ISSN: 2249-5789
P Subhadra Devi et al, International Journal of Computer Science & Communication Networks,Vol 8(4),25-31
Secure Cloud Data Storage with Efficient Key
Exposure
P. Subhadra Devi
M. Tech,
Department of CSE,
Shri Vishnu Engineering College for
Women (A),
Vishnupur, Bhimavaram, West
Godavari District, Andhra Pradesh.
Dr. P. Kiran Sree Ph.D.
Professor
Department of CSE
Shri Vishnu Engineering College for
Women (A),
Vishnupur, Bhimavaram, West
Abstract— The world just witnessed the surge of a
new adversarial model. We then propose Bastion, a
new and powerful attacker, which was able to coerce
primitive that is secure according to our definition
operators and acquire the necessary keys to break the
and, therefore, guarantees data confidentiality even
privacy of users. Once the encryption key is exposed,
when the encryption key is exposed, as long as the
the
data
adversary cannot compromise all storage servers. We
confidentiality is to limit the adversary’s access to the
analyze the security of Bastion, and we evaluate its
ciphertext. This may be achieved, for example, using
performance
multi-cloud storage systems. These systems spread
implementation.
data
only
viable
across
measure
multiple
to
servers
preserve
in
different
administrative domains, to cater for availability and
fault tolerance. If the adversary can only compromise
a subset of these domains, multi-cloud storage
systems may prevent the adversary from accessing
Godavari District, Andhra Pradesh.
by
means
of
a
prototype
Keywords— Key exposure, cryptography,
Security, Data confidentiality.
1
Introduction
the entire ciphertext. However, if data is encrypted
The world became aware of a massive surveillance
using existing encryption schemes, spreading the
program which mined data from operators and ISPs,
ciphertext on multiple servers does not entirely solve
and performed illegal taps on digital communication
the problem since an adversary which has the
channels. This surveillance program was not hindered
encryption key, can still compromise single servers
by the various security measures deployed within the
and decrypt the ciphertext stored therein. In this
targeted services. For instance, although these
paper, we leverage multi-cloud storage systems to
services relied on secure encryption mechanisms, the
provide data confidentiality against an adversary
necessary keying material was acquired, e.g., by
which has access to the encryption key, and can
means of backdoors, bribe, or coercion. In addition to
compromise a large fraction of the storage servers.
the public and governmental outrage, another
For this purpose, we first introduce a novel security
immediate reaction from the industry was an even
definition that captures data confidentiality in the
IJCSCN | August-September 2018
Available
25
ISSN: 2249-5789
P Subhadra Devi et al, International Journal of Computer Science & Communication Networks,Vol 8(4),25-31
larger apprehension to use third-party services, and in
block. For example, if each server stores at least two
particular cloud services.
ciphertext blocks, a (n− 2)ke secure scheme clearly
preserves data confidentiality unless all servers are
If the encryption key is exposed, the only viable
countermeasure is to limit the adversary’s access to
the ciphertext, e.g., by spreading it across multiple
administrative domains, in the hope that the
adversary cannot compromise all of them. However,
compromised, even when the adversary has access to
the encryption key. We analyze the security of
Bastion, and we compare its performance in a
realistic implementation setup with a number of
existing encryption schemes.
this countermeasure does not entirely solve the
problem. Even if the data is encrypted and dispersed
We consider a multi-cloud storage system
across different administrative domains, an adversary
which can leverage a number of commodity cloud
equipped with the appropriate keying material can
providers with the goal of distributing trust across
compromise a single server and decrypt ciphertext
different administrative domains. This “cloud of
blocks stored therein. In this paper, we leverage
clouds” model is receiving increasing attention
multi-cloud
data
nowadays with leading cloud storage providers such
confidentiality against an adversary which knows the
as EMC, IBM, and Microsoft, offering products for
encryption key, and can compromise a large fraction
multi-cloud systems. In particular, we consider a
of the storage servers. The adversary can acquire the
system of s storage servers S1, . . . , Ss, and a
keys either by exploiting flaws or backdoors in the
collection of users. We assume that each server
key-generation software , or by compromising the
appropriately authenticates users.
storage
systems
to
provide
devices that store the keys (e.g., at the user-side or in
the cloud). As far as we are aware, this adversary
2
Related Works
i)
Enhancing
invalidates the security of most cryptographic
solutions, including those that protect the keys by
means of secret-sharing (since the keys are leaked at
generation time).
Data
Security
In
Cloud Storage Auditing With Key
Abstraction
In this work, we propose Bastion, an efficient
primitive that requires only one round of block cipher
encryption, followed by a linear transformation.
Bastion is (n− 2)ke secure, i.e., it ensures that
plaintext data cannot be recovered as long as the
adversary has access up to all but two ciphertext
blocks, even when the encryption key is exposed. As
such, Bastion relaxes the notion of all-or-nothing at
the benefit of improved performance. This is
reasonable since, in a multi-cloud storage system,
each server is likely to store more than one ciphertext
IJCSCN | August-September 2018
Available
We
have
examined
working
framework plan for information security in
distributed
storage.
we
talked
about
engineering segments for giving information
security at the two levels (User furthermore,
Administrator). To guarantee the rightness
of client's information in cloud information
stockpiling, we proposed a powerful and
adaptable
appropriated
conspire
with
unequivocal powerful information bolster,
26
ISSN: 2249-5789
P Subhadra Devi et al, International Journal of Computer Science & Communication Networks,Vol 8(4),25-31
counting square refresh, erase, and attach
initially as opposed to applying the visually
tasks. For information security we have
impaired strategy amid the reviewing
utilized
procedure.
DES
which
lets
Discovering
that
the
information put away in the database as
information
figure content and on ask for information is
remain online practically speaking, with a
accessible in the required organization. We
specific end goal to keep the capacity
depend on erasure correcting code in the
accessible what's more, certain after a
document dissemination planning to give
noxious defilement, we present a semi-
excess equality vectors and certification the
confided
information steadfastness. By using the
framework show and give a benefit to the
Homomorphic token with dispersed check
intermediary to deal with the reparation of
of eradication coded information, our plan
the coded squares and authenticators. To
accomplishes the mix of capacity rightness
better correct for the recovering code-
protection
and
information
situation, we outline our authenticator in
limitation,
i.e.,
at
whatever
blunder
proprietor
in
can't
intermediary
generally
into
the
point
view
been
authenticator can be proficiently produced
distinguished amid the capacity rightness
by the information proprietor all the while
confirmation over the appropriated servers,
with the encoding methodology. Extra
we can nearly ensure the synchronous
investigation demonstrates that our plan is
distinguishing
provable
information
ii)
calculation,
debasement
proof
of
has
the
acting
of
the
BLS
secure,
and
signature.
the
This
execution
mischievously server(s). We have utilized
assessment demonstrates that our plan is
DES calculation with eradication adjusting
profoundly viable t and can be plausibly
procedure for giving information security
incorporated into a recovering code-based
respectability.
distributed storage framework.
An Efficient Cloud Storage Batch
Auditing Without Key Exposure
iii)
Privacy-preserving public auditing
for secure cloud storage
Resistance Using Public Verifier
Utilizing distributed storage, clients
From this the giving safeguarding
open examining to recovering code based
distributed storage has been actualized.
Where the information proprietors are
special to assign TPA for checking entitled
their information legitimacy. To secure the
first information protection against the
TPA,
we
randomize
IJCSCN | August-September 2018
Available
the
coefficients
can remotely store their information and
appreciate
the
on-request
amazing
applications and administrations from a
common pool of configurable processing
assets, without the weight of nearby
information stockpiling and support. In any
case, the way that clients never again have
physical
ownership
of
the
outsourced
27
ISSN: 2249-5789
P Subhadra Devi et al, International Journal of Computer Science & Communication Networks,Vol 8(4),25-31
information
makes
the
uprightness
security
information
in
information
distributed
in
outsourcing
stockpiling administration. In this
computing an imposing errand, particularly
paper, we propose an agreeable
for clients with obliged registering assets. In
addition, clients ought to have the capacity
provable
to simply utilize the distributed storage as
conspire in half and half mists to
though
without
help adaptability of administration
agonizing over the need to confirm its
and information movement, in which
it
is
neighborhood,
trustworthiness. In this manner, empowering
information
ownership
we consider the presence of various
open auditability for distributed storage is of
basic significance with the goal that clients
cloud specialist organizations to
can turn to an outsider inspector (TPA) to
helpfully store and keep up the
check the trustworthiness of outsourced
customers' information. Our tests
information and be straightforward. To
demonstrate that the check of our
safely
present
a
powerful
TPA,
the
plan requires a little, steady measure
evaluating procedure ought to acquire no
new
vulnerabilities
toward
of
client
overhead,
which
limits
correspondence many-sided quality.
information protection, and acquaint no
extra online weight with client. In this paper,
we propose a safe distributed storage
framework supporting security safeguarding
open reviewing. We additionally stretch out
our outcome to empower the TPA to
perform reviews for numerous clients at the
same time and effectively. Broad security
and execution examination demonstrate the
proposed plans are provably secure and very
proficient. Our primer analysis directed on
Amazon
EC2
occurrence
additionally
exhibits the quick execution of the plan.
iv)
Efficient provable data possession
for hybrid clouds
Provable
ownership
is
information
method
Algorithms
Algorithm 1 Encryption in Bastion:1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
procedure Enc(K, x = x[1] . . . x[m])
n = m + 1, Storage blocks : n
y′[n] {0, 1}l ⊲ y’[n] is the IV for CTR
for i = 1 . . . n − 1 do
y′[i] = x[i] FK(y′[n] + i)
end for
t = 0l
for i = 1 . . . n do
t = t y′[i]
end for
for i = 1 . . . n do
y[i] = y′[i] t
end for
return y ⊲ y = y[1] . . . y[n]
end procedure
for
A plaintext of m blocks, the CTR
guaranteeing the respectability of
encryption mode outputs n = m + 1
IJCSCN | August-September 2018
Available
a
3
28
ISSN: 2249-5789
P Subhadra Devi et al, International Journal of Computer Science & Communication Networks,Vol 8(4),25-31
ciphertext blocks. computed with (n− 1)
Algorithm
block cipher operations and (n
− 1) XOR
Algorithm:-
operations. The CTR encryption mode is
secure.
Polynomial
Time
input:n: number keys
Algorithm 2 Decryption in Bastion:1. procedureDec(K, y = y[1] . . . y[n])
,K ∈ {0, 1}k
2. t = 0l
3. for i = 1 . . . n do
4. t = t y[i]
5. end for
6. for i = 1 . . . n do
7. y′[i] = y[i] t
8. end for
9. for i = 1 . . . n − 1 do
10. x[i] = y′[i] F−1
11. K (y′[n] + i)
12. end for
13. return x ⊲ x = x[1] . . . x[n − 1]
14. end procedure
3
output:Polynomial keys
1.
2.
3.
4.
5.
6.
7.
8.
proceed(an , x[n] = x[1] . . . x[n])
if x not null
for x[i] in Xn do i ∈n
an = randomValue(); // to generate
random values
expireAfterWrite(10,
TimeUnit.MINUTES);
end for
end if
return an ;
Any party with access to all the
ciphertext blocks and the encryption key can
recover the plaintext. This requirement is
essential
subsequent
for
the
linear
correctness
of
the
transform
on
the
ciphertext blocks. That is, if m is even, then
the transform is not invertible. l is the block
size of the particular block cipher used. 0l
and 1l denote a bit-string of l zeros and a bitstream of l ones, respectively.
IJCSCN | August-September 2018
Available
29
ISSN: 2249-5789
P Subhadra Devi et al, International Journal of Computer Science & Communication Networks,Vol 8(4),25-31
4
Results
No. of File Character
No of different Characters
200
180
160
140
120
100
80
60
40
20
0
NO. Of character
1
0
2
40
3
80
4
120
5
160
No of different characters
0
50
100
150
200
Fig: Performance Analysis Graph
NO. Of character
No of different characters
0
0
40
50
80
100
120
150
160
200
Table: Performance Analysis
The graph is drawn between the
the system that has been proposed by the
number of file character that are being used
author. Uses the characters on completion of
for the encryption and decryption v/s
the rotation, this makes the algorithms to
number of different characters that are using
take little less character than this proposed
by the algorithm. the algorithm used in this
method.
system takes more characters to replace than
IJCSCN | August-September 2018
Available
30
ISSN: 2249-5789
P Subhadra Devi et al, International Journal of Computer Science & Communication Networks,Vol 8(4),25-31
5
Conclusion
Advances in Cryptology (CRYPTO), 1998, pp. 390–
407.
In this paper, we tended to the issue of
[4] C. Basescu, C. Cachin, I. Eyal, R. Haas, and M.
securing information outsourced to the cloud against
Vukolic, "Ground-breaking Information Sharing with
an enemy, which approaches the encryption key. For
Key-regard Stores," in ACM SIGACTSIGOPS
that reason, we presented a novel security definition
Symposium on Principles of Distributed Computing
that catches information privacy against the new
(PODC), 2011, pp. 221– 222.
adversary. We at that point proposed Bastion, a plan
[5] A. Beimel, "Riddle sharing designs: A survey," in
that
International Workshop on Coding and Cryptology
guarantees
the
classification
of
encoded
information notwithstanding when the enemy has the
(IWCC), 2011, pp. 11– 46.
encryption key, and everything except two figure
[6] A. Bessani, M. Correia, B. Quaresma, F. André,
content blocks. Bastion is most appropriate for
and P. Sousa, "DepSky: Dependable and Secure
settings where the cipher text blocks are put away in
Storage in a Cloud-ofclouds," in Sixth Conference on
multi-Cloud storage frameworks. In these settings,
Computer Systems (EuroSys), 2011, pp. 31– 46.
the foe would need to get the encryption key and to
[7] G. R. Blakley and C. Glades, "Security of incline
bargain all servers, keeping in mind the end goal to
designs," in Advances in Cryptology (CRYPTO),
recoup any single piece of plaintext. We broke down
1984, pp. 242– 268.
the security of Bastion and assessed its execution in
[8] V. Boyko, "On the Security Properties of OAEP
sensible settings. Bastion consider capably enhances
as an Allor-nothing Change," in Advances in
the execution of existing natives which offer
Cryptology (CRYPTO), 1999, pp. 503– 518.
practically identical security under key presentation.
[9] R. Canetti, C. Dwork, M. Naor, and R. Ostrovsky,
6. Reference
"Deniable Encryption," in Proceedings of CRYPTO,
1997.
[1] M. Abd-El-Malek, G. R. Ganger, G. R. Goodson,
[10]
M. K. Reiter, in addition, J. J. Wylie, "Reprimand
/>
Scalable Byzantine Fault-Tolerant Administrations,"
[11] C. Charnes, J. Pieprzyk, and R. Safavi-Naini,
in
"Prohibitively secure secret granting plans to
ACM
Symposium
on
Operating
Systems
Cavalry,
"Encryption
Engine
Dongle,"
Principles (SOSP), 2005, pp. 59– 74.
disenrollment capacity," in ACM Conference on
[2] M. K. Aguilera, R. Janakiraman, and L. Xu,
Computer and Communications Security (CCS),
"Using Erasure Codes Efficiently for Storage in a
1994, pp. 89– 95.
Distributed System," in Universal Conference on
[12] A. Desai, "The security of win enormous or bust
Dependable Systems and Networks (DSN), 2005, pp.
encryption: Protecting against far reaching key
336– 345.
request," in Advances in Cryptology (CRYPTO),
[3] W. Aiello, M. Bellare, G. D. Crescenzo, and R.
2000, pp. 359– 375.
Venkatesan, "Security strengthening by piece: The
case of doublyiterated, culminate figures," in
IJCSCN | August-September 2018
Available
31