Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
‹date/time›
=
‹fo=
oter›
=
‹#&=
#8250;
=
------=_NextPart_01C4F1BA.74EC70C0
content-location: file:///C:/2AA2647F/4A-03-Bhargava__files/master12.xml
content-transfer-encoding: quoted-printable
content-type: text/xml;
charset="utf-8"
Click to edit Master title style=
Click to edit Master subtitle st=
yle
‹=
;date/time›
‹footer›
‹#›&=
#13;
------=_NextPart_01C4F1BA.74EC70C0
content-location: file:///C:/2AA2647F/4A-03-Bhargava__files/master13.xml
content-transfer-encoding: quoted-printable
content-type: text/xml;
charset="utf-8"
‹header›
‹date/time›
‹footer›
‹#›
------=_NextPart_01C4F1BA.74EC70C0
content-location: file:///C:/2AA2647F/4A-03-Bhargava__files/pres.xml
content-transfer-encoding: quoted-printable
content-type: text/xml;
charset="utf-8"
![](3D"master13_image001.gif")
1
Trust-based
Privacy Preservation for Peer-to-peer
Data Sharing
Y. Lu=
, W.
Wang, D. Xu, and B. Bhargava
yilu,
wangwc, dxu, bb @ cs.purdue.edu
Depar=
tment
of Computer Sciences
Purdue University=
The
work is supported by NSF ANI-0219110 and IIS-0209059
![](3D"master12_image002.gif")
![](3D"master12_image003.gif")
![](3D"master12_image004.gif")
![](3D"master12_image005.gif")
![](3D"master12_image006.gif")
2
=
Problem statement
nPrivacy in peer-to-peer systems is <=
span
style=3D'position:absolute;top:28.74%;left:13.85%;width:89.13%;height:7.5=
%'>different
from the anonymity problem
=
nPrese=
rve
privacy of requester
nA me=
chanism
is needed to remove the association between the identity of=
the requester
and the data needed3
Proposed solution
nA
mechanism is proposed that allows the peers to acquire data th=
rough
trusted proxies to preserve privacy of requester
nThe data request is handled through the peer’s
proxies
nThe proxy can become a supplier later and mask
the original requester
4
Related work
nTrust
in privacy preservation
nAuthorization
based on evidence and trust, [Bhargava and Zhong,
DaWaK’02]
nDeveloping pervasive trust [Lilien, CGW’03]
=
nHiding
the subject in a crowd
nK-anonymity [Sweeney, UFKS’02]
nBroadcast and multicast [Scarlata et al, INCP’01]
5
Related work (2)
nFixed
servers and proxies
nPublius [Waldman et al, USENIX’00]
nBuilding
a multi-hop path to hide the real source and destination&#=
13;
nFreeNet [Clarke et al, IC’02]
=
nCrowds [Reiter and Rubin, ACM TISS’98]
nOnion routing [Goldschlag et al, ACM Commu.’99]
6
Related work (3)
n [Sherwood e=
t al,
IEEE SSP’02]
n provides sender-rece=
iver
anonymity by transmitting packets to a broadcast group
<=
/span>
nHerbivore
[Goel et al, Cornell Univ Tech Report’03]
nProvides
provable anonymity in peer-to-peer communication systems by
adopting dining cryptographer networks
7
Privacy measurement
nA tuple <requester ID, data handle, data content> is defined to describe a data =
acquirement.
nFor each element, “0” means that the =
peer knows nothing, while “1” means that i=
t knows
everything.
=
nA state in which the requester’s privacy is=
compromised can be represented as a vector <1, 1, y>, (y Є [0,1=
]) from
which one can link the I=
D of
the requester to the data that it is interested
in.
8
For example, line k
represents=
the
states that
the requester’s privacy is compromised.
Privacy measurement (2)
9
Mitigating collusion
=
nAn operation
“*” is defined as:
=
=
=
nThis
operation describes the revealed information
after a collusion of two peers when each
peer knows a part of the “secret”.
nThe number of col=
lusions
required to compr=
omise
the secret can be used to evaluate the
achieved privacy 10
Trust=
based
privacy preservation scheme
nThe
requester asks one proxy to look up the data on its behalf. =
Once
the supplier is located, the proxy will get the data and deliver
it to the requester
nAdvantage:
other peers, including the supplier, do not know the real
requester
nDisadvantage: The privacy solely depends on the
trustworthiness and reliability of the proxy
11
Trust=
based
scheme – Improvement 1
nTo avoid specifying the data handle in plain text, the requester calculates the hash code and =
only reveals a part of it to the proxy.
nThe proxy sends it to possible suppliers.
nReceiving the partial hash code, the supplier compares it to the hash codes of the data =
handles that it holds. Depending on the revealed part, multiple matches may be found.
=
;
nThe suppliers then construct a bloom filter based on the remaining parts of the matched hash codes and send it back. They also send back their public key certificates.=
12
Trust=
based
scheme – Improvement 1
nExamining the filters, the requester can eliminat=
e some candidate suppliers and finds some who may have t=
he data.
nIt then encrypts the full data handle and a data
transfer key
with the public key.
nThe supplier sends the data back using &nbs=
p;
through the
proxy
nAdvantages:
nIt is difficult to infer the data handle through =
the
partial hash code=
nThe proxy alone cannot compromise the privacy
=
;
nThrough adjusting the revealed hash code, the all=
owable
error of the bloom fil=
ter can
be determined
13
Data
transfer procedure after improvement 1
![](3D"slide0018_image022.gif")
R: requester S: supplier
Step 1, 2=
: R sends out the =
partial hash code of the data=
handle
Step 3, 4=
: S sends the bloom filter of the handles and the=
public key certificates
<=
/span>
Step 5, 6=
: R sends the data handle and &nbs=
p;
encrypted by the public key
Step 7, 8=
: S sends the required data encrypted by
![](3D"slide0018_image025.gif")
Requester Proxy of=
&nbs=
p;
Supplier
&nbs=
p; =
Requester
14
Trust=
based
scheme – Improvement 2
nThe above schem=
e does
not protect the privacy of the supplier
nTo a=
ddress
this problem, the supplier can respond to a request via =
its
own proxy
15
Trust=
based
scheme – Improvement 2
![](3D"slide0021_image027.gif")
![](3D"slide0021_image028.gif")
Requester Proxy of &nbs=
p;
Proxy of &=
nbsp;
Supplier
&nbs=
p;
Requester &nbs=
p;
Supplier
16
Trustworthiness of peers
nThe trust value =
of a
proxy is assessed based on its behaviors and other peers’ <=
/span>recommendations
nUsing=
Kalman
filtering, the trust model can be built as a multivariate, tim=
e-varying
state vector
17
Expe=
rimental
platform - TERA
nTrust
enhanced role mapping (TERM) server
assigns roles to users based on
nUncertain & subjective evidences
nDynamic trust
n Reputation server
nDynamic trust information repository
nEvaluate
reputation from trust information by using algorithms spec=
ified
by TERM server
18
Trust
enhanced role assignment architecture (TERA)
![](3D"slide0022_image030.gif")
19
Conclusion
nA trust based priv=
acy
preservation method for peer-to-peer data sharing is =
proposed
nIt adopts the proxy
scheme during the data acquirement
nExtensions
nSolid analysis and experimen=
ts on
large scale networks are required
nA security analysis of the
proposed mechanism is required
20
Related publication
nB. Bhargava and Y. Zhong, “Authorization ba=
sed on
evidence and trust,”=
; in Proc. of International Conference on Data =
Warehousing and Knowledge Discovery (DaWaK), <=
/i>2002
nB. Bhargava, “Vulnerabilities and fraud in
computing systems,” =
in Proc. of International Conference on Advances =
in Internet, Processing, Systems, and
Interdisciplinary R=
esearch
(IPSI), 2003.
=
nL. Lilien and A. Bhargava, “From vulnerabil=
ities
to trust: A road to trusted
computing,” in Proc. of
International C=
onference
on Advances in Internet, Processing, Systems, and Interdisciplinary Research (IPSI), 2003.
nL. Lilien, “Developing pervasive trust para=
digm
for authentication a=
nd
authorization,” in Proc. of=
Third
Cracow Grid Workshop=
(CGW) , 2003.
=
=
=
‹=
header›
=
=
=
‹date/time›
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
=
=
=
‹=
footer›
=
=
=
‹#›
' + ENDSHOW_MESG + '
![](3D"notes_flag.gif")
N=
otes![](3D"buttons.gif")
![](3D"slide0015_image008.gif"){kind=link}
![](3D"slide0015_image009.gif"){kind=link}
![](3D"slide0016_image011.gif"){kind=link}
![](3D"slide0017_image013.gif"){kind=link}
![](3D"slide0017_image015.gif"){kind=link}
![](3D"slide0017_image017.gif"){kind=link}
![](3D"slide0019_image019.gif"){kind=link}
![](3D"slide0019_image020.gif"){kind=link}
![](3D"slide0018_image023.gif"){kind=link}
![](3D"slide0018_image024.gif"){kind=link}