Iaetsd identifying and preventing resource depletion attack in by Iaetsd Iaetsd

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

ISBN: 378 - 26 - 138420 - 5

Identifying And Preventing Resource Depletion Attack In Mobile Sensor Network V.Sucharitha Associate Professor jesuchi78@yahoo.com Audisankara college of engineering and technology

M.Swapna M.Tech swapna.12b2@gmail.com

ABSTRACT: Ad-hoc low-power wireless networks are inspiring research direction in sense and enveloping computing. In previous security work in this area has focused primarily on inconsistency of communication at the routing or medium access control levels. This paper explores resource depletion attacks at the navigation protocol layer, which permanent disable networks by quickly draining nodes battery power. The “Vampire” attacks are not specific protocol, but rather rely on the properties of many popular classes of routing protocols. We find that all examined protocols are vulnerable to Vampire attacks, which are demolish and difficult to detect, and easy to carry out using as few as one malicious insider send only protocol compliant messages.

near future, such as omnipresent ondemand computing power, continuous connectivity, and instantly-deployable communication for military and first responders. Such networks already monitor environmental conditions, factory performance, and troop deployment, to name a few applications. As WSNs become more and more crucial to the everyday functioning of people and organizations, availability faults become less tolerable — lack of availability can make the difference between business as usual and lost productivity, power outages, environmental disasters, and even lost lives; thus high availability of these

1.INTRODUCTION: the last couple of years wireless communication has become of such fundamental importance that a world without is no longer imaginable for many of using. Beyond the establish technologies such as mobile phones and WLAN, new approaches to wireless communication are emerging; one of them are so called ad hoc and sensor networks. Ad hoc and sensor networks are formed by autonomous nodes communicating via radio without any additional backbone infrastructure. Adhoc wireless sensor networks (WSNs) promise exciting new applications in the

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

155

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

networks is a critical property, and should hold even under malicious conditions. Due to their ad-hoc organization, wireless ad-hoc networks are particularly vulnerable to denial of service (DoS) attacks, and a great deal of research has been done to enhance survivability. While these schemes can prevent attacks on the short-term availability of a network, they do not address attacks that affect long-term available â&#x20AC;&#x201D; the most permanent denial of service attack is to entirely deplete nodesâ&#x20AC;&#x2122; batteries. This is an instance of a resource depletion attack, with battery power as the resource of interest. this paper we consider how routing protocols, even those designed to be secure, lack protection from these attacks, which we call Vampire attacks, since they drain the life from networks nodes. These attacks are distinct from previouslystudied DoS, reduction of quality (RoQ), and routing infrastructure attacks as they do not disrupt immediate availability, but rather work over time to entirely disable a network. While some of the individual attacks are simple, and powerdraining and resource exhaustion attacks have been discussed before, prior work has been mostly confined to other levels of the protocol stack, e.g. medium access control (MAC) or application layers, and to our knowledge there is little discussion, and no thorough analysis or mitigation, of routing-layer resource exhaustion attacks. Vampire attacks are not protocolspecific, in that they do not rely on design properties or implementation faults of particular routing protocols, but rather exploit general properties of protocol classes such as link-state, distance-vector, source routing and geographic and beacon routing. Neither

ISBN: 378 - 26 - 138420 - 5

do these attacks rely on flooding the network with large amounts of data, but rather try to transmit as little data as possible to achieve the largest energy drain, preventing a rate limiting solution. Since Vampires use protocol-compliant messages, these attacks are very difficult to detect and prevent. This paper makes three primary contributions. First, we thoroughly evaluate the vulnerabilities of existing protocols to routing layer battery depletion attacks. We observe that security measures to prevent Vampire attacks are orthogonal to those used to protect routing infrastructure, and so existing secure routing protocols such as Ariadne, SAODV, and SEAD do not protect against Vampire attacks. Existing work on secure routing attempts to ensure that adversaries cannot cause path discovery to return an invalid network path, but Vampires do not disrupt or alter discovered paths, instead using existing valid network paths and protocol compliant messages. Protocols that maximize power efficiency are also inappropriate, since they rely on cooperative node behavior and cannot optimize out malicious action. Second, we show simulation results quantifying the performance of several representative protocols in the presence of a single Vampire (insider adversary). Third, we modify an existing sensor network routing protocol to provably bound the damage from Vampire attacks during packet forwarding. 1.1.Wireless Adhoc Network: An ad hoc wireless network is a collection of wireless mobile nodes that self-configure to form a network without the aid of any established infrastructure, as shown in without an inherent infrastructure, the mobiles handle the

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

156

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

necessary control and networking tasks by themselves, generally through the use of distributed control algorithms. Multihop connections, whereby intermediate nodes send the packets toward their final destination, are supported to allow for efficient wireless communication between parties that are relatively far apart. Ad hoc wireless networks are highly appealing for many reasons. They can be rapidly deployed and reconfigured. They can be tailored to specific applications, as implied by Oxfordâ&#x20AC;&#x2122;s definition. They are also highly robust due to their distributed nature, node redundancy, and the lack of single points of failure.

ISBN: 378 - 26 - 138420 - 5

necessary to determine the time factor, economy n company strength. Once these things r satisfied, ten next steps are to determine which operating system and language can be used for developing the tool.

Once

the

programmers

start

building the tool the programmers need lot of external support. This support can be obtained from senior programmers, from book or from websites. Before building

the

system

the

above

consideration r taken into account for developing the proposed system. A wireless sensor network (WSN) consists

spatially

distributed

autonomous sensors to monitor physical Fig:Adhoc Network Structure

or environmental conditions, such as

Existing work on secure routing attempts to ensure that adversaries cannot cause path discovery to return an invalid network path, but Vampires do not disrupt or alter discovered paths, instead using existing valid network paths and protocol compliant messages. Protocols that maximize power efficiency are also inappropriate, since they rely on cooperative node behavior and cannot optimize out malicious action.

temperature, sound, pressure, etc. and to

2.LITERATURE REVIEW:

industrial and consumer applications,

Literature survey is the most important

such as industrial process monitoring

step in software development process.

and control, machine health monitoring,

Before

and so on.

developing

the

tool

cooperatively pass their data through the network to a main location. The more modern networks are bi-directional, also enabling control of sensor activity. The development of wireless sensor networks was motivated by military applications such as battlefield surveillance; today such networks are used in

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

157

many

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

The WSN is built of "nodes" â&#x20AC;&#x201C;

ISBN: 378 - 26 - 138420 - 5

3.IMPLIMENTATION:

from a few to several hundreds or even thousands, where each node is connected

to one (or sometimes several) sensors.

cooperatively build a Chord overlay

Each such sensor network node has

network over the sensor network. Cloned

typically

node

several

parts:

radio

prerequisite,

may

not

all

participate

nodes

this

transceiver with an internal antenna or

procedure, but it does not give them any

connection to an external antenna, a

advantage of avoiding detection. The

microcontroller, an electronic circuit for

construction of the overlay network is

interfacing with the sensors and an

independent of node clone detection. As

energy source, usually a battery or an

a result, nodes possess the information

embedded form of energy harvesting. A

of their direct predecessor and successor

sensor node might vary in size from that

in the Chord ring. In addition, each node

of a shoebox down to the size of a grain

caches information of its g consecutive

of dust, although functioning "motes" of

successors in its successors table. Many

genuine microscopic dimensions have

Chord systems utilize this kind of cache

yet to be created. The cost of sensor

mechanism to reduce the communication

nodes is similarly variable, ranging from

cost and enhance systems robustness.

a few to hundreds of dollars, depending

More importantly in our protocol, the

on the complexity of the individual

facility

sensor nodes. Size and cost constraints

contributes to the economical selection

on sensor nodes result in corresponding

of inspectors. One detection round

constraints on resources such as energy,

consists of three stages.

memory,

and

Stage 1: Initialization

The

To activate all nodes starting a new

topology of the WSNs can vary from a

round of node clone detection, the

simple star network to an advanced

initiator uses a broadcast authentication

multi-hop wireless mesh network. The

scheme to release an action message

propagation technique between the hops

including a monotonously increasing

of the network can be routing or

nonce, a random round seed, and an

flooding.

action time. The nonce is intended to

computational

communications

speed

bandwidth.

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

158

the

successors

table

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

ISBN: 378 - 26 - 138420 - 5

prevent adversaries from launching a

period, during which nodes randomly

DoS attack by repeating broadcasting

pick up a transmission time for every

action messages. The action message is

claiming message.

deﬁned by

Stage 3: Processing claiming messages A claiming message will be forwarded to its destination node via several Chord intermediate nodes. Only those nodes in

Stage 2: Claiming neighbors information

the overlay network layer (i.e., the

Upon receiving an action message, a

source node, Chord intermediate nodes,

node veriﬁes if the message nonce is

and the destination node) need to process

greater than last nonce and if the

a message,

message signature is valid. If both pass,

whereas other nodes along the path

the node updates the nonce and stores

simply route the message to temporary

the seed. At the designated action time,

targets. Algorithm 1 for handling a

the node operates as an observer that

message is the kernel of our DHT-based

generates a claiming message for each

detection protocol. If the algorithm

neighbor (examinee) and transmits the

returns NIL, then the message has

message through the overlay network with

respect

the

arrived at its destination. Otherwise, the

claiming

message will be subsequently forwarded

probability .The claiming message by observer for examinee is

to the next node with the ID that is

constructed

returned by Algorithm 1.

Criteria

determining

inspectors:

During handling a message in Algorithm 1, the node acts as an inspector if one of Where

, are locations of

,respectively.

Nodes

can

and

the following conditions is satisﬁed.

start

transmitting claiming messages at the same time, but then huge trafﬁc may cause serious interference and degrade the network capacity. To relieve this problem, we may specify a sending

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

159

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

ISBN: 378 - 26 - 138420 - 5

performance. By Algorithm 1, roughly

4.ALGORITHMS:

of all claiming messages related to a same examineeâ&#x20AC;&#x2122;s ID will pass through one

the

predecessors

the

destination. Thus, those nodes are much more likely to be able to detect a clone than randomly selected inspectors. As a result, this criterion to decide inspectors can increase the average number of witnesses at a little extra memory cost. We will theoretically quantify those performance measurements later. In Algorithm 1, to examine a message for node clone detection, an inspector will invoke Algorithm 2, which compares the message

with

inspected

messages that are buffered in the cache table. Naturally, all records in the cache table should have different examinee IDs, as implied in Algorithm 2. If detecting a clone, which means that there exist two messages satisfying

and

, the

1) This node is the destination node of

witness

the claiming message.

evidence to notify the whole network.

2) The destination node is one of the g

All integrity nodes verify the evidence

node

successors of the node. In other words,

then

broadcasts

message

the destination node will be reached in

the

and

stop communicating with the cloned

the next Chord hop. While the ďŹ rst

nodes. To prevent cloned nodes from

criterion is intuitive, the second one is

joining the network in the future, a

subtle and critical for the protocol

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

160

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

revocation list of compromised nodes

expense

IDs may be maintained by nodes

probability. The RDE protocol shares the

individually. It is worth noting that

major merit with broadcasting detection:

messages

are

Every node only needs to know and

and,

buffer a neighbor-list containing all

respectively. Therefore, the witness does

neighbors IDs and locations. For both

not need to sign the evidence message. If

detection

a malicious node tries to launch a DoS

constructs a claiming message with

attack by broadcasting a bogus evidence

signed version of its neighbor-list, and

message,

node

then tries to deliver the message to

receiving it can immediately detect the

others which will compare with its own

wicked

the

neighbor-list to detect clone. For a dense

before

network, broadcasting will drive all

and

authenticated

the

behavior

signatures

observers

integrity

verifying

and

with

ISBN: 378 - 26 - 138420 - 5

adequate

procedures,

detection

every

node

neighbors of cloned nodes to ﬁnd the

forwarding to other nodes.

attack, but in fact one witness that The DHT-based detection protocol

successfully catches the clone and then

can be applied to general sensor

notiﬁes the entire network would sufﬁce

networks, and its security level is

for the detection purpose. To achieve

remarkable, as cloned nodes will be

that in a communicatively efﬁcient way,

caught by one deterministic witness plus several

probabilistic

we bring several

witnesses.

Chord

overlap

network

protocol. First, a claiming message

incurs

needs to provide maximal h op limit, and

considerable communication cost, which

initially it is sent to a random neighbor.

may not be desired for some sensor

Then,

networks that are extremely sensitive to

(RDE),

and

presents

optimal

subsequent

helps a message go through the network

which

as fast as possible from a locally optimal

tremendously reduces communication cost

message

line. The line transmission property

challenge, we propose the randomly exploration

the

transmission will roughly maintain a

energy consumption. To fulﬁll this

directed

and

effectively construct a multicast routing

However, the message transmission over a

mechanisms

perspective. In addition, we introduce

storage

border

determination

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

161

mechanism to

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

ISBN: 378 - 26 - 138420 - 5

signiďŹ cantly reduce communication cost. We can do all of those because every node is aware of its neighbors locations, which is a basic assumption for all

where is time to live (a.k.a. message

witness-based detection protocols but

maximum hop). Since tt1 will be altered

rarely utilized by other protocols.

intermediate

transmission,

nodes

should

during not

authenticated. The observer willdeliver the claiming message r times. In each time, the node transmits it to a random neighbor as indicated. Note that can be a real

number,

and

accordingly

observer transmits its claiming message at least[r] ,up to ,[r] and on average r times. When an intermediate node receives a claiming message

4.1 Protocol Description:

launches , which is described by pseudo code in Algorithm 3, to

One round of clone detection is still

process

the

message.

During

the

activated by the initiator. Subsequently, processing, node

at the designated action time, each node

compares its own neighbor-list to the

creates its own neighbor-list including

neighbor-list in the message, checking if

the neighbors IDs and locations, which

there is a clone. Similarly, if detecting a

constitutes the sole storage consumption

clone,

of the protocol. Then, it, as an observer

the

claiming message containing its own ID, and

its

claiming

message

neighb-list. by

node

the

witness

node

will

broadcast an evidence messageto notify

for all its neighbors, starts to generate a

location,

, as an inspector,

whole

that

The

network

such

the

cloned

nodes are expelled from the sensor

network. To deal with routing, node

constructed by

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

162

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

ISBN: 378 - 26 - 138420 - 5

decreases the messageâ&#x20AC;&#x2122;s by 1 and

can be directly discarded. In our

discards the message if reaches zero;

proposal for border local determination, another parameter

4.4target range : This is used along with ideal direction to determine a target zone. When no neighbor is found in this zone, the current node will conclude that the message has reached a border, and thus throw it away.

Essentially, Algorithm 4 contains the following three mechanisms.

4.2Deterministicdirected transmission: When node receives a claiming message from previous node, the ideal direction can be calculated. In order to achieve the best effect of line transmission, the next destination node should be node , which is closest to the ideal direction. Fig:Loose source routing performance

4.3Networkborder

compared to optimal, in a network with

determination: This

takes

diameter slightly above 10. The dashed into

trend line represents expected path

the

length when nodes store logN local state,

communication cost. In many sensor

and the solid trend line shows actual

network applications, there exist outside

observed performance.

consideration

network to

shape reduce

borders of network due to physical

5.CONCLUSION:

constrains. When reaching some border

We defined Vampire attacks, a new class

in the network, the claiming message

of resource consumption attacks that use

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

163

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

ISBN: 378 - 26 - 138420 - 5

routing protocols to permanently disable

[5] J. Bellardo and S. Savage, “802.11

ad-hoc wireless sensor networks by

Denial-of-Service

depleting nodes’ battery power. These

Vulnerabilities and Practical Solutions,”

attacks do not depend on particular

Proc. 12th Conf. USENIX Security,

protocols or implementations, but rather

2003.

expose vulnerabilities in a number of

[6] D. Bernstein and P. Schwabe, “New

popular protocol classes. We showed a

AES Software Speed Records,” Proc.

number

of proof-of-concept

Ninth Int’l Conf. Cryptology in India:

against

representative

attacks

examples

Attacks:

Real

Progress in Cryptology (INDOCRYPT),

existing routing protocols using a small

2008.

number

[7] D.J. Bernstein, “Syn Cookies,”

weak

adversaries,

and

measured their attack success on a

http://cr.yp.to/syncookies.html, 1996.

randomly-generated topology of 30

[8] I.F. Blaked, G. Seroussi, and N.P.

nodes.

Smart, Elliptic Curves in cryptography, vol. 265. Cambridge Univ. , 1999.

REFERENCES:

[9] J.W. Bos, D.A. Osvik, and D. Stefan, “Fast

[1] “The Network Simulator - ns-2,”

Various Platforms,” Cryptology ePrint

http://www.isi.edu/nsnam/ns,2012.

Archive,

[2] I. Aad, J.-P. Hubaux, and E.W.

501,

and Privacy in Sensor Networks,”

MobiCom, 2004.

Computer, vol. 36, no. 10, pp. 103-105,

[3] G. Acs, L. Buttyan, and I. Vajda,

Oct. 2003.

“Provably Secure On-Demand Source

[11] J.-H. Chang and L. Tassiulas,

Routing in Mobile Ad Hoc Networks,”

“Maximum

IEEE Trans. Mobile Computing, vol. 5,

Lifetime

Routing

Wireless Sensor Networks,” IEEE/ACM

no. 11, pp. 1533-1546, Nov. 2006. Aura,

2009/

[10] H. Chan and A. Perrig, “Security

in Ad Hoc Networks,” Proc. ACM

Report

http://eprint.iacr.org, 2009.

Knightly, “Denial of Service Resilience

[4]

Implementations of AES on

Trans. Networking, vol. 12, no. 4, pp.

“Dos-Resistant

609-619, Aug. 2004.

Authentication with Client Puzzles,” Proc. Int’l Workshop Security Protocols, 2001.

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

164

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

ISBN: 378 - 26 - 138420 - 5

[12] T.H. Clausen and P. Jacquet, Optimized

Link

State

Routing

Protocol(OLSR), IETF RFC 3626, 2003. [13] J. Deng, R. Han, and S. Mishra, “Defending against Path-Based DoS Attacks in Wireless Sensor Networks,” Proc. ACM Workshop Security of Ad Hoc and Sensor Networks, 2005. [14] J. Deng, R. Han, and S. Mishra, “INSENS: Intrusion-Tolerant Routing for

Wireless

Sensor

Networks,”

Computer Comm., vol. 29, 1. 2, pp. 216230, 2006. [15] S. Doshi, S. Bhandare, and T.X. Brown, “An On-Demand Minimum Energy Routing Protocol for a Wireless Ad Hoc Network,” ACM SIGMOBILE Mobile Computing and Comm. Rev., vol. 6, no. 3, pp. 50-66, 2002. [16] J.R. Douceur, “The Sybil Attack,” Proc.

Int’l

Workshop

Peer-to-Peer

Systems, 2002. [17] H. Eberle, A. Wander, N. Gura, C.S.

Sheueling,

and

Gupta,

“Architectural Extensions for Elliptic Curve Cryptography over GF(2m) on 8bit Microprocessors,” Proc. IEEE Int’l Conf’ Application- Specific Systems, Architecture Processors (ASAP), 2005.

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

165