraceable Electronic Cash. Advances in Cryptology -
Proceedings of Crypto '88, pp. 319-327.
|R. Hauser=
and G.
Tsudik. (1996) On Shopping Incognito. Proceedings of the 2nd USENIX Works=
hop
on Electronic Commerce, pp.251-7.
|J. Go and=
K. Kim.
(2001) Wireless authentication protocol preserving user anonymity. In Pro=
c. of
the 2001 Symposium on Cryptography and Information Security (SCIS 2001).=
3;
|R. Molva,=
D.
Samfat, and G. Tsudik. (1994) Authentication of Mobile Users. IEEE Networ=
k,
Special Issue on Mobile Communications, March/=
April
1994, pp. 26-34.
|G. Ng-Kru=
elle, P.
A. Swatman, D. S. Rebne and J. F. Hampe. (2002) The Price of Convenience:
Privacy and Mobile Commerce. Proceedings of the 3rd World Congre=
ss on
the Management of Electronic Commerce.
|A. Lysyan=
skaya, R.
Rivest, A. Sahai, and S. Wo=
lf
(1999) Pseudonym Systems. Proceedings of Selected Areas in =
Cryptogra=
phy 99,
Lecture Notes in Computer Science, Springer-Verlag, vol. 1758, pp.
184-199.
|D. O’Mahony, M. Peirce, H. =
Tewari
(1997) Electronic Payment Systems. Artech House, 1996.
=
|I. Nonaka=
and J.
Reinmoeller. (1998) The ART of Knowledge: Systems to Capitalize on Market
Knowledge, European Management Journal, vol. 16, no. 6, pp. 673-684.
|M. Polany=
i. (1966)
The Tacit Dimension. London: Routledge & Kegan Paul
|M. K. Rei=
ter and
A. D. Rubin (1997) Crowds: Anonymity for web transactions. Technical Repo=
rt
97-15, DIMACS
|E. Sallis=
and G.
Jones. (2002) Knowledge Management in Education: Enhancing Learning &
Education, London: Kogan Page.
|A. L. Sch=
irmer.
(2003) Privacy and Knowledge Management: Challenges in the Design of the =
Lotus
Discovery Server, IBM Systems Journal, vol 42, no 3, pp. 519-531.&=
#13;
|P. F. Syv=
erson, D.
M. Goldschlag, and M. G. Reed. (1997) Anonymous Connections and Onion Rou=
ting.
Proceedings of the 18th Annual Symposium on Security and
Privacy, pp. 44-54.
‹header›
̸=
9;date/time›
按一下以編輯=
597;片
第二層
第三層
第四層
第五層
‹footer›
&=
#8249;#›
' + ENDSHOW_MESG + '
![](3D"notes_flag.gif")
N=
otes![](3D"buttons.gif")
![](3D"master71_image001.png"){kind=link}
![](3D"master71_image002.png"){kind=link}