Survey of the end to end secure communication in ADHOC networks

ISSN 2319 – 1953
International Journal of Scientific Research in Computer Science Applications and Management Studies
IJSRCSAMS
Volume 1, Issue 3 (NOVEMBER 2012) www.ijsrcsams.com
Survey of End to End Secure Communication in
Adhoc Networks
Ahmad Sharifi#1
, Saeideh Taheri*2
, Mohsen Khosravi@3
,G.Praveen Babu^4
#
School ofInformation Technology,Jawaharlal Nehru Technological University, Hyderabad, India
*
Computer Engineering Department, Iran University of Science & Technology,Tehran,Iran
@
Faculty of Information and Communication Technolog, International Islamic university Malaysia (IIUM)
^
School ofInformation Technology,Jawaharlal Nehru Technological University, Hyderabad, India
1
ahmadsharifi@ieee.org
2
sa_taheri@vu.iust.ac.ir
3
mohsen.khosravi@live.iium.edu.my
4
pravbob@jntuh.ac.in
Abstract— several algorithms are proposed which uses the basic
scheme by predistributing random keys into nodes. The
drawback is that one pair wise key may be shared by multiple
links. Chan et al. presented two schemes. In their q-composite
scheme, multiple keys are required to establish a secure link,
which makes a trade-off between connectivity and security. In
their random pair wise-key scheme, a unique pair wise key is
assigned to each node and every one of a random set. This
scheme provides high security but poses an upper bound on
network size. Du et al. proposed the pair wise key predistribution
scheme based on both the basic scheme and Blom’s scheme, from
which it inherits the threshold property. On the contrary, our
scheme utilizes Blom’s scheme more smoothly.
Keywords— Wireless, Authentication, Sensor network, Trust
management
I. INTRODUCTION
Du et al. and Liu and Ning independently proposed to
utilize deployment knowledge to improve the performance of
key establishment. Our scheme outperforms Du’s deployment
knowledge scheme in terms of connectivity and security. Liu
and Ning’s polynomial-based key predistribution scheme also
has the threshold property for the use of bivariate polynomials,
which is a special form of Blom’s scheme. As we mentioned,
our scheme utilizes the deployment knowledge in a smoother
way. Different from all of these, Zhu et al. presented LEAP by
introducing a weaker model, which assumes that there exists a
short time interval within which nodes can establish pair wise
keys securely. However, this time interval is often very hard
to estimate accurately. Once it is overestimated, all links may
be compromised.
Probabilistic Key Sharing discussed most of the proposed
symmetric key cryptography protocols for establishing a pair
wise shared key between two nodes make use of an on-line
key server. Mitchell and Piper proposed a solution based on
probabilistic key sharing that does not depend on such an on-
line server. However, the storage complexity imposed on each
participant in their scheme seems to be unaffordable in the
context of ad hoc networks.
The probabilistic keying scheme in our protocol is similar
to schemes that have been used by other researchers.
Eschenauer and Gligor introduced a key management scheme
based on probabilistic key sharing for distributed sensor
networks (DSN) with central key servers (e.g., base stations).
Chan et al. extended this scheme by presenting three new
mechanisms for key establishment in sensor networks based
on the framework of probabilistic key predeployment,
including a mechanism for pair wise shared key establishment
called multipath key reinforcement. Our work differs from the
previous ones in several aspects.
First, in our scheme, a node can deduce the set of keys it
shares with any other node (which may be an empty set) only
based on the latter’s identity. In contrast, the approaches
require each node to exchange the ids of the keys it possesses
with its neighbours. Thus, our approach trades computation
for communication, which is desirable in ad hoc networks.
Second, Eschenauer and Gligor proposed using the
predeployed keys for encrypting all communication between
nodes. A session key between two nodes can also be
established using a logical path secured by the predeployed
keys. However, it seems that the established session key might
not be exclusively known to the two nodes involved, because
each predeployed key is known to several nodes. In contrast,
we propose using the predeployed keys for establishing a
shared pair wise key that is exclusively known to two nodes
with overwhelming probability.

ISSN 2319 – 1953
IJSRCSAMS
II. KEY SCHEMES
We note that the multipath key reinforcement scheme
shares some similarities with our scheme in that both the
schemes use the idea of secret sharing; however, our scheme
differs from theirs in the following aspects. First, their scheme
uses multiple physically disjoint paths between two nodes in
establishing a pair wise key, while our scheme can use a
single physical path as long as the shares are transmitted over
multiple logically disjoint paths. Second, we present a detailed
security and performance analysis of the idea of combining
probabilistic key sharing and secret sharing, and also present
an algorithm for deciding the number of secret shares to be
used for establishing a pair wise key based on the desired
level of security.
III. THRESHOLD SECRET SHARING
There has been a great deal of research on threshold secret
sharing Shamir and its applications. In one direction, Gong
proposed an approach in which threshold secret sharing is
used for increasing the availability of authentication services.
Our work bears the similarity that we also utilize secret
sharing techniques to establish pair wise keys. Unlike Gong’s
scheme, however, our scheme does not use any single on-line
key server. In another relevant direction, researchers have
extensively investigated the interplay of network connectivity
and secure and reliable communication (e.g., Dolev, Delev et
al., Franklin and Wright, Desmedt and Wang). We refer the
reader to Bagchi et al. For an overview and recent result in
this regard.
Network, Node and Security Assumptions First, we assume
network links are bidirectional, i.e., if node A can hear node B,
B can also hear A. This is true when all the nodes use
omnidirectional antennas and have equal power levels. Second,
we aim to provide solutions for low-end devices. The
resources of a node such as power, storage, computation and
communication capacity, are relatively constrained, making
public key techniques impractical. We assume that every node
has space for storing hundreds of bytes or a few kilobytes of
keying materials, depending on the security requirements.
Third, we do not assume a central key server exists in the
formed network, whereas it may exist off-line to initiate the
nodes prior to the formation of the network.
Fourth, we assume that if a node is compromised, all the
information it holds will also be compromised. We do not
distinguish between a compromised node and an attacker.
Moreover, all the compromised nodes may try to eavesdrop
on other nodes’ communications and collude to launch attacks
by sharing their keying materials.
IV.KEY DISTRIBUTION
Our pair wise key establishment protocol is based on two
techniques – probabilistic key sharing and threshold secret
sharing. Before the deployment of a network, i.e., during a
key pre-distribution phase, every node is loaded with a (small)
fraction of keys out of a large pool of keys by a key server.
Note that this phase occurs before the deployment of the
network, and the key server stays off-line after finishing this
phase. Keys are allocated to each node using a probabilistic
scheme that enables every pair of nodes to share one or more
keys with certain probability. The keys directly shared
between any two nodes can thus be used to encrypt messages
exchanged between them. Even if two nodes do not share any
keys directly, our probabilistic key sharing scheme enables
them to communicate securely using logical paths obtained
via a logical path discovery process[1].
To be concrete, consider two nodes u and v that wish to
communicate privately. u and v may already share one or
more keys from the pool of keys after the key predistribution
phase. However, these keys are not known exclusively to u
and v because every key in our key pool may be allocated to
multiple nodes; hence, they cannot be used for encrypting any
message that is private to u and v. Thus the goal of our
algorithm is to establish a key, S, that is known exclusively to
u and v. The basic idea underlying the establishment of such a
key S is as follows: The sender node splits S into multiple
shares using an appropriate secret sharing scheme. The sender
then transmits to the recipient node all these shares, using a
different logical path for each share. The recipient node then
reconstructs S after it receives all (or a certain number of) the
shares.
V. SENSOR NODES
Sensor networks are ideal candidates for applications such
as military target tracking, home security monitoring, and
scientific exploration in dangerous environments. Typically, a
sensor network consists of a potentially large number of
resource constrained sensors, which are mainly used to collect
data (e.g. temperature) from the environment, and a few
control nodes, which may have more resources and may be
used to control the sensors and/or connect the network to the
outside world (e.g. a central data processingServer). Sensors
usually communicate with each other through wireless
communication channels. Sensor networks may be deployed
in hostile environments, especially in military applications. In
such situations, the sensors may be captured, and the
data/control packets may be intercepted and/or modified.
Therefore, security services such as authentication and
encryption are essential to maintain the network operations.
However, due to the resource constraints on the sensors, many
security mechanisms such as public key cryptography are not
feasible in sensor networks. Indeed, providing security
services in sensor networks is by no means a trivial problem;
it has received a lot of attention recently [1].
A fundamental security service is the establishment of a
symmetric, pair wise key shared between two sensors, which
is the basis of other security services such as encryption and
authentication. Several key predistribution techniques have
been developed recently to address this problem. Eschenauer
and Gligor proposed the basic probabilistic key predistribution,
in which each sensor is assigned a random subset of keys from
a key pool before the deployment of the network. By doing

ISSN 2319 – 1953
IJSRCSAMS
this, two sensors can have a certain probability to share at
least one key. Chan et al. developed the q-composite key
predistribution and the random pair wise keys schemes. The q-
composite key predistribution scheme is based on the basic
probabilistic scheme, but it requires two sensors share at least
q predistributed keys to establish a pair wise key. The random
pair wise keys scheme pre distribute random pair wise keys
between a particular sensor and a random subset of other
sensors, and has the property that compromised sensors do not
lead to the compromise of pair wise keys shared between non-
compromised sensors. However, these approaches still have
some limitations. For the basic probabilistic and the q-
composite key predistribution, a small number of
compromised sensors may reveal a large fraction of pair wise
keys shared between non-compromised sensors. Though the
random pair wise keys scheme provides perfect security
against node captures, the maximum supported network size is
strictly limited by the storage capacity for pair wise keys and
the desired probability to share a key between two sensors.
Liu and Ning developed a framework to predistributepair wise
keys using bivariate polynomials and proposed two efficient
instantiations, a random subset assignment scheme and a grid-
based key predistribution scheme, to establish pair wise keys
in sensor networks.
Our second theme during this paper is thought-about
associate degree internal representation of this framework
however can do higher performance thanks to the specific
usage of location data. Recent advances in electronic and pc
technologies have paved the manner for the proliferation of
wireless detector networks (WSN). Detector networks
typically comprises an oversized variety of ultra-small
autonomous devices. every device, known as a detector node,
is battery battery-powered and equipped with integrated
sensors, processing capabilities, and short-range radio
communications. In typical application eventualities, detector
nodes square measure unfold every which way over the
preparation region beneath scrutiny and collect detector
information. samples of detector network comes embrace
SmartDust and WINS.
Sensor networks ar being deployed for a large kind of
applications, together with military sensing and pursuit,
atmosphere observation, patient observation and pursuit, good
environments, etc. once sensing element networks ar deployed
during a hostile atmosphere, security becomes extraordinarily
necessary, as they're vulnerable to differing kinds of malicious
attacks. for instance, associate individual will simply hear the
traffic, impersonate one in every of the network nodes1, or by
choice give deceptive data to different nodes. to supply
security, communication ought to be encrypted and echt.
associate open analysis drawback is the way to bootstrap
secure communications among sensing element nodes, i.e. the
way to discovered secret keys among human activity nodes?
VI. KEY MANAGEMENT
This key agreement drawback could be a a part of the key
management drawback, that has been wide studied generally
network environments. There ar 3 sorts of general key
agreement themes: trusted-server scheme, self-enforcing
theme, and key pre-distribution theme. The trusty-server
theme depends on a trusted server for key agreement between
nodes, e.g., Kerberos. this sort of theme isn't appropriate for
detector networks as a result of there's typically no trusty
infrastructure in detector networks. The self-enforcing theme
depends on uneven cryptography, like key agreement
mistreatment public key certificates. However, restricted
computation and energy resources of detector nodes usually
create it undesirable to use public key algorithms, like Diffie-
Hellman key agreement or RSA, as realized. The third form of
key agreement theme is vital pre-distribution, wherever key
info is distributed among all detector nodes before preparation.
If we all know that nodes ar additional seemingly to remain
within the same neighbourhood before preparation, keys are
often set a priori. However, attributable to the randomness of
the preparation, knowing the set of neighbours
deterministically may not be possible [5].
There exist variety of key pre-distribution schemes. A naive
resolution is to let all the nodes carry a master secret key. Any
try of nodes will use this international master secret key to
attain key agreement and acquire a brand new try wise key.
This theme doesn't exhibit fascinating network resilience: if
one node is compromised, the protection of the complete
detector network are compromised. Some existing studies
counsel storing the key in tamper-resistant hardware to cut
back the danger, however this will increase the price and
energy consumption of every detector. what is more, tamper-
resistant hardware won't perpetually be safe. Another key pre-
distribution theme is to let every detector carry N − one secret
try wise keys, every of that is thought solely to the present
detector and one in all the opposite N−1 sensors (assuming N
is that the total variety of sensors). The resilience of this
theme is ideal as a result of compromising one node doesn't
have an effect on the protection of communications among
alternative nodes; but, this theme is impractical for sensors
with a very restricted quantity of memory as a result of N may
be massive. Moreover, adding new nodes to a pre-existing
detector network is tough as a result of the prevailing nodes
don't have the new nodes’ keys.
The Eschenauer-Gligor theme has been delineated earlier
during this section. we'll provides a additional elaborated
description of this theme in Section II. supported the
Eschenauer-Gligor theme, Chan, Perrig, and Song projected a
q-composite random key pre-distribution theme. The
distinction between this theme and therefore the Eschenauer-
Gligor theme is that letter common keys (q nine 1), rather than
simply one one, ar required to ascertain secure
communications between a try of nodes. it's shown that, by
increasing the worth of letter, network resilience against node
capture is improved, i.e., associate offender needs to
compromise more nodes to realize a high chance of
compromised communication [3].
Du, Deng, Han, and Varshney projected a brand new key
predistribution theme, that well improves the resilience of the

ISSN 2319 – 1953
IJSRCSAMS
network compared to the present schemes. This theme
exhibits a pleasant threshold property: once the amount of
compromised nodes is a smaller amount than the edge, the
likelihood that any nodes apart from these compromised nodes
ar affected is near zero. This fascinating property lowers the
initial payoff of smaller scale network breaches to AN soul,
and makes it necessary for the soul to attack a big proportion
of the network. an identical methodology is additionally
developed by Liu and Ning.
The ideas delineated during this paper is applied to all or
any of the higher than pre-distribution schemes to additional
improve their performance. Blundo et al. projected many
schemes which permit any cluster of t parties to figure a
typical key whereas being secure against collusion between a
number of them. These schemes concentrate on saving
communication prices whereas memory constraints aren't
placed on cluster members. Perrig et al. projected SPINS, a
security design specifically designed for detector networks. In
SPINS, every detector node shares a secret key with the
bottom station. 2 detector nodes cannot directly establish a
secret key. However, they will use the bottom station as a sure
third party to line up the key.
Several other key distribution schemes have been proposed
for mobile computing, although they are not specifically
targeted at sensor networks. Tatebayashi, Matsuzaki, and
Newman consider key distribution for resource-starved
devices in a mobile environment. This work is further
improved by Park et al. Other key agreement and
authentication protocols include the one by Beller and Yacobi.
A survey on key distribution and authentication for resource-
starved devices in mobile environments is given. The majority
of these approaches rely on asymmetric cryptography, which
is not a feasible solution for sensor networks. Several other
methods based on asymmetric cryptography are also proposed:
Zhou and Hass propose a secure ad hoc network using secret
sharing and threshold cryptography. Kong et al. also propose
localized public-key infrastructure mechanisms, based on
secret sharing schemes.
VII. DISTRIBUTED NETWORKS
Distributed sensor networks have received a lot of attention
recently due to their wide application in military as well as
civilian operations. Example applications include target
tracking, scientific exploration, and monitoring of nuclear
power plants. Sensor nodes are typically low-cost, battery
powered, and highly resource constrained, and usually
collaborates with each other to accomplish their tasks [1].
Security services such as authentication and key
management are critical to secure the communication between
sensors in hostile environments. As one of the most
fundamental security services, pair wise key establishment
enables the sensor nodes to communicate securely with each
other using cryptographic techniques. However, due to the
resource constraints on sensor nodes, it is not feasible for
sensors to use traditional pair wise key establishment
techniques such as public key cryptography and key
distribution centre (KDC).
Eschenauer and Gligor proposed a probabilistic key
predistribution scheme recently for pair wise key
establishment. The main idea was to let each sensor node
randomly pick a set of keys from a key pool before
deployment so any two sensor nodes have a certain
probability of sharing at least one common key. Chan et al.
further extended this idea and developed two key
predistribution techniques: q-composite key predistribution
and random pair wise keys scheme [6]. The q-composite key
predistribution also uses a key pool but requires two sensors
compute a pair wise key from at least q predistributed keys
they share. The random pair wise keys scheme randomly picks
pairs of sensors and assigns each pair a unique random key.
Both schemes improve the security over the basic probabilistic
key predistribution scheme. However, the pair wise key
establishment problem is still not solved. For the basic
probabilistic and the q-composite key predistribution schemes,
as the number of compromised nodes increases, the fraction of
affected pair wise keys increases quickly. As a result, a small
number of compromised nodes may affect a large fraction of
pairwise keys. While the random pair wise keys scheme
doesn’t suffer from the above security problem, given a
memory constraint, the network size is strictly limited by the
desired probability that two sensors share a pair wise key and
the number of neighbour nodes that a sensor can communicate
with it.
VIII. RELATED WORKS
Some general key distribution and management approaches
aren't appropriate for wireless sensing element networks. First,
trivially storing in every node a combine wise key for each
different node poses a high memory demand unaffordable for
sensing element nodes. Second, on-line key distribution and
management offered by the bottom station is inefficient for
wireless sensing element networks owing to high
communication overhead. Third, public-key algorithms like
RSA, Diffie-Hellman, and Elliptic Curve Cryptography (ECC)
square measure too high-ticket to current sensing element
nodes for prime energy consumption and computation
overhead. Experimental results from existing analysis show
that the execution time of public key- primarily based
operations, like cryptography and cryptography, is of the order
of seconds or perhaps ten seconds. Moreover, wireless sensing
element networks might not be able to offer the specified
public-key infrastructure (PKI) for key distribution. we've to
either distribute public keys into nodes through the bottom
station on-line, which can cause high communication
overhead, or predistribute public keys into nodes offline,
which can want some theme like what we have a tendency to
gift during this paper to boost its potency.
Fortunately, the bootstrapping problem can be solved by
key predistribution schemes that predistribute secret
information in nodes to help them establish secure links after
deployment. Eschenauer and Gligor proposed basic scheme by

ISSN 2319 – 1953
IJSRCSAMS
utilizing probabilistic key predistribution, which was
improved byChanet al. And Duet al. Recently, Duet al.And
Liu and Ning independently proposed to make use of
deployment knowledge to further improve the performance of
key establishment. Different from all these schemes, LEAP
proposed by Zhu et al. assumes a weaker model, that is, there
exists a short time interval within which nodes can establish
pair wise keys safely after deployment.
IX.PREDISTRIBIUTION SCHEME
Group-based preparation information implies that every
cluster of nodes reside solely among alittle native space,
which implies that almost all of the neighbours of every node
come back from its own cluster and neighbour teams.
Therefore, to attain a extremely connected network, the key
purpose is to maximise the likelihood that nodes from
identical cluster and neighbour teams share keys. For this
purpose, we have a tendency to divide the links of detector
networks into 2 sorts, in-group links and intergroup links,
reckoning on whether or not the concerned nodes area unit
from identical cluster or not, severally. consequently, we have
a tendency to build 2 styles of secret matrices A and B,
severally [8]. Our theme consists of key predistribution part
and key discovery part. Key Predistribution Phase: during this
part, we have a tendency to generate a worldwide public
matrix G and variety of secret A and B matrices. All teams
share the world matrix G, which implies each node of a gaggle
can choose a corresponding column from G. Meanwhile,
every cluster is assigned a singular secret matrix A, that is,
each node of {a cluster|a gaggle|a bunch} can choose its
corresponding row from the distinctive matrix A assigned to
its group. Thus, we have a tendency to guarantee that any 2
nodes from identical cluster will forever notice a combine
wise key [10].
Then, we tend to assign every cluster some variety of B
matrices to ensure that every combine of neighbour teams
shares a minimum of one common matrix B. a lot of exactly,
we tend to 1st choose some teams and assign every of them a
unique secret matrix B, wherever these teams ar known as
basic teams, and that we decision alternative nonbasic teams
traditional teams. Then, for every cluster (including basic and
traditional groups), we tend to assign it all of the B matrices
that are allotted to its neighbour basic teams, that ar the
essential teams among its neighbour teams. After that, every
node picks corresponding rows from some or all (depending
on totally different methods) of the B matrices that are allotted
to its cluster. At last, we tend to set all nodes an equivalent
transmission vary and deploy them in teams.
X. CONCLUSIONS
After deployment, each node first probes its neighbours.
Then, neighbour nodes exchange their group indexes, indexes
of B matrices, and columns of matrix G. If two neighbours
come from the same group, they derive the pair wise key from
the common matrix A and G. If they are not from the same
group, but share one or more common B matrices, they can
also find out the pair wise key from one shared matrix B and
the global matrix G. Then, the neighbours with pair wise keys
established will build a secure link between each other and
start to transmit data securely through the link. Meanwhile,
those neighbours sharing no pair wise keys will no longer
communicate with each other. Note: Some nodes that find no
pair wise keys between them may still exploit other methods
such as multihop path reinforcement to establish pair wise
keys indirectly. However, this discussion is out of the scope of
our paper and we only focus on how to establish pair wise
keys through one-hop links.
REFERENCES
[1] H. Dai and H. Xu, ―Triangle-based key management
scheme for wireless sensor networks," Frontiers Electrical
Electron. Eng. China, vol. 4, no. 3, pp. 300-306, 2009.
[2] A. Poornima and B. Amberker, ―Tree-based key
management scheme for heterogeneous sensor networks," in
16th IEEE International Conf. Netw., 2008.
[3] Y. Zhang, W. Yang, K. Kim, and M. Park, ―An AVL tree-
based dynamic key management in hierarchical wireless
sensor network," in Proc. International Conf. Intelligent Inf.
Hiding Multimedia Signal Process., pp. 298-303, 2008.
[4] A. Poornima and B. Amberker, ―Key management
schemes for secure communication in heterogeneous sensor
networks," International J. Recent Trends Eng., 2009.
[5] A. Das, ―An unconditionally secure key management
scheme for largescale heterogeneous wireless sensor
networks," in Proc. First International Commun. Syst. Netw.
Workshops, pp. 1-10, 2009.
[6] Y. Yang, J. Zhou, R. Deng, and F. Bao, ―Hierarchical self-
healing key distribution for heterogeneous wireless sensor
networks," SecurityPrivacyCommun.Netw., pp. 285-295,
2009.
[7] B. Tian, S. Han, and T. Dillon, ―A key management
scheme for heterogeneous sensor networks using keyed-hash
chain," in 5th International Conf. Mobile Ad-hoc Sensor
Netw., 2010.
[8] J. Kim, J. Lee, and K. Rim, ―Energy efficient key
management protocol in wireless sensor networks,"
International J. Security its Appl., 2010.
[9] M. Wen, Z. Yin, Y. Long, and Y. Wang, ―An adaptive key
management framework for the wireless mesh and sensor
networks," Wireless Sensor Netw.J., 2010.
[10] H. Jen-Yan, I. Liao, and H. Tang, ―A forward
authentication key management scheme for heterogeneous
sensor networks," EURASIP J. Wireless Commun.Netw.,
2010.

Survey of the end to end secure communication in ADHOC networks

More Related Content

What's hot (20)

Viewers also liked (19)

Similar to Survey of the end to end secure communication in ADHOC networks (20)

Recently uploaded (20)

Survey of the end to end secure communication in ADHOC networks