International Journal of Network Security, Vol.21, No.1, PP.137-144, Jan. 2019 (DOI: 10.6633/IJNS.201901 21(1).17) 137
Multimedia Social Network Authorization
Scheme of Comparison-based Encryption
Cheng Li, Zhiyong Zhang, Guoqin Chang
(Corresponding author: Zhiyong Zhang)
School of Information Engineering, Henan University of Science and Technology,
Luoyang, Henan 471023, China
(Email: xidianzzy@126.com)
(Received Nov. 9, 2017; revised and accepted Mar. 5, 2018)
Abstract
In many Ciphertext-Policy Attributed Based Encryption
(CP-ABE) schemes, the level of attributes is ignored;
while the comparison based attribute encryption scheme
is not flexible enough. In this paper, an encryption scheme
based on comparative attributes is proposed. In this
scheme, users can’t only make more granular and flexi-
ble access control policies based on the level of attributes,
but also support more diverse forms of access control pol-
icy. At the same time, in order to solve the computational
pressure of the user terminal, a third-party proxy is added
to the solution to assist the user to decrypt the cipher-
text. Through the comparative and experimental data
analysis, the scheme can be better applied to multimedia
social networks.
Keywords: Comparison-Based Attribute; CP-ABE; Mul-
timedia Social Networks; Third-Party Decrypt
1 Introduction
With the development of multimedia social networks,
more and more people are willing to publish their per-
sonal life and privacy to the multimedia social network.
But the security problems caused by privacy leaking and
data authorization of social network users (hereinafter re-
ferred to as ”users”) are followed. The user uploads his or
her private data to a social network service provider, such
as the health condition of the user, travel information, and
payment (consumption) information, via a social network
provider or a third party storage agent to save the user
data. However, the social network provider and third-
party storage (or ”Cloud”) are often untrustworthy, and
they are likely to spy on the user’s private data or to leak
privacy data due to problems such as failures and mali-
cious user attacks, which leads to unnecessary problems
to the user.
In order to protect the user’s privacy, the user can en-
crypt the encrypted data and then upload the ciphertext
to the cloud; then the user uses a flexible authorization
method to share the encryption key, while users can also
specify a fine-grained access control strategy to achieve
efficient and secure data authorization. Sahai and Waters
first proposed attribute based encryption (ABE) scheme
in the [1], which can achieve fine-grained one to many au-
thorization. ABE encrypted data can not only ensure
the security and integrity of user data, but also have
good flexibility. In 2006, Goyal proposed a Key-Policy
Attributed Based Encryption (KP-ABE) and Ciphertext-
Policy Attributed Based Encryption (CP-ABE) in the [2],
and implemented the first KP-ABE algorithm. In 2007,
Bethencourt [3] and Cheung [4] implemented the CP-
ABE algorithm, respectively. After that, with the con-
tinuous development of ABE technology, has been widely
used in multimedia social networks [5–7], cloud comput-
ing [8,9], cloud storage [10,11] and electronic health man-
agement [12,13] and many other areas. At the same time,
users can use the ”Boolean expression” [14], ”and/or” ac-
cess structure [6], (t, n) threshold [15,16] and linear secret
sharing scheme (LSSS) [17], constructing a relatively flex-
ible access control policy to meet the user’s needs.
However, the current attribute-based encryption au-
thorization scheme is often used to use specific attributes,
such as the access strategy ”President AND July 1”,
which states that ”only the president has access rights
in July 1st”, in other words, Other people in this time
or ”President” in addition to this time cannot access the
data, although this is a relatively extreme example, but
it does show that most proposal is not flexible, because
”President” can be divided into ”president” and ”vice-
president”, and even more detailed division, the date is
the same reason, for these can be refined attribute autho-
rization program research is relatively less.
Therefore, the attributes can be divided into sub-
attributes according to a certain order, making data au-
thorization more in line with the actual needs. The or-
der relationship between these sub-attributes can be com-
pared. Only when the user attribute level satisfies the
access authorization policy can the data be decrypted.