A Key Management Approach For Wireless Sensor Networks

International Journal of Information Technology, Modeling and Computing (IJITMC) Vol. 2, No.3, August 2014
DOI : 10.5121/ijitmc.2014.2301 1
A KEY MANAGEMENT APPROACH FOR WIRELESS
SENSOR NETWORKS
Ali Bagherinia, Akbar Bemana, Sohrab Hojjatkhah, Ali Jouharpour
Department of Computer Engineering, Islamic Azad University-Dehdasht Branch,
Dehdasht, Iran
ABSTRACT
In this paper we presenta key management approach for wireless sensor networks. This approach
facilitating an efficient scalable post-distribution key establishment that provides different security services.
We have developed and tested this approach under TinyOs. Result shows that this approach provides
acceptable resistance against node capture attacks and replay attacks. The provision of security services is
completely transparent to the user of the WSNs. Furthermore, being highly scalable and lightweight, this
approach is appropriate to be used in a wireless sensor network of hundreds of nodes.
KEYWORDS
Sensor networks,key management , scalability , flexibility , resistant.
1. INTRODUCTION
Recent advances in electronic and computer technologies have paved the way for the proliferation
of wireless sensor networks (WSN) [1,2]. Sensor networks usually consist of a large number of
ultra-small autonomous devices. Each device, called a sensor node. Each sensor node measures
necessary parameters from round area and communicate it’s with radio sender through electrical
signal. Processing of this signal extracts specification such as object placement or around events.
Figure 1 shows modular structure of each multi sensing sensor node. Each sensor node consists
of: multi sensing interface and A/D (for sensing corresponding analog area such as pressure,
temperature …), memory, CPU, RF and controller [3].
Figure 1. Sensor node structure with multiple sensing units

2
This key agreement problem is a part of the key management problem, which has been widely
studied in general network environments. There are three types of general key agreement schemes:
trusted- server scheme, self-enforcing scheme, and key pre-distribution scheme. The trusted
server scheme depends on a trusted server for key agreement between nodes ,e.g.,
Kerberos[5].This type of scheme is not suitable for sensor networks because there is usually not
rusted infrastructure in sensor networks. The self-enforcing scheme depends on a symmetric
cryptography, such as key a agreement using public key certificates. However, limited
computation and energy resources of sensor nodes of ten make it undesirable to use public key
algorithms, such as Daffier-Hell man key agreement[6]or RSA[7],as pointed outing[8].The third
type of key agreement scheme is key redistribution, where key information is distribute among all
sensor no despair to deployment. If we know which nodes are more likely to stay in the same
neighborhood before deployment, key scan be decide dapriori. However, because of the
randomness of the deployment, knowing the set of neighbors deterministically might not be
feasible.
.
2. Related work
Key management is an essential challenge in a large-scale and resource-limited particularly
WSNS. In [28],[11], [12], [13], [14], [15], [16], [17] a number of pair-wise symmetric key
establishment schemes have been recently proposed. Most of them use the idea of probabilistic
key sharing [14] to establish trust between two nodes, each with different emphasis on enhanced
security protection [11], flexibility of security requirements [17], high probability of key
establishment and reduced overhead [15], or utilization of deployment knowledge [12]. Such
pairwise keys can be used to authenticate a node’s identity or messages; however, they cannot
handle the fabricated sensing data injected by compromised nodes. Instead, semantic verification
of the data is required to detect the fabricated ones. Secure Diffusion exploits location-based key
management to achieve this goal. Because the data authentication keys are bound to geographic
locations, the compromised nodes outside the targeted region, no matter how many there are,
cannot fabricate sensing data without being detected.
Secure routing has been extensively studied in the context of ad-hoc networks [18], [19], [20],
[21]. However, none of these protocols can be applied in sensor networks, because none
addresses the unique feature of data-centric communication, and the network scale is limited by
the excessive number of keys each node should store. The challenges of secure sensor routing are
discussed in [22], together with security threat and counter-measurement analysis on a few
popular routing protocols. However, it does not consider the fabricated data injection attacks
launched by compromised nodes.
Two recent studies of SEF [23] and Hop-by-Hop Authentication [24] address the problem of
filtering the fabricated data en-route in sensor networks. Such early drop of malicious traffic can
potentially save precious energy resources at forwarding nodes. Secure Diffusion takes a different
approach that quarantines the malicious traffic through implicit rate control and negative
reinforcement mechanisms. As a result, Secure Diffusion is resilient to an increasing number of
compromised nodes, whereas both SEF and Hop-by-Hop Authentication completely lose security
protection when the attacker has compromised beyond a small, fixed number of nodes.
There are a few recent security proposals that explicitly involve the geographic locations. The
Echo protocol [25] exploits an on-site verifier node with ultrasound transceiver to verify a
location claim. A recent secure routing proposal TRANS [26] monitors the behavior of static

3
sensor nodes, and then bypasses the areas of misbehaving nodes in the route. The pair wise key
establishment scheme in [16] exploits a location- aware deployment model and pre-distributes
pair wise keys between nodes that are expected to be close to each other. However, Secure
Diffusion differs from all these work in that it binds keys to locations, and provides a scalable
secure data dissemination protocol for sensor networks.
Thee exist a number of key pre-distribution schemes. A naive solution isotope tall the nodes carry
a Master secret key. Any pair of nodes can use this global master secret key to achieve key
agreement and obtain a new pair wise key. This scheme does not exhibit desirable network
resilience: if one node Is compromised, the security of the entire sensor network will be
compromised. Some existing studies Suggest storing the master key in tamper- resistant hardware
to reduce the risk, but this increases the Cost and energy consumption of each sensor.
Furthermore, tamper- resistant hardware might not always be safe[9]. Another key pre-
distribution scheme isolate each sensor carry N-1 secret pair Wise keys, each of which is known
only to this sensor and one of the other N-1 sensors(assuming Nis the total number of sensors).
The resilience of this scheme is perfect because compromising one Node does not affect the
security of communications among other nodes; however, this scheme is impractical for sensor
switch an extremely limited amount of memory because N could be large. Moreover, adding new
nodes to a pre-existing sensor network is difficult because the exist in nodes do not have the new
nodes ’keys. Because of their small size, limited processing power, and unattended deployment,
individual sensor nodes are highly prone to security compromises.
Therefore, it is important to build security in to the network architecture and protocols, so that a
sensor network can successfully operate in the presence of both component fail rues and
malicious attacks [10]. This paper consists of: related work (section 2), proposed approach
(section 3), simulation (section 4), results and conclusion.
3. Proposed approach
In this section we describe our key management approach. Our approach is a post-deployment
key management scheme which deal scalability and flexibility issues and is resistant to node
capture attacks.
All of the direct communications in wireless sensor networks can be divided into the two types
of one-to-one and one-to-many. To secure these communication sour key establishment approach
establishes the following kinds of keys:
i. Pair-wise(PW) key that is established between two neighbors to protect their for one-to-
one communications.
ii. Broadcast(BC)key that is established in order to secure the broad cast messages sent by a
node to its neighbors.
iii. Node-zase(NB)key that is established in order to secure the communication between a
node and the base station (note that this communication is not necessarily direct). A
message encrypted by this key, can only be decrypted by the base station.
Since the pair-wise and broad cast keys are essentially established among neighboring nodes ,
the first phase of key establishment is neighbor discovery. This is achieved in two steps by a pair

4
of hand shake messages. In the first step, nodes broad casts a specific type of message
containing its ID, so that every other node in s’s communication range (like r for example) can
receive it. Were fero this message as a ping message. Every node receiving the ping message
answers back to the sender(s) with a pong message containing its ID (steps1 and 2 in Figure 2)
.Nodes can then add r to its own neighbor list. After a sufficient amount of time (see Table IV
and more explanations in Section IV -B), s will discover all of its neighbors and this phase will
be finished.
When the neighbor discovery phase is over, node s computes its own node-base key and its pair-
wise keys with its neighbors as well as their broad cast keys as follows:
Nibs=Func(s||base Station Address||K)
PSs,r=F(min(s,r
)||max(s,r)||G
MK)
BSs=Func(s||G
MK)
where“||”is the concatenation operator and Func is a secure pseudo-random function usually
implemented by a hash function such MD5. GMK is a global master key that is distributed to
all nodes before deployment of the network. As we will explain later, GMK will eventually be
deleted from the memory of the nodes in order to make the approach more secure against node
capture attacks.

5
Figure 2. STEPSOFKEYESTABLISHMENTPROTOCOL
Whenthesecalculationsareover,nodeshasacompletetableofrelatedkeys.However,noder’skeytableis
notquitecompleteasitdoesnothaveanyentrycorrespondingtonodes.Thus,nodeshastosendamessage
M1containingthesekeystonoder.Obviously,M1shouldnotbesentinplain.Therefore,nodesshouldcal
culateanappropriatekeytoencryptM1withitandthensendtheencryptedversionofM1tonoder.Aproper
key,aswewillsee,isthenode-basekeyofnoderwhichcanbefollowedbysasfollows:
NibBr=Func(r||b
aseStationAddre
ss||GMK)
Having this key, node s can encrypt and send to r the key it shares with it as well as its own
broad cast key. The related messages are the following (Steps3 and 4 in Figure 2):
s→r:{s,PSsr,NIB A}NIBBs
s→r:{s,BSr,NB}NibBs
where Nib A and Nib B are two non cesto guarantee the freshness of these messages.
After sending these two messages, node s will delete the node-base key of node r from its
memory. Therefore the only non-base station node that can decrypt these message s is node r
(note that we assume the base station is secure). Node s will also delete the master key GMK
from its memory.
Step Message
1
2
3
4
5
6
s→r:{s}
r→s:{r}
s→r:{s,PSsr,NIB A}NIBBs
s→r:{s,BSr,NB}NibBs
r→s:{r,NibA,Nib B}PSsr

6
Upon receiving the keys, node r will answer back to node s by sending a message
containing the non ces NibA and NibB. This message is encrypted with the pair-wise key of s
and r (Sstep5 in Figure 2). At this point, key establishment is complete.
Notice how this message exchange enforces the scalability aspect of our protocol: related keys
can be established when a new node is added to a previously deployed network. Any new node
that joins the network (such as s) can initiate the key establishment phase by broad casting a
ping message. Following that, related keys are calculated by then ew node. Then the broad cast
keys of this added node, as well as its pair-wise keys with each of its neighbors are sent to related
neighbors, encrypted with their node-base keys. Note that using the node-base keys for this
purpose is quite an appropriate choice in order to make the protocol scalable and secure. This is
because the already available network nodes have already deleted the master key GMK from
their memory and consequently cannot use it to either calculate the keys orde cryptany message
encrypted with it. It is not a good idea touse the broad cast key of previously joined neighbor
nodes (similar to r) since other neighbors of r have that key available and can decrypt messages
encrypted with it; a fact that results in providing a looser security scheme.
The deletion of master key GMK and the temporarily calculated node-base key of r by s as
mentioned above, makes the protocol resilient to node capture attacks by reducing the
effects of capturing a node to its neighborhood and not the entire network. Since the needed
time for key establishment is negligible, we can assume that the adversary does not have
enough time to find the master key GMK before it is deleted from the memory of the nodes
(see also LEAP [4] for a similar assumption). On the other hand, newly joined nodes must
come with the master key GMK in order to calculate the cryptographic keys. Therefore, the
adversary cannot gain any use ful information by introducing new nodes to the network as a
result of not having access to GMK. In addition to that, it is important to note that if one of
the above mentioned messages in key establishment protocol is not delivered, the receiving
node will not get stuck. If node s does not receive the last message of the protocol (Step5 in
TableIII), it will not add any entry for node r in its key table.
4. Simulation
Our key management approach is implemented in Tiny Os[27] which is an event-driven
operating system commonly used on WSN nodes (motes). Results are shown in Table 1 and
Figure 3.
Table 1. Required energy and time before deleting the glbal key
Phase Neighbor discovery Key computation Key Sending
Energy (nJ) 1592640 157 38049000
Time (ms) 1000 10 10
Our key establishment approach is 10 bytes, which provides strong security (280 bit key space) fo
r sensor network applications. As a result, I kna very dense network where d = 50 will have
M≈1KB. Although this value of d is far more than enough to keep the network connected, this

7
memory over head is well within the memory capabilities of motes (MICA 2 motes have 4KB of
RAM).
During the key establishment phase, prior to deletion of the master key ,and versary has a
chancet of in d it and use it to derive all the other keys. However, this time is so small that
probability of having a nad versary capture a mote during it is minimal. Table IV shows there
lated duration t hat it takes to delete the master key from memory of a newly added mote during
its initialization phase. These results are of simulations using an internal simulator coming with
Tiny Os (Tossim).
The estimated amount of energy consumption for each phase of key establishment for the same
network (d=50) is presented in Table1 as well. This estimation was performed by multiplying
the total amount of communications by an average communications cost of 18 µJ/bit).As a
result, the estimated energy consumption of our key management scheme is approximately 0.4J
comparing to PIKE-2D [28] that is more than 8J or PIKE-3D[28] which is around 6J. This high
energy efficiency of our platform comes with a comparable cost in terms of memory over head;
it uses about 1000bytes of memory to establish and manage the keys while PIKE-2D and PIKE-
3D need around 600 bytes and 500 bytes respectively.
In our scheme the effects of having a node captured is reduced to its neighborhood, its broad
cast key and its node-base key are only keys that can be discovered by the adversary. This is a
small fraction of established keys and secure communication still remains possible in other parts
of the network.
Enegy consumption according to number of malicious nodes is shown in Figure 3. It is clear
that with larg number of malicious nodes consumption of energy is less than SEF and Hop-by-
Hop Authen- tication approchs.
Figure 2 Enegy consumption according to number of malicious nodes.
0
20
40
60
80
100
120
0 1000 2000 3000
Proposed approch
SEF
Hop-by-Hop

8
5. conclusion
In this paper we introduced a post-distribution key management approach that provides several
security services such as acceptable resistance against node capture attacks and replay attacks. It
is allows for high scalability while being easy to use and transparent to the users and light
weight. Simulation result shows that energy consumption in proposed approach with larg
number of malicious nodes in contast to other approaches is less.
References
[1] David W. Carman, Peter S. Kruus, and Brian J.Matt. Constraints and approaches for distributed
sensor network security. NAI Labs Technical Report #00-010, September 2000.
[2 ]C. Intanagonwiwat, R. Govindan, and D. Estrin. Directed diffusion: A scalable and robust
communication paradigm for sensor networks. In Proceedings of the Sixth Annual International
Conference on Mobile Computing and Networking, pages 56-67, Boston, MA, Aug. 2000. ACM
Press.
[3] Bagherinia, “OPTIMIZED TASK ALLOCATION IN SENSOR NETWORKS”, IJITMC, Vol.1,
No.3,August 2013.
[4] I. F. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, “A survey on sensor networks,” IEEE
Communications Magazine, vol. 40, no. 8, pp. 102–114, August 2002.
[5] B. C. Neuman and T. Tso, “Kerberos: An authentication service for computer networks,” IEEE
Communications, vol. 32, no. 9, pp. 33–38, September 1994.
[6] W. Diffie and M. E. Hellman, “New directions in cryptography,” IEEE Transactions on Information
Theory, vol. 22, pp. 644–654, November 1976.
[7] R. L. Rivest, A. Shamir, and L. M. Adleman, “A method for obtaining digital signatures and public-
key cryptosystems,” Communications of the ACM, vol. 21, no. 2, pp. 120–126, 1978.
[8] A. Perrig, R. Szewczyk, V. Wen, D. Cullar, and J. D. Tygar, “Spins: Security protocols for sensor
networks,” in Proceedings of the 7th Annual ACM/IEEE Internation Conference on Mobile
Computing and Networking (MobiCom), Rome, Italy, July 2001, pp. 189–199.
[9] R. Anderson and M. Kuhn, “Tamper resistance - a cautionary note,” in Proceedings of the Second
Usenix Workshop on Electronic Commerce, November 1996, pp. 1–11.
[10] C. Intanagonwiwat, R. Govindan, and D. Estrin. Directed Diffusion: AScalable and Robust
Communication Paradigm for Sensor Networks. InACM MOBICOM, 2000.
[11] H. Chan, A. Perrig, and D. Song. Random Key Predistribution Schemes for Sensor Networks. In
IEEE Symposium on Security and Privacy, 2003.
[12] W. Du, J. Deng, Y. Han, S. Chen, and P. Varshney. A Key ManagementScheme for Wireless Sensor
Networks Using Deployment Knowledge. In IEEE INFOCOM, 2004.
[13] W. Du, J. Deng, Y. Han, and P. Varshney. A Pairwise Key Predistribution Scheme for
Wireless Sensor Networks. In ACM CCS, 2003.
[14] L. Eschenauer and V. Gligor. A Key-Management Scheme for Distributed Sensor Networks. In
ACM CCS, 2002.
[15] D.LiuandP.Ning. EstablishingPairwiseKeysinDistributedSensorNetworks.InACMCCS,2003.
[16] D.LiuandP.Ning. Location-
BasedPairwiseKeyEstablishmentsforRelativelyStaticSensorNetworks.InACMSASN,2003.
[17] S. Zhu, S. Setia, and S. Jajodia. LEAP: Efficient Security Mechanism for Large-Scale Distributed
Sensor Networks. In ACM CCS, 2003.
[18] B. Awerbuch, D. Holmer, C. Nita-Rotaru, and H. Rubens. An OnDemand Secure Routing Protocol
Resilent to Byzantine Failures. InACM WiSe, 2002.
[19] Y.-C. Hu, A. Perrig, and D. B. Johnson. Ariadne: A Secure OndemandRouting Protocol for Ad Hoc
Networks. In ACM MOBICOM, 2002.

9
[20] Y.-C.Hu,D.B.Johnson, and A.Perrig.SEAD: Secure Efficient Distance Vector Routing in Mobile
Wireless AdHoc Networks .In IEEE Work-shop on Mobile Computing Systems and Applications
(WMCSA’02),2002.
[21] P. Papadimitratos and Z. Haas. Secure Routing for Mobile AdHoc Networks. In
Communication Networks and Distributed Systems Modeling and Simulation Conference (CNDS),
2002.
[22] C. Karlof and D. Wagner. Secure Routing in Wireless Sensor Networks: Attacks and
Countermeasures. In IEEE SPNA, 2002.
[23] F. Ye, H. Luo, S. Lu, and L. Zhang. Statistical En-route Filtering of Injected False Data in Sensor
Networks. In IEEE INFOCOM, 2004.
[24] S. Zhu, S. Setia, S. Jajodia, and P. Ning. An Interleaved Hop-by-Hop Authentication Scheme for
Filtering False Data in Sensor Networks. InIEEE Symposium on Security and Privacy, 2004.
[25] N. S. U. Shankar and D. Wagner. Secure verification of location claims.In ACM WISE, 2003.
[26] S. Tanachaiwiwat, P. Dave, R. Bhindwale, and A. Helmy. Secure locations: Routing on trust and
isolating compromised sensors in location- aware sensor networks. In ACM SENSYS, Poster
Abstract, 2003.
[27] J.Hill,etal,“Systemarchitecturedirectionsfornetworkedsensors”,inProceedingsofACMASPLOSIX,2000.
[28] [6] H. Chan, A. Perrig, “PIKE: Peer Intermediaries for Key Establishment in Sensor Networks”,
Proceedings of IEEE Infocom, 2005.

A Key Management Approach For Wireless Sensor Networks

More Related Content

Similar to A Key Management Approach For Wireless Sensor Networks (20)

Recently uploaded (20)

A Key Management Approach For Wireless Sensor Networks