Secure Communication in Dynamic Wireless Sensor Network by

Effective Key Management

first author1, second author2
1
PG Student, Department of Computer Science and Engineering,
2
Professor, Department of Computer Science and Engineering,
Poojya Doddappa Appa College of Engineering, Kalaburagi, Karnataka, India

Abstract— Nowadays sensor devices created much kind of flexible and scalable than a static wireless sensor network.
applications for hacking details from one system. Encrypting Elliptical Curve Cryptography (ECC) in existing were
techniques were used to secure the data. Malicious attacks were exposed to forgery of messaging details, compromised key
mostly exposed in sensor devices impersonation, cloning, and and known key attacks. But in dynamic WSN security were
interception, physical destruction, lapses of connectivity in provide authentication and access control for resources, so the
wireless network communication. Here Certificateless Effective internet works on totally trust one another. By utilizing
Key Management (CL-EKM) scheme were used for dynamic
certificateless Hybrid Signcryption scheme can generate
wireless sensor networks. Certificateless Effective Key
Management was used for the purpose of creating secure pairwise key between nodes by building CLEKM. CL-EKM is
pairwise node communication and group oriented key a light weight processes support mobile nodes and updates
communication within clusters. This key management cluster key during the movement occurred in node between
procedures were used to secure the node when it move across clusters. When the node is detected by the attacker or leave
different clusters and key revocation process for compromised from the cluster permanently, the method of key revocation is
node.CL-EKM protect against the various attacks when the possible.
sensor node gets communicated by maintaining the key
efficiently provided to it.
II. PROBLEM STATEMENT
Keywords— Wireless sensor networks, Mobile nodes are not feasible in symmetric key schemes and
certificateless public key cryptography, key
its past is based only on the static wireless sensor network. To
management scheme.
support dynamic wireless sensor network only few approaches
were based only on the Public Key Cryptography
I. INTRODUCTION (PKC).Based on Diffie-Hellman (DH) the existing system
uses two layered key management scheme and a dynamic key
Dynamic wireless sensor network allow large area of network
update protocol. In order to store the pairwise key the large
applications for the mobility of the sensor nodes and more
memory space is needed and suffers from the overhead of
exact than static wireless sensor networks. It includes
high communication. Impossible to support mobile nodes
applications of battlefield surveillance from the tracking of the
applications and won’t flexible against the attackers. It is not
target from the opposing team, medical report, vehicle status
suitable for the sensor with small resources and large key size
monitoring and dairy health care centre. Malicious attacks
computations were not performed.
were exposed by impersonation, cloning, duplication of files,
But here, goal is to providing node authentication and
interception in sensor devices. In dynamic wireless sensor
generating pairwise key to the nodes dynamically. A scheme
applications security is the main issues. The security
provided for the secure node movements across different
requirements were provided in dynamic wireless sensor
clusters, cloning, impersonation, and interception also ensures
network such as node authentication, data confidentiality, and
forward and backward secrecy. Certificateless Key
integrity and security were provided when the node gets
Management Scheme (CL-EKM) used for types of keys such
movement from its cluster. The security were get addressed by
as certificateless Public /private key, individual key, pairwise
the key management protocols for the dynamic wireless
key and cluster key. In deriving certificateless private/public
sensor networks. In existing, mobile sensor nodes were not
key and pairwise keys CLEKM uses CL-HSC scheme. By the
feasible in symmetric key encryption and get focused only on
use of CL-EKM scheme the compromised node won’t affect
the static wireless sensor network which leads to limited
the security key from the valid one. It support node
energy and handing out capabilities. So communication gets
authentication, replication and impersonation attacks also
overhead which requires large memory space .It uses two
ensures forward and backward secrecy. This scheme is secure
layered key management and dynamic key update protocols.
against the Known- Key attacks of the dynamic wireless
Not scalable and not flexible against the attacker nodes.
sensor networks.
Symmetric encryption scheme were not used in dynamic
wireless sensor. Asymmetric encryption key were mostly used
in dynamic WSN. By using Public Key Cryptography the key III. RELATED WORK
has been generated for each node for the security between the
node movements. Public Key Cryptography (PKC) is more Mobile nodes are not feasible for Symmetric key schemes [1],
the symmetric schemes are used only for the static nodes. A
small number of approaches were used for the dynamic sensor sends the encrypted message securely using pairwise
wireless sensor networks which based only on the Public Key key to the cluster. H sensor encrypts that required message to
Cryptography (PKC) [1]. In order to store the pairwise key it the cluster with the pairwise key.
requires a large number of spaces and suffers from overhead  Cluster key:
of high communication. Not scalable and not flexible against A key was shared to all the nodes in the cluster. The cluster
the attackers nodes. It fails to support the mobile nodes [1]. A key is used to secure the message which is sensitive. Cluster
sensor with large number of resources needed a large key size key was update by the cluster head during forward and
which is not suited [1]. Wireless sensor network were backward secrecy.
embarrassed in the resources of Energy-efficient distributed
deterministic key management [2]. The pairwise key was
maintained and estabilished only on the local clusters and
flaws are fixed in existing key management schemes. Replay
attacks were barely launched by adversaries by the use of
neighbour table [2]. A key predistribution scheme was
developed during the node operation in wireless sensor
network [3]. It significantly increases the connectivity of the
networks and large amount of memory were not used here. It Fig.1 System Architecture
is flexible against the node detain [3]. Here flexibility and
connectivity are the two differing properties, lower flexibility In fig1 it shows that the node were registered in the key
needs higher connectivity [3]. A new predistribution scheme generation centre and certain key were provided for the secure
increases the flexibility than the existing schemes [4].To store node movement.
pairwise key requires large number of space and suffers from
overhead of high communication [4]. In heterogeneous sensor V. PHASES OF CL_EKM
networks, the key were maintained efficiently to provide The CL_EKM includes the phases of system setup, pairwise
security [5]. Here routing is done by two phases 1) Intra key generation, cluster formation, key update, node movement
cluster router sends data to the cluster head (H sensor) by the and node revocation.
L sensor via multi hops of other L sensor. 2) Inter cluster A.SYSTEM SETUP- Base station generates the parameters
router collects multiple data from the L sensor and sends the and node gets register before the deployment of network by
data to the sink via the H sensor backbone [5]. This scheme including it in a member list.
was susceptible against message forgery and known key -Generation of System Parameters
attacks [5]. -Node Registration
The design, implementation of TinyECC includes ECC
operations were used in wireless sensor networks [6]. B. PAIRWISE KEY GENERATION-To activate the pairwise
TinyECC used PKC operations that can be flexible and key setup, a node transmit an advertised message to the
incorporated into wireless sensor network. Based on the neighbourhood cluster nodes after the network has deployed.
developer needs the turn of specific operations were on or off The advertised message includes its identifier and public key.
which includes number of optimization switches provided by i) PAIRWISE MASTER KEY ESTABILISHMENT- The
TinyECC. TinyECC increases energy consumption [6]. protocol was described for the establishment of pairwise
master key between the two nodes nA and nB. An
encapsulation process was used here between nA and nB for
secure message transmission.
IV. OVERVIEW OF CERTIFICATELESS KEY
ii) PAIRWISE ENCRYPTION KEY ESTABILISHMENT-
MANAGEMENT
The HMAC of KAB was generated when the pairwise master
key get generated. The node nA and nb was validated by
 Certificateless public /private key: HMAC. The pairwise encryption key estabilished the HMAC
The base station at the key generation centre (KGC) create a value once the validation gets succeed.
pair of certificateless public /private key and use that key to
the node before the node gets deployed. C. CLUSTER FORMATION-After the node deployment H
 Individual node key: sensor creates L sensor in the course of signal message
The different key was used for the each node with the base exchange and authenticated. If the process of authentication is
station. For example: L sensor sends the encrypted message succeeding, cluster were formed by H sensor with the help of
with the individual key to the H sensor and alert message were L sensor which shares a same cluster key. Each member of
sent to the base station or if it is fails to communicate cluster establishes a pairwise key.
 Pairwise key generation:
For a secure communication and node authentication a D. KEY UPDATE- Frequent key updating requires protecting
pairwise key were created between neighbouring nodes. For against the attacker and cryptanalysis. In this process pairwise
example: L sensor and H sensor uses the same pairwise key. L key update and cluster key update operations were done.
E. NODE MOVEMENT- In order to ensure forward and
backward secrecy the cluster key should be properly managed
by H sensor during the node movements. Thus the cluster key
was update by the H sensor and base station gets notified
about the changes of node status. By the use of this report the
base station instantly revise the node status in the list.

F. KEY REVOCATION-To perceive malicious attacks the

base station uses the intrusion detection system. By the use of
node status information cluster head examine the abnormal
node when the node leaves and node join the cluster the node
status information gets update to the base station in the
member list.

VI. RESULTS
After implementing the proposed system on NS2 Fig.4 Throughput Comparison graph
platform, the results obtained are as follows:

Fig.4 Energy Consumption Comparison graph

Fig.2 Network Creation

VII. CONCLUSION
Dynamic wireless sensor network provides a secure
communication between the nodes by the use of
Certificateless key management scheme (CL-EKM).CL-EKM
ensure the forward and backward secrecy and maintain the
secure node transmission process. The efficient key was
updated to the cluster whenever the nodes leaves and joins the
cluster. By using applications can secure the node
transmission against the attacker’s node.CL-EKM scheme
used for flexibility against negotiation nodes, provides
integrity, interception, cloning and replication of data files.
This process results in maintain the key efficiently by the
scheme of CL-EKM in resource embarrassed wireless sensor
network. The future plan is to conceive a frequency range
during the node movements within or across the cluster with
the help of Thold values based on CL-EKM scheme.

Fig.3 Network Configuration

REFERENCES
[1] H. Chan, A. Perrig, and D. Song, “Random key predistribution [18] K. Chatterjee, A. De, and D. Gupta, “An improved ID-based key
schemes for sensor networks,” in Proc. IEEE Symp. SP, May 2003, management scheme in wireless sensor network,” in Proc. 3rd Int.
pp. 197–213. Conf. ICSI, vol. 7332. 2012, pp. 351–359.
[2] W. Du, J. Deng, Y. S. Han, and P. K. Varshney, “A key [19] W. T. Zhu, J. Zhou, R. H. Deng, and F. Bao, “Detecting node
predistribution scheme for sensor networks using deployment replication attacks in mobile sensor networks: Theory and
knowledge,” IEEE Trans. Dependable Secure Comput., vol. 3, no. 1, approaches,” Secur. Commun. Netw., vol. 5, no. 5, pp. 496–507,
pp. 62–77, Jan./Mar. 2006. 2012.
[3] W. Du, J. Deng, Y. S. Han, P. Varshney, J. Katz, and A. Khalili, [20] M. A. Rassam, M. A. Maarof, and A. Zainal, “A survey of
“A pairwise key predistribution scheme for wireless sensor intrusion detection schemes in wireless sensor networks,” Amer. J.
networks,” ACM Trans. Inf. Syst. Secur., vol. 8, no. 2, pp. 228–258, Appl. Sci., vol. 9, no. 10, pp. 1636–1652, 2012.
2005.
[4] M. Rahman and K. El-Khatib, “Private key agreement and secure
communication for heterogeneous sensor networks,” J. Parallel
Distrib. Comput., vol. 70, no. 8, pp. 858–870, 2010.
[5] M. R. Alagheband and M. R. Aref, “Dynamic and secure key
management model for hierarchical heterogeneous sensor networks,”
IET Inf. Secur., vol. 6, no. 4, pp. 271–280, Dec. 2012.
[6] D. S. Sanchez and H. Baldus, “A deterministic pairwise key
predistribution scheme for mobile sensor networks,” in Proc. 1st Int.
Conf. SecureComm, Sep. 2005, pp. 277–288.
[7] I.-H. Chuang, W.-T. Su, C.-Y. Wu, J.-P. Hsu, and Y.-H. Kuo,
“Twolayered dynamic key management in mobile and long-lived
clusterbased wireless sensor networks,” in Proc. IEEE WCNC, Mar.
2007, pp. 4145–4150.
[8] S. Agrawal, R. Roman, M. L. Das, A. Mathuria, and J. Lopez, “A
novel key update protocol in mobile sensor networks,” in Proc. 8th
Int. Conf. ICISS, vol. 7671. 2012, pp. 194–207.
[9] S. U. Khan, C. Pastrone, L. Lavagno, and M. A. Spirito, “An
energy and memory-efficient key management scheme for mobile
heterogeneous sensor networks,” in Proc. 6th Int. Conf. CRiSIS, Sep.
2011, pp. 1–8.
[10] X. Zhang, J. He, and Q. Wei, “EDDK: Energy-efficient
distributed deterministic key management for wireless sensor
networks,” EURASIP J. Wireless Commun. Netw., vol. 2011, pp. 1–
11, Jan. 2011.
[11] N. Gura, A. Patel, A. Wander, H. Eberle, and S. C. Shantz,
“Comparing elliptic curve cryptography and RSA on 8-bit CPUs,” in
Proc. 6th Int. Workshop Cryptograph. Hardw. Embedded Syst., 2004,
pp. 119–132.
[12] S. S. Al-Riyami and K. G. Paterson, “Certificateless public key
cryptography,” in Proc. 9th Int. Conf. ASIACRYPT, vol. 2894. 2013,
pp. 452–473.
[13] S. Seo and E. Bertino, “Elliptic curve cryptography based
certificateless hybrid signcryption scheme without pairing,”
CERIAS, West Lafayette, IN, USA, Tech. Rep. CERIAS TR 2013-
10, 2013. [Online]. Available:
https://www.cerias.purdue.edu/apps/reports_and_papers/.Seung-
Hyun
[14] S. H. Seo, J. Won, and E. Bertino, “POSTER: A pairing-free
certificateless hybrid sign cryption scheme for advanced metering
infrastructures,” in Proc. 4th ACM CODASPY, 2014, pp. 143–146.
[15] Q. Huang, J. Cukier, H. Kobayashi, B. Liu, and J. Zhang, “Fast
authenticated key establishment protocols for self-organizing sensor
networks,” in Proc. 2nd ACM Int. Conf. WSNA, 2003, pp. 141–150.
[16] X.-J. Lin and L. Sun, “Cryptanalysis and improvement of a
dynamic and secure key management model for hierarchical
heterogeneous sensor networks,” in Proc. IACR Cryptol. ePrint
Archive, 2013, pp. 698–698.
[17] P. Szczechowiak, L. B. Oliveira, M. Scott, M. Collier, and R.
Dahab, “NanoECC: Testing the limits of elliptic curve cryptography
in sensor networks,” in Proc. 5th Eur. Conf. WSN, vol. 4913. 2008,
pp. 305–320.