Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide


  1. 1. Rajkumar, Sunitha K R, Dr.H.G.Chandrakanth, / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-August 2012, pp.1684-1691 A Survey on Security Attacks in Wireless Sensor Network Rajkumar Sunitha K R Dr.H.G.ChandrakanthAssistant Professor, Dept. of ISE, Lecturer, Dept. of ISE, Sri Krishna Institute of Technology,Sambhram Institute of Technology Sambhram Institute of Bangalore Bangalore Technology BangaloreAbstract A wireless sensor network (WSN) has key consideration for most sensor network protocols.important applications such as remote Second, Sensor nodes may be deployed in publicenvironmental monitoring and target tracking. hostile locations, which make sensor nodesThis has been enabled by the availability, vulnerable to physical attacks by adversaries.particularly inrecent years, of sensors that are Generally, adversaries are assumed to be able tosmaller, cheaper, and intelligent. These sensors undetectably take control of a sensor node and extractare equipped with wireless interfaces with which all secret data in the node. Furthermore, the scale ofthey can communicate with one another to form a sensor networks is considerably large, and thenetwork. In this paper we deal with the security of network topology is dynamically adjusted, becausethe wireless sensor networks. Staring with a brief some nodes may die out of running out of energy oroverview of the sensor networks, and discusses the failure, and new nodes may join the network tocurrent state of the security attacks in WSNs. maintain desirable functionality. At last, sensorVarious types of attacks are discussed and their networks use insecure wireless communicationcountermeasures presented. A brief discussion on channel and lack infrastructure. As a result, existingthe future direction of research in WSN security is security mechanisms are inadequate, and newalso included approaches are desired.Keywords: Wireless Sensor Networks (WSNs), Since large number of sensor nodes areAttacks, Security, Threats. densely deployed, neighbor nodes may be very close to each other. Hence, multihop communication in1 INTRODUCTION sensor networks is expected to consume less power Wireless sensor networks (WSNs) are than the traditional single hop communication.innovative large-scale wireless networks that consist Furthermore, the transmission power levels can beof distributed, autonomous, low-power, low-cost, kept low, which is highly desired in covertsmall-size devices using sensors to cooperatively operations. Multihop communication can alsocollect information through infrastructureless ad-hoc effectively overcome some of the signal propagationwireless network. The development of wireless effects experienced in long-distance wirelesssensor networks was originally motivated by military communication. One of the most importantapplications such as battlefield surveillance. constraints on sensor nodes is the low powerHowever, wireless sensor networks are now used in consumption requirement. Sensor nodes carrymany civilian application areas, including limited, generally irreplaceable, power sources.environment and habitat monitoring, healthcare Therefore, while traditional networks aim to achieveapplications, home automation, and traffic control. high quality of service (QoS) provisions, sensorSecurity plays a fundamental role in many wireless network protocols must focus primarily on powersensor network applications. Because sensor conservation. They must have inbuilt trade-offnetworks pose unique challenges, security techniques mechanisms that give the end user the option ofused in conventional networks cannot be directly prolonging network lifetime at the cost of lowerapplied to WSNs because of its unique throughput or higher transmission delay. Manycharacteristics. First, sensor nodes are very sensitive researchers are currently engaged in developingof production cost since sensor networks consist of a schemes that fulfill these requirements. In this paper,large number of sensor nodes. [1] argued that the cost we present a survey of protocols and algorithmsof a sensor node should be much less than one dollar proposed thus far for sensor networks. Our aim is toin order for sensor networks to be feasible. Therefore, provide a better understanding of the current researchmost sensor nodes are resource restrained in terms of issues in this field. We also attempt an investigationenergy, memory, computation, and communication into pertaining design constraints and outline the usecapabilities. Normally sensor nodes are powered by of certain tools to meet the design objectives [2].batteries, and recharging batteries are infeasible inmany circumstances. Energy consumption becomes a 1684 | P a g e
  2. 2. Rajkumar, Sunitha K R, Dr.H.G.Chandrakanth, / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-August 2012, pp.1684-1691WSNs are intelligent compared with traditional three parts: (i) energy for the sensor transducer, (ii)sensors, and some WSNs are designed to use in- energy for communication among sensor nodes, andnetwork processing, where sensed data can be (iii) energy for microprocessor computation. Thegathered in situ and transformed to more abstract and study in [4] found that each bit transmitted in WSNsaggregated high-level data before transmission. The consumes about as much power as executing 800 tocombination of processing power, storage and 1000 instructions. Thus, communication is morewireless communication also means that data can be costly than computation in WSNs. Any messageassimilated and disseminated using smart algorithms. expansion caused by security mechanisms comes at aThe vast number of sensor nodes planned for many significant cost. Further, higher security levels inapplications also implies a major portion of these WSNs usually correspond to more energynetworks would have to acquire self organization consumption for cryptographic functions. Thus,capability. Intuitively, a denser infrastructure would WSNs could be divided into different security levelscreate a more effective sensor network. It can provide depending on energy cost [5, 6]. Memory limitations:higher accuracy and has more energy available for A sensor is a tiny device with only a small amount ofaggregation. If not properly handled, a denser memory and storage space. Memory is a sensor nodenetwork can also lead to collisions during usually includes flash memory and RAM. Flashtransmission, and network congestion. This will no memory is used for storing downloaded applicationdoubt increase latency and reduce efficiency in terms code and RAM is used for storing applicationof energy consumption. One distinguishing programs, sensor data, and intermediate results ofcharacteristic of WSNs is their lack of strong computations. There is usually not enough space toboundaries between sensing, communication and run complicated algorithms after loading the OS andcomputation. Unlike the Internet, where data application code. In the SmartDust project, forgeneration is mostly the province of end points, in example, TinyOS consumes about 4K bytes ofsensor networks every node is both a router and a instructions, leaving only 4500 bytes for security anddata source[30]. applications [4]. A common sensor type- TelosB- has a 16-bit, 8 MHz RISC CPU with only 10K RAM, 48K program memory, and 1024K flash storage [7]. The current security algorithms are therefore, infeasible in these sensors [8]. Unreliable communication: Unreliable communication is another serious threat to sensor security. Normally the packet-based routing of sensor networks is based on connectionless protocols and thus inherently unreliable. Packets may get damaged due to channel errors or may get dropped at highly congested nodes. Furthermore, the unreliable wireless communication channel may also lead to damaged or corrupted packets. Higher error rate also mandates robust error handling schemes to be implemented leading to higher overhead. In certain situation evenFigure 1: Common Wireless Sensor if the channel is reliable, the communication may notNetwork Architecture be so. This is due to the broadcast nature of wireless communication, as the packets may collide in transit2. Constraints in Wireless Sensor Networks and may need retransmission [1]. A wireless sensor network consists of alarge number of sensor nodes which are inherently Higher latency in communication: In a WSN,resource-constrained. These nodes have limited multi-hop routing, network congestion andprocessing capability, very low storage capacity, and processing in the intermediate nodes may lead toconstrained communication bandwidth. These higher latency in packet transmission. This makeslimitations are due to limited energy and physical synchronization very difficult to achieve. Thesize of the sensor nodes. Due to these constraints, it is synchronization issues may sometimes be verydifficult to directly employ the conventional security critical in security as some security mechanisms maymechanisms in WSNs. In order to optimize the rely on critical event reports and cryptographic keyconventional security algorithms for WSNs, it is distribution [9]. Unattended operation of networks:necessary to be aware about the constraints of sensor In most cases, the nodes in a WSN are deployed innodes [3]. Some of the major constraints of a WSN remote regions and are left unattended. Theare listed below. Energy constraints: Energy is the likelihood that a sensor encounters a physical attackbiggest constraint for a WSN. In general, energy in such an environment is therefore, very high.consumption in sensor nodes can be categorized in Remote management of a WSN makes it virtually 1685 | P a g e
  3. 3. Rajkumar, Sunitha K R, Dr.H.G.Chandrakanth, / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-August 2012, pp.1684-1691impossible to detect physical tampering. This makes In addition, WSNs have following specific securitysecurity in WSNs a particularly difficult task. objects:3 SECURITY REQUIREMENTS - Forward secrecy: preventing a node from The goal of security services in WSNs is to decrypting any future secret messages after it leavesprotect the information and resources from attacks the networkand misbehavior. The security requirements in WSNs - Backward secrecy: preventing a joining node frominclude: decrypting any previously transmitted secret message• Availability, which ensures that the desired network - Survivability: providing a certain level of service inservices are available even in the presence of denial- the presence of failures and/or attacksof-service attacks - Freshness: ensuring that the data is recent and no• Authorization, which ensures that only authorized adversary can replay old messagessensors can be involved in providing information to - Scalability: supporting a great number of nodesnetwork services -Efficiency: storage, processing and communication• Authentication, which ensures that the limitations on sensor nodes must be consideredcommunication from one node to another node isgenuine, that is, a malicious node cannot masquerade 5. CHALLENGESas a trusted network node Providing efficient data aggregation while• Confidentiality, which ensures that a given message preserving data privacy and integrity is a challengingcannot be understood by anyone other than the problem in wireless sensor networks due to thedesired recipients following factors:• Integrity, which ensures that a message sent from 1. Trust management in WSN is very challenging.one node to another is not modified by malicious Users in the wireless sensor networks can be veryintermediate nodes curious to learn others’ private information, and the• No repudiation, which denotes that a node cannot communication is over public accessible wirelessdeny sending a message it has previously sent links, hence the data collection is vulnerable to• Freshness, which implies that the data is recent and attacks which threaten the privacy. Without properensures that no adversary can replay old messages protection of privacy, the communication of privacy-Moreover, as new sensors are deployed and old sensitive data over civilian wireless sensor networkssensors fail, we suggest that forward and backward is considered impractical.secrecy should also be considered: 2. During in-network aggregation, adversaries can• Forward secrecy: a sensor should not be able to read easily alter the intermediate aggregation result andany future messages after it leaves the network. make the final aggregation result deviate from the• Backward secrecy: a joining sensor should not be true value greatly. Without protection of dataable to read any previously transmitted message. The integrity, the data aggregation result is notsecurity services in WSNs are usually centered on trustworthy.cryptography. However, due to the constraints in 3. Data collection over wireless sensor networks doesWSNs, many already existing secure algorithms are not rely on dedicated infrastructure. In many cases,not practical for use. the number of nodes answering a query is unknown before the data aggregation is conducted.4. Security Goals 4. Resource limited portable devices cannot afford Wireless sensor networks are vulnerable to heavy computation and communication load.many attacks because of broadcast nature of 5. The requirement on accuracy of informationtransmission medium, resource limitation on sensor collection (i.e., aggregated result) makes the existingnodes and uncontrolled environments where they are randomized privacy-preserving algorithms notleft unattended. Similar to other communication suitable. Besides the above mentioned factors, it issystems, WSNs have the following general security very challenging to protect privacy and integrity ofgoals: data aggregation simultaneously, because usually- Confidentiality: protecting secret information from privacy-preserving schemes disable traffic peerunauthorized entities monitoring mechanisms, which reduces the- Integrity: ensuring message has not been altered by availability of information in a neighborhood tomalicious nodes verify data integrity.- Data Origin Authentication: authenticating thesource of message; 6. Security Vulnerabilities in WSNs-Entity Authentication: authenticating the user / node Wireless Sensor Networks are vulnerable to/ base - station is indeed the entity various types of attacks. These attacks are mainly ofwhom it claims to be three types [87]: Attacks on secrecy and- Access control : restricting access to resources to authentication: standard cryptographic techniquesprivileged entities can protect the secrecy and authenticity of- Availability: ensuring desired service may be communication channels from outsider attacks suchavailable whenever required as eavesdropping, packet replay attacks, and 1686 | P a g e
  4. 4. Rajkumar, Sunitha K R, Dr.H.G.Chandrakanth, / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-August 2012, pp.1684-1691modification or spoofing of packets. Attacks on distributed nature, the nodes in a WSN are highlynetwork availability: attacks on availability of WSN susceptible to physical attacks [65]. The physicalare often referred to as denial-of-service (DoS) attacks may cause irreversible damage to the nodes.attacks. Stealthy attack against service integrity: in a The adversary can extract cryptographic keys fromstealthy attack, the goal of the attacker is to make the the captured node, tamper with its circuitry, modifynetwork accept a false data value. For example, an the program codes or even replace it with a maliciousattacker compromises a sensor node and injects a sensor [15]. It has been shown that sensor nodes suchfalse data value through that sensor node. In these as MICA2 motes can be compromised in less thanattacks, keeping the sensor network available for its one minute time [14].intended use is essential. DoS attacks against WSNsmay permit real-world damage to the health and 6.1.2 Link layer attackssafety of people [11]. The DoS attack usually refers The link layer is responsible forto an adversary’s attempt to disrupt, subvert, or multiplexing of data-streams, data frame detection,destroy a network. However, a DoS attack can be any medium access control, and error control [12].event that diminishes or eliminates a network’s Attacks at this layer include purposefully createdcapacity to perform its expected functions [29]. collisions, resource exhaustion, and unfairness in allocation.6.1 Denial of Service (DoS) attacks A collision occurs when two nodes attempt Wood and Stankovic have defined a DoS to transmit on the same frequency simultaneouslyattack as an event that diminishes or attempts to [11]. When packets collide, they are discarded andreduce a network’s capacity to perform its expected need to re-transmitted. An adversary mayfunction [81]. There are several standard techniques strategically cause collisions in specific packets suchexisting in the literature to cope with some of the as ACK control messages. A possible result of suchmore common denial of service attacks, although in a collisions is the costly exponential back-off. Thebroader sense, development of a generic defense adversary may simply violate the communicationmechanism against DoS attacks is still an open protocol and continuously transmit messages in anproblem. Moreover, most of the defense mechanisms attempt to generate collisions. Repeated collisionsrequire high computational overhead and hence not can also be used by an attacker to cause resourcesuitable for resource constrained WSNs. Since DoS exhaustion [11]. For example, a naïve link layerattacks in WSNs can sometimes prove very costly, implementation may continuously attempt toresearchers have spent a great deal of effort in retransmit the corrupted packets. Unless theseidentifying various types of such attacks, and retransmissions are detected early, the energy levelsdevising strategies to defend against them. Some of of the nodes would be exhausted quickly. Unfairnessthe important types of DoS attacks in WSNs are is a weak form of DoS attack [11]. An attacker maydiscussed below. cause unfairness by intermittently using the above link layer attacks. In this case, the adversary causes6.1.1 Physical layer attacks degradation of real-time applications running on The physical layer is responsible for other nodes by intermittently disrupting their framefrequency selection, carrier frequency generation, transmissions.signal detection, modulation, and data encryption[12]. As with any radio-based medium, the possibility 6.1.3 Network layer attacksof jamming is there. In addition, nodes in WSNs may The network layer of WSNs is vulnerable tobe deployed in hostile or insecure environments the different types of attacks such as: (i) spoofedwhere an attacker has the physical access. routing information , (ii) selective packet forwarding,Two types of attacks in physical layer are (i) (iii) sinkhole, (iv) Sybil, (v) wormhole, (vi) hellojamming and (ii) tampering. flood, (vii) acknowledgment spoofing etc[25, 26, 27]. These attacks are described briefly in the following:Jamming: it is a type of attack which interferes with Spoofed routing information: the most direct attackthe radio frequencies that the nodes use in a WSN for against a routing protocol is to target the routingcommunication [11,24]. A jamming source may be information in the network. An attacker may spoof,powerful enough to disrupt the entire network. Even alter, or replay routing information to disrupt trafficwith less powerful jamming sources, an adversary in the network [12]. These disruptions includecan potentially disrupt communication in the entire creation of routing loops, attracting or repellingnetwork by strategically distributing the jamming network traffic from selected nodes, extending orsources. Even an intermittent jamming may prove shortening source routes, generating fake errordetrimental as the message communication in a WSN messages, causing network partitioning, andmay be extremely time-sensitive [11]. increasing end-to-end latency. Selective forwarding: in a multi-hop network like aTampering: sensor networks typically operate in WSN, for message communication all the nodes needoutdoor environments. Due to unattended and to forward messages accurately. An attacker may 1687 | P a g e
  5. 5. Rajkumar, Sunitha K R, Dr.H.G.Chandrakanth, / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-August 2012, pp.1684-1691compromise a node in such a way that it selectively attacker is able to disseminate wrong informationforwards some messages and drops others [43]. about the status of the nodes.Sinkhole: In a sinkhole attack, an attacker makes acompromised node look more attractive to its 6.1.4 Transport layer attacksneighbors by forging the routing information The attacks that can be launched on the[13,12,11]. The result is that the neighbor nodes transport layer in a SN are flooding attack and de-choose the compromised node as the next-hop node synchronization attack.to route their data through. This type of attack makes Flooding: Whenever a protocol is required toselective forwarding very simple as all traffic from a maintain state at either end of a connection, itlarge area in the network would flow through the becomes vulnerable to memory exhaustion throughcompromised node. flooding [81]. An attacker may repeatedly make newSybil attack: it is an attack where one node presents connection request until the resources required bymore that one identity in a network. It was originally each connection are exhausted or reach a maximumdescribed as an attack intended to defeat the objective limit. In either case, further legitimate requests willof redundancy mechanisms in distributed data storage be ignored.systems in peer-topeer networks [11]. Newsome et al De-synchronization: De-synchronization refers todescribe this attack from the perspective of a WSN the disruption of an existing connection [81]. An[13]. In addition to defeating distributed data storage attacker may, for example, repeatedly spoofsystems, the Sybil attack is also effective against messages to an end host causing the host to requestrouting algorithms, data aggregation, voting, fair the retransmission of missed frames. If timedresource allocation, and foiling misbehavior correctly, an attacker may degrade or even preventdetection. Regardless of the target (voting, routing, the ability of the end hosts to successfully exchangeaggregation), the Sybil algorithm functions similarly. data causing them instead to waste energy attemptingAll of the techniques involve utilizing multiple to recover from errors which never really exist. Theidentities. For instance, in a sensor network voting possible DoS attacks and the correspondingscheme, the Sybil attack might utilize multiple countermeasures are listed in Table 1.identities to generate additional “votes”. Similarly, toattack the routing protocol, the Sybil attack would Table 1. Attacks on WSNs and countermeasuresrely on a malicious node taking on the identity of Layer Attacks Defensemultiple nodes, and thus routing multiple paths Jamming Spread-spectrum,through a single malicious node. Physical riorityWormhole: a wormhole is low latency link between messages, lower dutytwo portions of a network over which an attacker cycle,region mapping,replays network messages [12]. This link may be modechangeestablished either by a single node forwarding Collision Error-correction codemessages between two adjacent but otherwise non- Link Exhaustion Rate limitationneighboring nodes or by a pair of nodes in different Unfairness Small framesparts of the network communicating with each other. Spoofed Egress filtering,The latter case is closely related to sinkhole attack as routing authentication,an attacking node near the base station can provide a information & monitoringone-hop link to that base station via the other selectiveattacking node in a distant part of the network. forwardingHello flood: most of the protocols that use Hello Sinkhole Redundancy checkingpackets make the naïve assumption that receiving Sybil Authentication,such a packet implies that the sender is within the Network monitoring,redundancradio range of the receiver. An attacker may use a yhigh-powered transmitter to fool a large number of Wormhole Authentication,nodes and make them believe that they are within its probingneighborhood [12]. Subsequently, the attacker node Hello Flood Authentication, packetfalsely broadcasts a shorter route to the base station, leashes by usingand all the nodes which received the Hello packets, geographic andattempt to transmit to the attacker node. However, temporal infothese nodes are out of the radio range of the attacker. Ack. flooding Authentication, bi-Acknowledgment spoofing: some routing algorithms directional linkfor WSNs require transmission of acknowledgment authenticationpackets. An attacking node may overhear packet verificationtransmissions from its neighboring nodes and spoof Transpor Flooding Client puzzlesthe acknowledgments thereby providing false t De- Authenticationinformation to the nodes [12]. In this way, the synchronizatio 1688 | P a g e
  6. 6. Rajkumar, Sunitha K R, Dr.H.G.Chandrakanth, / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-August 2012, pp.1684-1691 n server, Eavesdropping on these messages prove more effective for an adversary. Traffic analysis: In order to make an effective attack on privacy, eavesdropping should be combined with a traffic analysis. Through an effective analysis of6.2 Attacks on secrecy and authentication traffic, an adversary can identify some sensor nodes There are different types of attacks under with special roles and activities in a WSN. Forthis category as discussed below. example, a sudden increase in message communication between certain nodes signifies that6.2.1 Node replication attack those nodes have some specific activities and events In a node replication attack, an attacker to monitor. Deng et al have demonstrated two typesattempts to add a node to an existing WSN by of attacks that can identify the base station in a WSNreplication (i.e. copying) the node identifier of an without even underrating the contents of the packetsalready existing node in the network [17]. A node being analyzed in traffic analysis [21].replicated and joined in the network in this manner Camouflage: An adversary may compromise acan potentially cause severe disruption in message sensor node in a WSN and later on use that node tocommunication in the WSN by corrupting and masquerade a normal node in the network. Thisforwarding the packets in wrong routes. This may camouflaged node then may advertise false routingalso lead to network partitioning, communication of information and attract packets from other nodes forfalse sensor readings. In addition, if the attacker gains further forwarding. After the packets start arriving atphysical access to the entire network, it is possible the compromised node, it starts forwarding them tofor him to copy the cryptographic keys and use these strategic nodes where privacy analysis on the packetskeys for message communication from the replicated may be carried out systematically.node. The attacker can also place the replicated node It may be noted from the above discussionin strategic locations in the network so that he could that WSNs are vulnerable to a number of attacks ateasily manipulate a specific segment of the network, all layers of the TCP/IP protocol stack. However, aspossibly causing a network partitioning. pointed out by authors in [22], there may be other types of attacks possible which are not yet identified.6.2.2 Attacks on privacy Securing a WSN against all these attacks may be a Since WSNs are capable of automatic data quite challenging task.collection through efficient and strategic deploymentof sensors, these networks are also vulnerable topotential abuse of these vast data sources. Privacy 7 FUTURE TRENDSpreservation of sensitive data in a WSN is Though significant research in WSNs andparticularly difficult challenge [18]. Moreover, an mobile computing continues, issues concerning theadversary may gather seemingly innocuous data to enablement of seamless and transparent interactionderive sensitive information if he knows how to between each domain need to be resolved. A numberaggregate data collected from multiple sensor nodes. of issues are now identified. Communication protocolThis is analogous to the panda hunter problem, where issues: In order for a PDA (Personal Digitalthe hunter can accurately estimate the location of the Assistants) to communicate with a sensor network, itpanda by monitoring the traffic [19]. is necessary that both PDAs and WSNs use the same communication protocol. At present, off the shelf The privacy preservation in WSNs is even PDAs have the Bluetooth protocol for short rangemore challenging since these networks make large communication provided. Unfortunately, studies ofvolumes of information easily available through the Bluetooth architecture (Leopold, 2003) showedremote access mechanisms. Since the adversary need the unsuitability of such a protocol for wirelessnot be physically present to carryout the surveillance, sensor networks. On the other hand, although recentthe information gathering process can be done advances propose a vast number of protocols tailoredanonymously with a very low risk. In addition, to WSNs, the communication compatibility betweenremote access allows a single adversary to monitor the two technologies is still an open issue. Ontologymultiple sites simultaneously [20]. Following are issues: Such kinds of issues arise after PDAs andsome of the common attacks on sensor data privacy sensors agree which communication protocol to use.[18,20]: In the context of knowledge sharing between PDAsEavesdropping and passive monitoring: This is and sensors at the application layer, they should agreemost common and easiest form of attack on data with the specification of a conceptualization, alsoprivacy. If the messages are not protected by known as an ontology. Although some researchcryptographic mechanisms, the adversary could propose the study of semantic techniques for wirelesseasily understand the contents. Packets containing sensor networks (Whitehouse, 2006), acontrol information in a WSN convey more comprehensive methodology of PDA/sensorinformation than accessible through the location interaction is still an open issue to be addressed. Trust management issues: Requests of m-commerce- 1689 | P a g e
  7. 7. Rajkumar, Sunitha K R, Dr.H.G.Chandrakanth, / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-August 2012, pp.1684-1691related information from sensors to PDAs and vice communication security in wireless ad-hocversa raises issues of trust management. In fact, sensor networks”, In Proceedings of 11thsensors should trust the quality of service offered by IEEE International Workshop on Enablingthe PDA protocol. On the other side, PDAs should Technologies: Infrastructure fortrust sensors when, for example, product availability Collaborative Enterprises (WETICE’02),or machinery condition are sent to a PDA. While the 2002, pp. 139-144.latter case can be considered as an instance of [6] L. Yuan and G. Qu, “Design spaceinternet trust management, the former case needs to exploration for energy-efficient secureconsider the issue of memory capability constraints sensor networks”, In Proceedings of IEEEof sensors. Procedures for realizing trust management International Conference on Application-on individual sensors, for example, through Specific Systems, Architectures, andintelligent agent technologies, need further research. Processors, July 2002, pp. 88-100.The big “umbrella” of trust management also [7] http://www.xbow.com/wireless_home.aspx,includes more specific issues of security. In fact, the 2006.multi-hop routing of WSNs together with the [8] A. Perrig, R. Szewczyk, V. Wen, D.E.relatively simple architecture of sensors pose an Culler, and J.D. Tygar,“SPINS: Securityinherent risk, as an attacker may only need to protocols for sensor networks”,compromise one device to compromise the security WirelessNetworks, Vol.8 , No. 5, pp. 521-of the entire network. This concern is amplified in 534, September 2002.applications like m-commerce where private [9] J.A. Stankovic et al, “Real-timecredentials must be fully safely encoded. communication and coordination in embedded sensor networks”, In Proceedings8 CONCLUSIONS of the IEEE, Vol. 91,No. 7, , pp. 1002-1022, Security is becoming a major concern for July 2003.energy constrained wireless sensor network because [10] E. Shi and A. Perrig, “Designing secureof the broad security-critical applications of WSNs. sensor networks”, Wireless CommunicationThus, security in WSNs has attracted a lot of Magazine, Vol. 11, No. 6, pp. 38-43,attention in the recent years. The salient features of December 2004WSNs make it very challenging to design strong [11] A.D. Wood and J.A. Stankovic, “Denial ofsecurity protocols while still maintaining low service in sensor networks”, IEEEoverheads. Computer, Vol. 35, No. 10, pp. 54-62, 2002.In this paper, we introduce sensor networks, its [12] C. Karlof and D. Wagner, “Secure routing inrelated security problems, threats, risks and wireless sensor networks: Attacks andcharacteristics. Network security for WSNs is still a countermeasures”, In Proceedings of the 1stvery fruitful research direction to be further explored. IEEE International Workshop on Sensor Network Protocols and Applications, MayREFERENCES 2003, pp. 113-127. [1] I. F. Akyildiz, W. Su, Y. [13] J. Newsome, E. Shi, D. Song, and A. Perrig, Sankarasubramaniam, and E. Cayirci. A “The Sybil attack in sensor networks: survey on sensor networks. IEEE Analysis and defenses”, In Proceedings of Communications Magazine, 40(8):102–114, the 3rd International Symposium on 2002. Information Processing in Sensor Networks, [2] Wireless sensor networks: a survey I.F. pp. 259-268, ACM Press 2004. Akyildiz, W. Su*, Y. Sankarasubramaniam, [14] C. Hartung, J. Balasalle, and R. Han, “Node E. Cayirci compromise in sensor networks: The need [3] D.W. Carman, P.S. Krus, and B.J. Matt, for secure systems”, Technical Report CU- “Constraints and approaches for distributed CS-988-04, Department of Computer sensor network security”, Technical Science, University of Colorado at Boulder, Report 00-010, NAI Labs, Network 2004. Associates Inc., Glenwood, MD, 2000. [15] X. Wang, W. Gu, S. Chellappan, Dong [4] J. Hill, R. Szewczyk, A. Woo, S. Hollar, Xuan, and Ten H. Laii, “Search-based D.E. Culler, and K. Pister, “System physical attacks in sensor networks: architecture directions for networked Modeling and defense, Technical report, sensors”, In Proceedings of the 9th Department of Computer Science and International Conference on Architectural Engineering, Ohio State University, Support for Programming Languages and February 2005. Operating Systems, New York, ACM Press, [16] J. Douceur, “The Sybil attack”, In 2000, pp. 93-104. Proceedings of the 1st International [5] S. Slijepcevic, M. Potkonjak, V. Tsiatsis, S. Workshop on Peer-to-Peer Systems Zimbeck, and M.B. Srivastava, “On (IPTPS’02),February 2002. 1690 | P a g e
  8. 8. Rajkumar, Sunitha K R, Dr.H.G.Chandrakanth, / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue4, July-August 2012, pp.1684-1691 Systems (SensSys’04), Baltimore, MD,[17] B. Parno, A. Perrig, and V. Gligor, November 2004, pp. 162-175. “Distributed detection of node replication [29] N. Xu, S. Rangwala, K. Chintalapudi, D. attacks in sensor networks”, In Proceedings Ganesan, A. Broad, R. Govindan, and D. of IEEESymposium on Security and Privacy, Estrin, “A Wireless Sensor Network for May 2005. Structural Monitoring,” Proceedings of the[18] M. Gruteser, G. Schelle, A. Jain, R. Han, ACM Conference on Embedded Networked and D. Grunwald, “Privacy-aware location Sensor Systems, Baltimore, MD, November sensor networks”, In Proceedings of the 9th 2004. USENIX Workshop on Hot Topics in [30] D. Culler, D. Estrin, and M. Srivastava, Operating Systems, (HotOS IX), 2003. “Overview of Sensor Networks,” IEEE Computer,August 2004.[19] C. Ozturk, Y. Zhang, and W. Trappe, “Source-location privacy in energy- constrained sensor network routing “, In Proceedings of the 2nd ACM Workshop on Security of Ad Hoc and Sensor Networks, 2004.[20] H. Chan and A. Perrig, “Security and privacy in sensor networks”, IEEE Computer Magazine, pp. 103-105, 2003.[21] J. Deng, R. Han, and S. Mishra, “Countermeasures against traffic analysis in wireless sensor networks”, Technical Report CU-CS-987-04, University of Colorado at Boulder, 2004.[22] A. Perrig, J. Stankovic, and D. Wagner, “Security in wireless sensor networks”, Communications of ACM , Vol 47, No. 6, pp. 53-57, 2004[23] X. Wang, W. Gu, K. Schosek, S. Chellappan, and D. Xuan, “Sensornetwork configuration under physical attacks,”, Technical report (OSU-CISRC-7/04-TR45), Department of Computer Science and Engineering, Ohio State University, July 2004.[24] E. Shi and A. Perrig, “Designing secure sensor networks”, Wireless Communication Magazine, Vol. 11, No. 6, pp. 38-43, December 2004.[25] Douceur, J. “The Sybil Attack”, 1st International Workshop on Peer-to-Peer Systems (2002).[26] Newsome, J., Shi, E., Song, D, and Perrig, A, “The sybil attack in sensor networks: analysis & defenses”, Proc. of the third international symposium on Information processing in sensor networks, ACM, 2004, pp. 259 – 268[27] Karlof, C. and Wagner, D., “Secure routing in wireless sensor networks: Attacks and countermeasures”, Elseviers Ad Hoc Network Journal, Special Issue on Sensor Network Applications and Protocols, September 2003, pp. 293-315[28] C. Karlof, N. Sastry, and D. Wagner, “TinySec: a link layer security architecture for wireless sensor networks”, In 2nd ACM Conference on Embedded Networked Sensor 1691 | P a g e