International Journal of Scientific & Engineering Research, Volume 3, Issue 9, September-2012 1
ISSN 2229-5518
Analysis of Various Attacks in Routing Protocols for Wireless Sensor Network
Muhammad Danish Qureshi, Hassam Ishtiaq, Arsalan Farhat, Obaid Rehman
Abstract— Demand of wireless sensor network is increasing for various smart applications and provides unlimi ted opportunities. But with the increase in usage of W SN it imposes some challenges including limited power resources and security threats which needs to be identified and its mitigation techniques requires further development. In this paper we analyzed different routing protocols including SEER, Direct Diffusion, Tiny OS Beaconing, Geographic routing and Rumor Routing and its various attacks on these routing protocols. Our work also analyzes the design issues of W SN by comparing different design parameters including power usage, scalability, data aggregation, overhead, fault tolerance and quality of service. After analysis of these protocols we present its comparison which shows the important features required for consideration while suggesting routing prot ocols for W SN. Furthermore, as a result optimum protocol is suggested in term of security and its energy efficiency.
Index Terms— Design parameter, Evasdropping, QOS, Routing protocol, Sybil attack, Wormhole, W SN.
—————————— ——————————
Wireless sensor network is consisting of different nodes that maintain a cooperative network. Every node has RF (Radio Frequency) transceiver, power source, various sensors and actuators and processing capabilities. Nodes transmit and receive data wirelessly and it can auto- matically self organized into Ad hoc fashion after spreading into network. WSN is a need of smart environment that wants information about its internal work and about surrounding. This step includes in building, home, industrial and shipboard and transport systems automation. There are many applica- tions of WSN in real world. It can be implemented for envi- ronment monitoring, for military operation, in factories for maintenance, health monitoring and even in bodies of patient. As every network need a routing protocol for implementation and running of network, WSN also has some routing proto- cols. Routing in WSN is different from routing in other ordi- nary networks. In WSN there is infrastructure, wireless links are unreliable that may fail, routing protocols of WSN have to meet strict energy saving and security requirements [1], [2], [7].
WSN uses broadcast nature, so an attacker having strong receiver can intercept and eavesdrop data. It can attract data like location of node, Node ID, Message ID etc [12].
Fig. 1 Eavesdropping
Attacker wants to disrupt, corrupt or destroy a network. Its task is to jam a node or set of nodes. It simply transmits radio signals that create interference with the radio frequency used by the network [13].
Fig. 2 DOS
Attacker receives the message and then forwards it to other node after tempering it. So when data reaches to the sink is not useful [13].
Fig. 3 Message tempering
IJSER © 2012
The research paper published by IJSER journal is about Analysis of Various Attacks in Routing Protocols for Wireless Sensor Network 2
ISSN 2229-5518
Attacker node receives the message but it refuses to forward it and simply drop it, ensuring that the message cannot be prop- agate further. Attacker acts like a hole and don’t forward packet [13].
Fig. 6 Wormhole attacks
Fig. 4 Selective forwarding
Attacker is very powerful in this attack. It attracts all the nodes from a particular area and creates itself as a sink for other nodes. Now it can change the data as well as drop it. It also leads to the other attacks [14].
Fig. 5 Sinkhole attacks
Attacker receives the information by making a path of low latency link between two different parts of the network. It forwards the message of one part to other part for making confusion. It also allows sinkhole to occur as the attacker on other side of the wormhole can shows to have a high quality path to the sink. An attacker that is located near to the sink may completely disrupt routing by creating a best location wormhole [13].
————————————————
Muhammad Danish Qureshi is currently pursuing bachelor degree pro- gram in electric engineering in Sarhad University of Science & IT, Pakis- tan, PH-+923333633945. E-mail: mdanishqureshi@yahoo.com
Hassam Ishtiaq, Arsalan Farhat, Obaid Rehman is currentlyaffiliated with
Sarhad University of Science & IT, Pakistan.
Attacker shows multiple identities to other nodes in the net- work. It can decrease the effectiveness of fault tolerant such as storage distribution, disparity and multi path routing, and maintenance of topologies. By multiple identities it makes such a condition that every message should be go through that attacker [13].
Fig. 7 Sybil attacks
Following are the some design parameters which has much importance while suggesting a protocol:
Sensor nodes have much chances of failure due to odd and difficult deployment environments. Thus, sensor nodes should have ability of fault tolerant and have the abilities of testing themselves, auto celebrate, auto repair and self recovering [3].
In WSN, every application may have different quality of ser- vice (QoS) requirements in terms of delivery latency path and packet loss. Network protocol design should consider the QoS requirements of applications [3].
Sensor nodes use the battery for power and it is very difficult to charge or recharge their batteries, so it is very important to control and reduce the power consumption of nodes so that the lives of the sensor nodes increases, as well as the network will live for longtime [5].
The sensor nodes in WSN are in the order of tens, hundreds,
IJSER © 2012
The research paper published by IJSER journal is about Analysis of Various Attacks in Routing Protocols for Wireless Sensor Network 3
ISSN 2229-5518
or thousands, routing protocols designed for sensor networks should be scalable to different network sizes [5].
Routing protocols designed for WSN must have error control and correction functions to ensure reliability of data delivery over noise and time varying channels [3].
WSN have limited bandwidth resources. Routing protocols designed for sensor networks must use the whole bandwidth very efficiently to improve channel utilization [5].
Overhead is also a special design parameter. In many cases overheads are not allowed. Overhead may be in form of bandwidth, space area, large number of nodes etc [3].
There are 7 major types that includes (1) Centric Protocols (2)Hierarchical Protocols (3)Location based Protocols (4)Mobility Based Protocols (5)Multi path based Protocols (6) Heterogeneity based Protocols (7) QoS based Protocols. In next section we will study some of the routing protocols of WSN and compare them on the basis of security and design parameters. We will compare the following protocols:
SEER (Data centric, Location based) Directed Diffusion (Data centric) TinyOS beaconing (Hierarchical)
Geographic routing (Location based Protocols) Rumor Routing (Data centric)
SEER uses multi paths between two nodes one by one to in- crease the life of the network. Multi path specially uses for two functions. One is for balancing of load and the other is for making data delivery reliable. The sink first broadcasts a ND (Neighbors Discovery) packet to the network. The sink broad- casts another packet NC. Information about nodes is collected during the previous broadcasting (Energy levels of the node etc) [18], [19].
Fig. 8 SEER Operation
When node gets the NC message, it replies a NCR packet to the sink. The NCR packet has the location and the information about the node and the list of all addresses of its neighbor nodes. Path is only selected by the sink only. Base Station or sink select different paths after a certain period ac- cording to current energy level of node. This gives the surety that if the attacker advertises; it has no affect on path selection process. After the sink selects another path, the attacker can- not attack any more [17].
Fig. 9 SEER defending
Directed Diffusion is a type of data centric protocol for com- municating data in sensor network. Here if base station wants data he has to first broadcast interest packet. An interest pack- et is a request that should be done by the base station. Every node keeps moving with that interest packet until that packet reaches to the node that is interested or satisfies the interest condition. Each node that gets the interests packet set up a slope value toward the origin node. A slope value contains direction and attributes value. As shown in Figure when node
IJSER © 2012
The research paper published by IJSER journal is about Analysis of Various Attacks in Routing Protocols for Wireless Sensor Network 4
ISSN 2229-5518
―B‖ gets an interest packet from node ―A‖, it includes ―A (Δ)‖ in its slope value. Similarly when node ―C‖ receives an inter- est from node ―A‖ and node ―D‖, it includes ―B (2Δ)‖ and ―D (2Δ)‖. When interest packet reached to the place of event, then the sink strictly forces one or more neighbor’s nodes to reply at a higher data rate. Also sink can negatively force the nodes to leave high data flow that are not in use at that event [15], [10].
In that scenario attacker can eavesdrop the interest.
After an attacker gets an interest packet from a sink, it can
simply reply with the message that ―I am the node that is in-
terested‖. When the reply for that interest is sent, then after
sink the attacker would also be receiving them.
Fig. 10 Directed Diffusion
When source generate data events, an attacker node
can attack a data flow and cause to suppression. It is denial of
service attack.
It is a hierarchical Protocol. It builds a tree with a sink as the parent for all the other nodes in the sensor network. After a certain period the sink always broadcasts new route information to the neighbor nodes, nodes that receives that information also forward it to their neighbors. Nodes that get the new route path information mark the sink as its parent and rebroadcast the update [14].
That is a very simple protocol that makes it so much
susceptible to all the attacks discussed above. Since new
routing path information are not authenticated, so an attacker
can easily say that or claim himself as a parent of the all other
nodes of sensor network. An attacker who is interested in ea-
vesdropping or suppressing packets in a particular area can
easily do it by creating a combination of wormhole or sinkhole
attack [14], [6].
Fig. 11 TinyOS beaconing
GEAR (Geographic and Energy Aware Routing) use nodes location addresses to inform node that is near to it about inter- est and also express geographic message destinations to effi- ciently propagate queries and route replies in the WSN. A black hole is created when there is no further node near to the event other than it. Whenever a message is received by a node it checks the nearby node that is closest to the target. If there is more than one node, then it will choose that is much closer to the target [6], [9].
And if there is only one node it forwards the packet to that node. If there is no node near to that, then it will choose the node using a function known from above, that this protocol uses location and energy levels information. In this protocol an attacker can increase his strength of attack by creating a Sybil attack. Attacker can create Sybil attack by covering up the target node and its path totally by number of compromis- ing nodes [13].
Fig. 12 Geographic routing
As shown in figure that an attacker ―A‖ have actual address of location (3, 2) also advertized some fake locations that actually not exists (i.e. A1, A2 and A3). Now if ―B‖ wants to communi- cate with node ―C‖ at (0, 2) it will communicate through ―A3‖ because of fake advertisement. So that type of communication is an overhead that is handled by an attacker. Attacker can easily do selective forwarding here [14].
IJSER © 2012
The research paper published by IJSER journal is about Analysis of Various Attacks in Routing Protocols for Wireless Sensor Network 5
ISSN 2229-5518
Rumor routing is such a protocol that actually inter- sects the paths of queries and data events. It is very efficient at the situation where high flooding is not possible. It does not allow the whole network to match or found the event (inter- est). Rumor routing uses a healthy living message or packet knows as AGENT. Source node generates an agent whenever it observes an event. Agent propagates along the whole net- work and forwards the information about the event and re- mote nodes [16], [11].
Fig. 13 Rumor Routing
Agent have the information like list of events, hope path to the event, list of nodes from which it is coming and a TTL (Time to Live) field. When it reaches to new node it actually tells the node about the event and adds that node in an event list. It also decreases the TTL field by one at every new node visit. Now if TTL field is more than zero it selects the agent’s next hope from its neighbors in the table and subtracts the pre- viously visited node from table. Similarly sink or base station creates an agent to propagate the queries into network. So a point comes when both the paths (i.e. queries path and data event path will intersect) is our desired point. After that a final path is decided for communication between event and base station [15], [16].
After studying different protocols deeply we analyze that there is no such a protocol that gives 100% surety of security. But there are some protocols that are secure enough from oth- er protocols. Below are the tables of comparision of routing protocols on the basis of attacks and design parameters.
TABLE 1
COMP ARISION ON THE B ASIS OF ATTACKS
[5], [6], [18], and Analysis
TABLE 2
COMP ARISION ON THE B ASIS OF DESIGN PARAMETERS
[6], [20] and Analysis
After studying deeply these five protocols we realize that SEER is a best protocol for security as well as energy efficiency as compared to other routing protocols, As in SEER, the path is changing continuously for next transmission. So it monitors each nodes energy level and position. If attackers wants to take control on network then there should be number of at- tackers are required to create a path between sink and event. After comparing in both cases security and design parameters we conclude that SEER is the best in both cases. SEER is ac- tually created for energy efficiency but it also gives best securi- ty.
In the name of Allah, the Most Gracious and the Most Merciful Alhamdulillah, all praises to Allah for the strengths and His blessing in completing this research paper. After that I would like to thanks my beloved parents ―Izhar Ahmad Qureshi‖ who supported me alot. My work is also supported by my supervisor ―Obaid Rehman‖ and ―Sarhad University of Science & IT‖.
[1] John A. Stankovic, ―Wireless Sensor Networks‖, Department of
Computer Science University of Virginia Charlottesville, Virginia
22904, June 2006.
[2] F. L. LEWIS, Associate Director for Research, ―Wireless Sensor Networks1‖, The University of Texas at Arlington 7300 Jack Newell Blvd. S Ft. Worth, Texas 76118-7115 August 2009.
[3] Kay Romer and Friedemann Mattern, ―The Design Space of Wire-
less Sensor Networks‖ IEEE Wireless Communications, pp. 54-61, December 2004.
IJSER © 2012
The research paper published by IJSER journal is about Analysis of Various Attacks in Routing Protocols for Wireless Sensor Network 6
ISSN 2229-5518
[4] John Paul Walters et al, A Survey: ―Wireless Sensor Network Securi- ty‖, Department f Computer Science Wayne State University, A u- gust 2008.
[5] Shio Kumar Singh, M P Singh et al, A Survey: ―Routing Protocols in Wireless Sensor Networks‖, Maintenance Engineering Department Jamshedpur, Jharkhand, India, 2011.
[6] S.K. Singh, M.P. Singh and D.K. Singh, ―A survey of Energy-
Efficient Hierarchical Cluster-based Routing in Wireless Sensor
Networks‖, International Journal of Advanced Networking and
Application (IJANA), vol. 02, issue 02, pp. 570–580, October 2010.
[7] S. Misra et al, ―Guide to Wireless Sensor Networks, Computer Communications and Networks‖, DOI: 10.1007/978-1-84882-218-4 4, Springer Verlag London Limited, 2009.
[8] E. Zanaj, M. Baldi, and F. Chiaraluce, ―Efficiency of the Gossip Al-
gorithm for Wireless Sensor Networks‖, In Proceedings of the 15th
International Conference on Software, Telecommunications and Computer Networks (SoftCOM), Split–Dubrovnik, Croatia, Septem- ber, 2007.
[9] Yu, R. Govindan, and D. Estrin, "Geographical and energy aware
routing: A recursive data dissemination protocol for wireless sensor networks", Technical Report UCLA/CSD-TR-01-0023, UCLA Com- puter Science Department, May 2001.
[10] C. Intanagonwiwat, R. Govindan, and D. Estrin, "Directed diffusion: A scalable and robust communication paradigm for sensor net- works", Proceedings ACM MobiCom'00, Boston, MA, pp. 56-67, August 2004.
[11] D. Braginsky and D. Estrin, "Rumor routing algorithm in sensor
networks", Proceedings ACM WSNA, in conjunction with ACM MobiCom'02,Atlanta, GA, pp. 22-31, Sept. 2002.
[12] J. Yick, B. Mukherjee and D. Ghosal ―Wireless Sensor Network Sur-
vey‖, Computer networks, Vol. 52, no. 12, pp 2292- 2330, 2008.
[13] R. El-Kaissi, A. Kayssi, A. Chehab, and Z. Dawy., DAWWSEN: ―A Defence mechanism Against Wormhole attacks in Wireless Sensor Networks‖, 2009.
[14] C. Karlof and D. Wagner, ―Secure Routing in Wireless Sensor Net- works: Attacks and Countermeasures‖, University of California at Berkeley, 2010.
[15] S. Shanmugham, ―Secure Routing in Wireless Sensor Networks‖
Scholarly Paper Advisor: Dr. Jens-Peter Kaps, 2011.
[16] D. Braginsky and D. Estrin, ―Rumour routing algorithm for sensor
networks‖, in First ACM International Workshop on Wireless Se n-
sor Networks and Applications, 2002.
[17] Ch. Phani kanth and Pardhav Lingam, ―Secure Routing Protocols in
Wireless Sensor Networks‖, Software Developer Technology, Gun- tur University, India, 2008.
[18] Nidal Nasser and Yunfeng Chen, ―Secure Multipath Routing Proto-
col for Wireless Sensor Networks‖, ICDCSW 07 Proceedings of the
27th International Conference, 2006.
[19] Hancke, GP & Leuschner, CJ 2007, ―SEER : a simple energy efficient routing protocol for wireless sensor networks‖, South African Com- puter Journal, vol. 39, pp. Dec 2007.
[20] Rajashree. V. Biradar and V.C. Patil ―Special Issue on Ubiquitous
Computing Security Systems‖, UbiCC Journal Volume 4,. May 2011.
IJSER © 2012