Journal of Information & Computational Science 11:17 (2014) 6221–6228
Available at http://www.joics.com
November 20, 2014
A Data-encrypted Distributed Simulation Framework ?
Hongqian Chen a,b , Fengxia Li b,∗, Yuehong Sun a , Li Liu a
a School
of Computer and Information Engineering, Beijing Technology and Business University
Beijing 100048, China
b Beijing
Key Laboratory of Intelligent Information Technology, Beijing Institute of Technology
Beijing 100081, China
Abstract
To add the feature of data security for the distributed simulation system, the paper propose an dataencrypted distributed simulation framework. The framework introduces the encryption module into the
HLA simulation system. The framework adds an encryption module in every federator to obtain more
security in data transportation. The encrypt module adopt the chaotic key and the composite discrete
chaotic system as encryption scheme. The chaotic system can produce continuous acyclic encrypt key
for every data transportation during the whole of simulation. The encryption algorithm can achieve
self-synchronization due to its own feature. This framework provides data-protected service for each
group of federators with different authority. This framework has been proved feasible and effective.
Keywords: Distributed Simulation; Data Security; Composite Discrete Chaotic System; Distributed
Network
1
Introduction
With the development of the scale and complexity in distributed simulation system, the security
of the data transportation processing in the simulation system is one of the most important
topics [1]. Encryption is the popular method to achieve the security purpose currently [2]. The
simulation system set different key which used to decrypt the cipher data within its authority
for each member. But the number of key rises rapidly with the members in simulation system
increase [3]. It will be very difficulty to manage huge number of keys.
The Chaotic system provides a platform for producing infinite erratic output stream. The chaos
system is suit for creating continuous acyclic encrypt key in distributed simulation because of its
complex kinetics feature.
?
Project supported by the Science and Technology Project of Beijing Municipal Commission of Education (No. PXM2014-014213-000004). Twelfth Five-Year National Science and Technology Support Project
(No. 2012BAD29B01-2) and Funding Project for Innovation on Science, Technology and Graduate Education
in Institutions of Higher Learning under the Jurisdiction of Beijing Municipality (No. PXM2013-014213-00003000042300).
∗
Corresponding author.
Email address: lfx [email protected] (Fengxia Li).
1548–7741 / Copyright © 2014 Binary Information Press
DOI: 10.12733/jics20104952
6222
H. Chen et al. / Journal of Information & Computational Science 11:17 (2014) 6221–6228
We focus on introducing the chaos-encryption system to the High Level Architecture (HLA).
The framework designs an encrypted module for federators, which encrypt any input or output
stream for its host. The security of the simulator is obtained via logic and flexible federation
construction. The framework can encrypt the data using a synchronous updating stream key
while simulation processing. All interfaces used in the framework comply with the standard HLA
interface specification.
The remainder of this paper is organized as follows: In Section 2, an overview of the HLA
architecture, the encryption and related work is given. Section 3 describes in detail this distributed
simulation framework design and technical issues arising from this approach. Section 4 presents
the analysis of the simulation framework. Finally, Section 5 summarizes the framework.
2
Related Work
Data exchange and other services in the HLA framework are realized by the Runtime Infrastructure (RTI). The RTI provides lots of services to support simulations and to carry out federateto-federate interactions. Each of federator in the federation can access the transferring data via
RTI services. But it can not prevent data from being revealed in the public network. The data
could be exposited un-secretly to non-authority federator in the HLA architecture.
Wang [4] designed a distributed simulation architecture based on mobile agents in the existing
distributed simulation architecture based on HLA. A new message-oriented active communication
mechanism is adopted to ensure the communication among mobile agents reliably and efficiently.
Iazeolla [5] introduces a new distributed simulation environment and a new distributed simulation
language to help implement the wireless distributed system.
The security is one of the most important topics in communication in network. Anagnostou
[6] presents a distributed hybrid agent-based discrete event simulation model for the distributed
system. Both the ABS and the DES models were developed to achieve communication between
models. Ceccarelli [7] apply the biometrics in the management of sessions via a secure protocol
for perpetual authentication through continuous user verification.
Park [8] replace transport layer of HLA-based distributed simulation system using Data Distribution Service. The wrapper API is adopted to achieve the network control mechanism for data
transmission in distributed system. Nouman [9] developed an agent-based model of an ambulance
service based on the high level architecture technology.
Brito [10] proposes a solution for embedded modeling and simulation of heterogeneous models of
computation in a distributed way. Pfeifer [11] presents an expression-level distributed simulation
system to observe a decrease the transient analysis time. Duan [12] proposes some management
and control techniques for the building and executing to enhance the efficiency of HLA-based
distributed simulation system.
Chaos is one form of nonlinear dynamic system [13]. It is pseudo-random phenomena between
certainty and randomness. It has some interesting features as following. A) It is extremely
sensitive to initial condition and parameters. B) It has topological transitivity, which means
high randomness. C) The nonlinear system where chaos located is a certain system, which states
the system has strong certainty and regularity. The features of chaos are match closely to the
requirement in cryptology.
H. Chen et al. / Journal of Information & Computational Science 11:17 (2014) 6221–6228
6223
The paper proposed a framework for distributed simulation system. The framework has the
following feature comparing to the traditional one. A) This framework can achieve the accesscontrolling by authority for members. The encryption function can integrate into the federators
as a module. B) The chaos system automatic reproduces the stream encrypt key along with
advancing of simulation, while the traditional encryption scheme need to the management and
distribution of encrypt key. C) The chaos-based encryption algorithm works on the continuous
real numbers set, while the traditional algorithm works on the discrete integer number set.
3 Distributed Simulation Framework with Data-encrypted
Transport Module
Encryption is the popular method of preventing illegal accessing to data. The fitting encryption
method in simulation should achieve the following features: A) The encrypt algorithm is suitable
with various block size data. B) The algorithm is efficient to avoid affect the simulation processing.
The real time interaction is required generally among federators in simulation. C) The algorithm
has special encryption key mechanism to prevent being decoded during the simulation.
According to the requirement in the encryption in distributed simulation system, this paper
proposes the chaos-based encryption method. The method generates the chaos sequence which
can be transformed into binary value via chaos iteration. The binary value sequence was combined
into chaos key.
The chaos key can be synchronously updated during simulation via chaos iteration. The chaosbased algorithm can achieve the aim of one-time pad encryption because of the updating chaos key
[14]. Each federator can obtain the same chaos key at the same time by the time-synchronization
in the simulation framework.
The procedure of the chaos-based encryption method can be described as Fig. 1.
Initial
condition
Chaotic
iteration
Clipper
data
Key
XOR
Chaos
key
Fig. 1: The sketch map of chaos-based encryption method
The chaos system accomplishes its initialization according to user’s setting. The chaos-based
algorithm encrypts data using the chaos key in each data transportation processing. The chaos
system rebuilds the encryption key when the simulation system advanced.
3.1
Generating the Logistic-based Encryption Key
Logistic mapping equation can be expressed as Eq. (1)
xn+1 = µ · xn (1 − xn )
(1)
6224
H. Chen et al. / Journal of Information & Computational Science 11:17 (2014) 6221–6228
The equation has the certainly form and includes none of randomness. The iterate result is
extremely sensitive to the mapping parameter u. When 3.5699456 . . . ≤ µ ≤ 4, the result of each
iteration will be various extremely, which is the irregular chaos phenomenon.
The distributed function ρ(x) of the Logistic sequence xi can be expressed as Eq. (2)
ρ(x) =

 √
π
1
x(1−x)

0<x<1
0
other
(2)
The average x of the sequence xi can be calculated by Eq. (3)
Z 1
N −1
1 X
x¯ = lim
xi =
xρ(x)dx
N →∞ N
0
i=0
(3)
In the Eq. (3), N states the length of the selected sequence produced by Logistic iteration.
The stream encryption key in our simulation framework can be obtained via combining binary
sequence value. The binary value can be calculated by Eq. (4)
(
λi =
3.2
1 xi ≥ x¯
(4)
0 xi < x¯
Composite Discrete Chaotic-based Encryption
The iteration in the composite discrete chaotic system [16] depends on both the frontier iteration
result and the plain data. The theory of the composite discrete chaotic can be described as Fig. 2.
Plain data
Composite
discrete
chaotic
Chaos key
Encryption
algorithm
Clipper data
Fig. 2: The encryption based on the composite discrete chaotic system
There are two iteration equations f0 (x) and f1 (x) in one composite discrete chaotic system. It
select one equation f0 (x) or f1 (x) to execute depends on the plain data and the frontier iteration
result in each step of iteration.
The executing of composite discrete chaotic iteration n depends on the plain data mn−1 and
the frontier iteration xn−1 . In the other word, the Ciphers data can not be decrypted if have not
the frontier part of the data. The feature can take more security when transport.
H. Chen et al. / Journal of Information & Computational Science 11:17 (2014) 6221–6228
6225
The iteration formula of the composite discrete chaotic can be described as Eq. (5).
(
xn+1 =
f0 (xn ), m = 0
(5)
f1 (xn ), m = 1
In Eq. (5), m is the current bit of plain data, f0 (x) is the corresponding iteration equation for
the current bit 0, while f1 (x) is corresponding to the current bit 1. They can show respectively
as Eq. (6) and Eq. (7).
(
√
1 − 1 − 2x, 0 ≤ x < 21
f0 (x) =
(6)
√
1
2x − 1,
≤
x
<
1
2
( √
1 − 2x,
0 ≤ x < 21
f1 (x) =
(7)
√
1 − 2x − 1, 21 ≤ x < 1
3.3
Chaos Synchronization Among Federators
In the distributed simulation system, the chaos key should be consistent between sender and
receiver. If they hold different key, the receiver can not decode the data receiving from the
sender.
The chaos keys are generated on their own federator. As a result, they must have the same
iteration sequence among federators at the same time.
There are two discrete chaotic systems in sender federator and receiver federator. The system
executed in sender federator can be expressed as Eq. (8)
(8)
X(k + 1) = F (X(k))
In the Eq. (8), the function X and function F can be processed by Eq. (9)
X(k) = (x1 (k), x2 (k), . . . xn (k))T ∈ Rn
F (X(k)) = (f1 (X(k)), f2 (X(k)), . . . fn (X(k)))T
(9)
The system executed in receiver federator can be expressed as Eq. (10)
(10)
Y (k + 1) = G(Y (k), Xm (k))
In the Eq. (10), the function Y and function G can be processed by Eq. (11)
Y (k) ∈ Rm

g1 (Y (k), Xm (k))

 g2 (Y (k), Xm (k))
G (Y (k), Xm (k)) = 

···

gm (Y (k), Xm (k))






(11)
6226
H. Chen et al. / Journal of Information & Computational Science 11:17 (2014) 6221–6228
There is one mapping H : Rn → Rm and one number set B = Bx × By ⊂ Rn × Rm . The two
systems have the initial condition (X(0), Y (0)) ∈ B . The system in Eq. (10) has the solution
(X(k), Y (k)). If the solution can satisfy Eq. (12), the two systems can reach the chaotic selfsynchronization.
°
°
lim °Xm (k) − H −1 (Y (k))° = 0
(12)
k→∞
The chaotic encryption algorithm based on self-synchronization is unsymmetrical encryption
system. It adopts two difference chaotic systems between sender and receiver. The sender federator and the receiver federator can adopt difference encryption key, which obtain higher security
than symmetrical encryption system.
The sender federator can encrypts the plain text using x(k) sequence, and the receiver federator
should use the corresponding v(Y (k)) sequence to decrypt the cryptograph text.
3.4
Distributed Simulation Framework with Data-encrypted Transportation
The encryption function is designed to integrate into the federators as a module. In the encryption
module-based framework, the federator modifies its source code to add the encryption module on
basis of original simulation system. All of the data are encrypted by the module when federator
sends data. And all the receiving data also are decrypted by this module before using by the
federator. The structure of module-based framework can be described as Fig. 3.
Synchronization mechanism
Federator
Federator
Chaos
iteration
Chaos key
Chaos
iteration
Encryption
Encryption
Chaos key
RTI
Fig. 3: The structure of encryption module-based framework
The framework based on encryption module has high security and have not to modify the logical
relation among the federators. The path of data transferring also has not to be redesigned. The
OMT (Object Model Template) files should be modified before simulation.
4
Application and Analysis
We realize a small distributed simulation system base on the framework proposed in this paper.
The federators in the system have the chaos encryption module. The system encrypts the sim-
H. Chen et al. / Journal of Information & Computational Science 11:17 (2014) 6221–6228
6227
ulation data using the chaos key. The experimental environment includes Intel Core2 3.0 GHz
CPU, 4 GB RAM, Windows 7 OS. The analysis result of the chaos encryption and decryption
processing can be shown as Table 1. The analysis includes the average time cost, the key space
and the key distribution.
Table 1: The analysis result of distributed simulation
Cost Time
Key Space
Key Distribution
35.827 ns
1016+15
0.57
The result demonstrates that the framework can qualified the encryption requirement in the
distributed simulation system. The simulation data can be accurately encrypted and decrypted
at sender and receiver respectively. The time cost of chaos encryption and decryption is less than
100 ns.
5
Conclusions
This paper proposes a distributed simulation framework with data encryption function. The
framework obtain more security in data transportation via adding a encrypt module in every
federator. The framework has some advantages as following. 1) The encryption Scheme is designed
according to the feature of the data security in the system. The Logistic-based algorithm can
suit for encrypting any length simulation data. 2) The encryption algorithm can achieve selfsynchronization due to adopting the composite discrete chaotic-based system. The continuous
acyclic encrypt key can provide higher security than symmetrical encryption system. 3) The
encrypted module integrated into the federator can keep the original logical relation among the
federators and data transferring path. This framework has been proved utility and feasible and
can provides data-protected service for each group of federators.
References
[1]
Richard M. Fujimoto, Parallel simulation: Parallel and distributed simulation systems, Proceedings of the 33rd Conference on Winter Simulation, Arlington, Virginia, 2001, 147-157
[2]
N. Smaoui, A. Karouma, M. Zribi, Secure communications based on the synchronization of the
hyperchaotic Chen and the unified chaotic systems, Communications in Nonlinear Science and
Numerical Simulation, 16(8), 2011, 3279-3293
[3]
R. Ewald, J. Himmelspach, A. M. Uhrmacher, An algorithm selection approach for simulation
systems, 22nd Workshop on Principles of Advanced and Distributed Simulation, PADS’08, 2008,
91-98
[4]
Cheng Wang, Hua Wu, Xiyu Pang, Guangyuan Zhang, The application research of mobile agent to
distributed simulation, International Conference on Computer Science and Service System (CSSS),
Nanjing, 2012, 1158-1161
[5]
G. Iazeolla, A. Pieroni, D’Ambrogio Andrea et al., A distributed approach to wireless system
simulation, Sixth Advanced International Conference on Telecommunications (AICT), Barcelona,
2010, 252-262
6228
H. Chen et al. / Journal of Information & Computational Science 11:17 (2014) 6221–6228
[6]
A. Anagnostou, A. Nouman, S. J. E. Taylor, Distributed hybrid agent-based discrete event emergency medical services simulation, Simulation Conference (WSC), Washington, D. C., 2013, 16251636
[7]
A. Ceccarelli, L. Montecchi, F. Brancati et al., Continuous and transparent user identity verification for secure internet services, IEEE Transactions on Dependable and Secure Computing,
V(PP), Issue 99, 2014, 1-2
[8]
Yunjung Park, Dugki Min, Development of HLA-DDS wrapper API for network-controllable distributed simulation, The 7th International Conference on Application of Information and Communication Technologies (AICT), Baku, 2013, 1-5
[9]
A. Nouman, A. Anagnostou, S. J. E. Taylor, Developing a distributed agent-based and des simulation using portico and repast, IEEE/ACM 17th International Symposium on Distributed Simulation and Real Time Applications (DS-RT), Delft, 2013, 97-104
[10] A. V. Brito, A. V. Negreiros, C. Roth et al., Development and evaluation of distributed simulation
of embedded systems using ptolemy and HLA, IEEE/ACM 17th International Symposium on
Distributed Simulation and Real Time Applications (DS-RT), Delft, 2013, 189-196
[11] D. Pfeifer, A. Gerstlauer, Expression-level parallelism for distributed spice circuit simulation,
IEEE/ACM 15th International Symposium on Distributed Simulation and Real Time Applications
(DS-RT), Salford, 2011, 12-17
[12] Wei Duan, Yuanzheng Ge, Xiaogang Qiu, Management and control techniques for distributed
simulation system, Fifth International Conference on Frontier of Computer Science and Technology
(FCST), Changchun, Jilin Province, 2010, 9-16
[13] M. Eisencraft, R. D. Fanganiello, J. M. V. Grzybowski, D. C. Soriano et al., Chaos-based communication systems in non-ideal channels, Communications in Nonlinear Science and Numerical
Simulation, 17(12), 2012, 4707-4718
[14] Qiyuan Liu, Youxin Luo, Bin Zeng, The research of composite nonlinear discrete chaos dynamical
systems to mechanism synthesis, Advanced Materials Research, 2011, 204-210
[15] Ashraf A. Zaher, Abdulnasser Abu-Rezq, On the design of chaos-based secure communication
systems, Communications in Nonlinear Science and Numerical Simulation, 16(9), 2011, 3721-3737