Get documents about "Packet Forwarding Encouragement Scheme in a Wireless Sensor Network
Description
IJCSIS, call for paper, journal computer science, research, google scholar, IEEE, Scirus, download, ArXiV, library, information security, internet, peer review, scribd, docstoc, cornell university, archive, Journal of Computing, DOAJ, Open Access, April 2011, Volume 9, No. 4, Impact Factor, engineering, international, proQuest, computing, computer, technology
Document Sample
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 4, 2011
Packet Forwarding Encouragement Scheme in a
Wireless Sensor Network
Praveen Kaushik Jyoti Singhai
Department of CSE Department of ECE
MANIT MANIT
Bhopal, India Bhopal, India
praveenkaushikmanit@gmail.com j.singhai@gmail.com
Abstract- The most common issue in wireless sensor networks is lazy nodes and constrained nodes (e.g. energy constrained or
limited availability of energy within network nodes. A widely field strength constrained). A routing algorithm plays a crucial
used energy-saving technique is to place nodes in sleep mode, role in transmission of data among nodes. Also dynamic
corresponding to low-power consumption as well as to reduce behavior of nodes causes several challenges in design of
operational capabilities. To conserve energy, sometime nodes
may try to maximize the benefits of its own without participating
routing scheme. A lot of algorithms have been proposed
in routing services (e.g. without forwarding packets of other claiming fairness in routing. DSR is most widely adapted
nodes). To conquer with this problem, several schemes have been algorithm for routing. But the original algorithm is known to
investigated in the literature. In this paper such nodes are fail in a scenario containing non cooperative nodes. In [1]
detected and evicted from the network using rating based proposed a mechanism for identifying and isolating selfish
multipath dynamic source routing. In our approach after route nodes to attain fairness. However after decaling a node selfish,
discovery, multiple paths are returned by the destination node, source node needs to identify alternate path using Route
which are stored by source node in route cache. Once source Discovery. This causes a lot many packet drops before it
node identifies such node in its path based on its rating threshold, identifies alternate path.
an alternative path is selected from the route cache. In sensor
network a node may be overused due to network topology. In
Multipath Source Routing [10] can increase performance by
order to mitigate this problem, this paper uses a minimum providing applications the liberty to use multiple paths during
battery cost routing (MBCR).The simulation results for rating packet forwarding. Also it stores multiple paths required. With
based multipath dynamic source routing indicates significant Multipath Source Routing data can be forwarded on arbitrary
improvement in aggregate network throughput. routes, which makes it very easy to forward data to multiple
paths without use of path using Route Discovery. This causes
Keywords- minimum cost; multipath ; reluctant node; wireless a lot many packet drops before it identifies alternate path.
sensor network;
II. NODE MODEL
I. INTRODUCTION
A. Malicious nodes
A Wireless Sensor Network (WSN) is a collection of low-end
Nodes which drops packet with the intention to cause network
devices whose size can range from a few hundred sensors to a
attack.
few hundred thousand or possibly more. The sensors do not
rely on any pre-deployed network architecture. The power B. Reluctant nodes
supply of each individual sensor is provided by a battery, Reluctant nodes try to save their own resources since
whose consumption for both communication and computation resources are very constrained in wireless network. Reluctant
activities must be optimized. A WSN may consist of many nodes may decide to conserve their resources by not
nodes without any grouping hierarchy. Nodes can forwarding data packets for other nodes:
communicate using any wireless technology such as Bluetooth, This can be achieved in two ways:
ZigBee or IEEE standards. 1) Reluctant node of type 1: In this model the reluctant
In wireless sensor networks, two nodes can exchange data node do not participate in the Route Discovery phase of DSR
when they are located within one another’s communication protocol[10] . Such reluctant nodes not only drop packets that
range. In WSN all nodes cooperates and forward packet for have a source address or destination address different from
each other as a router. Here it is possible that nodes may not
these reluctant nodes but also drop all RREQ packet they
be within the communication range of each other. At this stage,
receive or not forward RREP packet to some destination. If the
nodes extend the transmission range by multi hop packet
node does not participate in the route discovery process, then
forwarding. In ad hoc network nodes can be of two types’
cooperative nodes and inactive nodes. Cooperative nodes there will be no route with that reluctant node. The outcome is
comply with the standard at all times. Inactive nodes include that these reluctant nodes will be isolated i.e. these nodes will
151 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 4, 2011
never participate in packet forwarding. A reluctant node of this make a decision whether the node is regular or misbehaved,
type uses the node energy only for its own communication. while trust rating is used to choose whether the node is
trustworthy or not as a recommender.
2) Reluctant node of type 2: In this type of model, the Michiardi and Molva proposed CORE (Collaborative
reluctant node does not forward packets which are destined to Reputation mechanism) [3]. It defers from Watchdog and
other nodes[10]. These reluctant nodes participate correctly CONFIDANT by bringing aging factor with more weight on
in the routing function by advertising available roots. When past observations and isolating nodes having bad reputation.
this behaviour is chosen, reluctant nodes do not cooperate in Three types of reputations are used in the CORE.
packet forwarding function. However, a reluctant node that Subjective reputation of a target node is the reputation
operates complying with this model, participates in the route calculated directly from a subject’s inspection of the target
discovery and route maintenance phase of the DSR protocol. node’s behavior. Indirect reputation is evaluated only
considering the direct communication between a subject and
The outcome of this type of behaviour is that the reluctant
its neighbours. Function reputation is the subjective and
node will save a major amount of energy (battery life) by
indirect reputation calculated with respect to different
dropping great no of data packets.
functions such as forwarding a data packet, reply route request.
III. RELATED WORK The final reputation information is combined from the three
reputations with different weight associated to the functional
A lot of algorithm has been proposed claiming fairness in reputation value. CORE consists of two basic components:
routing.L Wang, Y Shu, M Dong, L Zhang proposed a Reputation Table (RT) is a data structure stored in each
multipath source routing scheme [9]. This scheme makes use network entity, keeping the reputation data pertaining to the
of DSR’s route discovery mechanism whereby multiple routes nodes in the network. The Watchdog mechanism (WD) is used
can be returned. The more alternative routes available in route to detect misbehaving nodes. With CORE only positive rating
cache, the more possibility that a node can find a perfect route. factors are distributed among the entities to avoid a
To get multiple route dsr optional feature is enabled. All the misbehaving entity to distribute false information about other
routes discovered are stored in the route cache with a unique entities in order to initiate a denial of service (DoS) attack.
route index. Route index is used in selecting multiple paths Miranda and Rodrigues classified nodes into friends, foes
from the route cache. and reluctant but at the cost of memory and message overhead
S. Marti, T. J. Giuli, K. Lai and M. Baker proposed two [4]. W. J. Adams, G. C. Hadjichristofi and N. J. Davis used
techniques that improve throughput of a network in the Reputation Indexing Windows (RIW) to handle misbehaving
presence of reluctant nodes [1]. The watchdog method is used nodes [5]. Unlike CORE it emphasised on current feedback
for each node to detect misbehaving nodes in the network. items rather than old ones.
When a node sends a packet to next hop, it tries to overhear S. Zhong, J. Chen and R. Yang proposed Sprite [13], a
the packet forwarded by next hop. If it hears that the packet is credit-based system for MANET. As opposed to Nuglets or
forwarded by next hop and the packet matches the previous Counter they do not require tamper-proof hardware to prevent
packet that it has sent itself, it considers the next hop node the fabrication of payment units.
behaves well. Otherwise it considers the next hop node is
misbehaving. The pathrater uses the knowledge about IV. DYNAMIC SOURCE ROUTING
misbehaving nodes acquired from watchdog to pick the route
that is most likely to be reliable. Each node maintains a trust A. Overview
rating for every other node. When watchdog detects a node is The DSR is an entirely on demand routing protocol which
misbehaving, the trust rating of the node is updated in negative composed of two parts: route discovery and route maintenance
way. When a node wants to choose a safe route to send [6].
packets, pathrater calculates a path metric by averaging the
node ratings in the path. B. Route Discovery
S. Marti, T. J. Giuli, K. Lai and M. Baker implemented the In DSR, whenever a node needs to send a packet to some
solutions on DSR protocol using ns2 as simulation destination for which, it does not currently have a route in its
environment. The simulation result shows the throughput of route cache, it initiates Route Discovery to find a route. The
the network is increased by up to 27% in a network where initiator broadcasts a Route Request packet to its neighbours,
packet drop attack happens. However, routing overhead is also specifying the target and a unique identifier from the initiator.
increased up to 24%. Each node receives the route request. If it has recently seen
A lot of work has been proposed on sensor network using this request identifier from the initiator, discards the request.
reputation to bring fairness in routing [11]. S. Buchegger and Otherwise it appends its own node address to a list in the
J.Y. LeBoudec proposed CONFIDANT algorithm using both request and rebroadcasts the request. When the route request
direct and indirect observations to measure fairness in reaches its target node, the target sends a route reply back to
Dynamic Ad -hoc Networks [2]. CONFIDANT makes a the initiator of the request, including a copy of accumulated
distinction trust from reputation. For each node, reputation list of address from the request. When the reply reaches the
rating signifies how well a node behaves while trust rating initiator of the request, it caches the new route in the route
stands for how honest a node is. Reputation rating is used to
152 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 4, 2011
cache. The intermediate node also sends a route reply, if it has A. Multipath dynamic source routing
a route to the destination.
B. Efficient Packet forwarding and load balancing
C. Route maintenance (Reputation based) Multipath Dynamic Source Routing
It is a mechanism by which a node sending a packet along
a specified route to some destination detects if that route has Multipath routing makes use of DSR’s route discovery
broken. If after a limited no of local retransmission of the mechanism whereby multiple routes can be returned. With
packet a node in the route is unable to make this confirmation, standard DSR, a node selects a shortest route to the destination
it returns a route error to the original source of the packet, from the route cache. In this mechanism, it works as the more
identifying the link from itself to the next node as broken. alternative routes available in route cache, the more possibility
that a node can find a perfect route. To get multiple route DSR
optional features is enabled. All the routes discovered are
E stored in the route cache with a unique route index. Route
G I index is used in selecting multiple paths from the route cache.
A
C F Efficient Packet forwarding and load balancing
J
B H Since this mechanism makes use of source routing
D intermediate nodes forwards the packet as the route indicated
in packet’s header. Reputation based packet forwarding to
handle reluctant behaviour. The mechanism counts on the
(a)
principle that a node freely evaluates its neighbours based on
the completion of requested service. The core principle is that
E when a node forwards packet through one of its neighbours, it
G I holds the neighbour responsible for the correct delivery of the
A packet to destination. Fig. 2 illustrates the operation of
C F
J mechanism.
B H
D C D
E
B
(b) F G H
E A
G I
I J K
A
C F
J
B H Figure 2.Sensor network with reluctant node G.
D
(c) Consider two flows A-F-G-H-E, B-F-G-H-K, in fig.
Figure 1.Example of Route Request: Node C is source node while Node J 1 Suppose node G starts acting reluctantly, dropping all
is destination. (a) Node C broadcasts a route request packet to find a path to
node J. (b) Neighbour nodes broadcasts route request in the same manner but the packets that it is likely to forward.
adding its own address to the packet. In the same manner route request is 2 Once, neighbour nodes of G will notice that packets
forwarded throughout the network. (c) J sends back a route reply to C using are not being delivered to their intended destinations.
the path contained in one of the route request packet.
C D
V. REPUTATION BASED MULTIPATH DYNAMIC
E
SOURCE ROUTING
B
F G H
In this work we try to find a hybrid approach that reduces
A
packet drops caused by identification and isolation of reluctant
nodes using algorithm proposed by [9] with the help of I J K
multipath source routing. Reputation based multipath dynamic
mechanism has two elements –
153 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 4, 2011
C D Ci is the remaining battery capacity of node. As the battery
E
capacity decreases, the value of cost function for node n will
increase. Therefore to find a route with maximum remaining
B battery capacity we should select a route π that has minimum
F G H battery cost
C π = min{Ci | i Є A} where A is the set containing all
possible routes. Since battery capacity is directly incorporated
A into the routing protocol, this metric prevent hosts from being
I J K overused, thereby increasing their lifetime and the time until
the network is partitioned
VI. ALGORITHM
Figure 3.Reluctant node G dropping packets forwarded from node F 1. During route discovery a source node sends a RREQ packet
to its neighbouring nodes. RREQ message must contain
source address, destination address, hop count, sequence no.
3 Node F will reduce the reputation index it assigns to G
twice due to G’s effect on both flows. 2. The intermediate nodes after appending their addresses
4 Immediate neighbours of node F and H will in turn again broadcast RREQ to the hop-2 neighbours. This
reduce the reputation index; they give to node F and H. process continues until RREQ get to the destination node.
5 Once G’s reputation index goes below a certain
3. Destination node sends a route reply back to the source node,
threshold, a new path is selected from the route cache.
including a copy of accumulated list of address from the
RREQ.
In a network of nodes N = {1,2,...,n}, each node i maintains a
reputation table where a reputation index Rij is stored for each 4. When the reply reaches the initiator of the request, it caches
of the node’s immediate neighbours j. A node credits a the new route in the route cache. Here multiple paths are
reputation index to each of its neighbours based on successful returned to source node. All the routes discovered are stored
delivery of packets forwarded through that neighbour. On the in the route cache with a unique route index. Route index is
other hand delivery failures penalize such neighbours by used in selecting multiple paths from the route cache.
decreasing the reputation index. Success or Failure is decided During data transfer, when a node receives a data packet, it
based on feedback obtained from the destination. checks if received packet is a retransmission. If data packet
In this mechanism, each node maintains a lookup table to is a retransmission then node decrements the reputation
keep information about data packets forwarded through it. It index of the neighbour through which the original packet
stores source and destination IP, port number, sequence was forwarded.
number and the next hop address. Each node also maintains a
5. If Rik be the reputation index of previous hop k and
reputation table in which, a node stores two thing neighbour
rthrash be the reputation threshold then -
node id and its reputation. Following algorithm illustrates the
operation of mechanism. 6. if (Rik < Rthrash )
Load balancing 7. drop the packet
8. else
In order to perform load balancing this paper uses Minimum
Battery Cost Routing [7] in MP-DSR where accumulated path 9. store related information and forward the packet
cost is used. If length of path P is L then its accumulated path 10. If a node receives acknowledgement packet, it
cost is: increments the reputation of its neighbour node.
(1) 11. Once a node’s reputation index goes below a certain
threshold, a new path is selected from the route cache using
Where, C {Di, Di+1} implies that the link cost in path P from MBCR (minimum battery cost routing.) and data is sent
Di to Di+1 for linkl . through new path.
To consume the energy in a more balance manner, an intuitive
technique is to utilize cost function based on the node's VII. SIMULATION SETUP AND EVALUATION
remaining battery capacity. In Minimum Battery Cost Routing Using ns-2[14] we study the performance of our mechanism.
(MBCR) [8] the battery cost C for a particular route π with n In our simulation we placed 20% nodes randomly in 670×670
nodes is defined as: area. Each node has a transmission range of 250m and moves
at a speed of 10m/s. We randomly generate CBR (Constant
(2)
154 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 4, 2011
Bit Rate) TCP flows. Neighbours identified as reluctant (i.e.
whose reputation index is below threshold). All route requests
(RREQs) and route replies forwarded by such neighbours will
be ignored. Simulation parameters are shown in table I.
Table I. SIMULATION PARAMETER
Parameter Value
Number of nodes 20
Routing scheme DSR
Packet size 512 bytes
Traffic model CBR
Propagation Two Ray Ground Figure 4.Total Number of packet drop in DSR vs. Number of packet
dropped in reputation based multipath DSR
Max speed of nodes 15 ms
As shown in the above graph x axis indicates Simulation time
Initial energy of node 1000 joules and y axis indicates number of packets received. Red line
Simulation Time 200 sec indicates number of packet received vs. simulation time for
original method. Blue line indicates number of packet received
vs. simulation time. From the above graph it is observed that
number of packet recieved in original method is notably less.
A. Simulation result In above graph initially for 10 seconds both lines are touching
The impact of reluctant nodes was measured in terms of total x axis because in the network traffic flow starts at 10 second.
number of packet dropped, number of packet received, packet
delivery ratio and network throughput. The measurements of
the network were made using a AWK script that parses and
analyse the trace file output provided by ns-2 software. The
trace file gives information about a ser of defined events
which occurred in the simulation such as routing events and
agent event. Through agent event trace not only it is possible
to estimate the total no packet sent by every node of the
MANET but also the total number of packets which have been
dropped.
Aggregate Network throughput (AGT) is the ratio of total
number of bits received to the total number of bits sent
AGT = Total number of received bits / Total number of sent
bits
Packet delivery ratio (PDR) is the ratio of total number of
packets sent the total number of packet received. Figure 5.Total Number of packet received in original DSR vs. Number
of packet received in reputation based multipath DSR.
PDR = Total number of sent packets / Total number of
Table II. AGGREGATE NETWORK THROUGPUT
received
Parameter % reluctant nodes DSR RB-
MPDSR
Aggregate 25 33.04 43.06
network
throughput
155 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 4, 2011
VIII. CONCLUSION AND FUTURE WORK
In wireless sensor network, due to limited energy of nodes,
some nodes may do not participate in routing functions. This Dr. Jyoti Singhai is Associate professor in Maulana
paper deals with detection and isolation of such nodes. The Azad National Institute of Technology (MANIT),
proposed method has certain limitations such as in proposed Bhopal, India. She holds Ph..D degree from MANIT
method cooperative nodes may be accused of reluctant India. Her general research interests include wireless
communication, image processing, WSN and network
behaviour on account of packet drop due to collision, memory security.
overflow, energy. Currently we are working to mitigate these
limitations.
REFERENCES
[1] S. Marti, T. J. Giuli, K. Lai and M. Baker “Mitigating routing
misbehavior in mobile sensor networks,” Proceedings of Sixth Annual
IEEE/ACM Intl. Conference on Mobile Computing and Networking ,
April 2000,PP. 225-256.
[2] S. Buchegger and J.Y. LeBoudec, “Performance analysis of
CONFIDENT protocol: cooperation of nodes – fairness in dynamic
sensornetworks,” Proceedings of the ACM MobiHoc, June 2002.
[3] P. Michiardi and R. Molva, “CORE: A collaborative reputation
mechanism to enforce node cooperation in mobile sensor networks,”
Proceedings of the 6th Joint Working Conference on Communication
and Multimedia Security , September 2002, pp. 107-121.
[4] H. Miranda and L. Rodrigues, “Friend and Foes: Preventing
Reluctantness in open Mobile Sensor Networks,” Proceedings of Intl.
conf. Distributed Computing System Workshop 2003, pp. 440-445.
[5] W. J. Adams, G. C. Hadjichristofi and N. J. Davis, “Calculating a
Node’s Reputation in a Mobile Adhoc Network,” In Proceedings of
IEEE Intl. Performance Computing and Communications Conference
(IPCCC)2005, pp. 303-307.
[6] D. Johnson , D. A. Maltz, and J. Broch. The Dynamic Source Routing
Protocol for Mobile Sensor Networs (Internet-Draft). Mobile Ad-hoc
Network (MANET) Working Group, IETF, October 1999.
[7] C. K. Toh, "Maximum Battery Life Routing", IEEE Communication
Magazine, June, 2001.
[8] M. Tamer Rafaei, Vivek Srivastav, Luiz DaSilva, ”A Reputation-based
Mechanism for Isolating Reluctant Nodes in Adhoc
Networks,”Proceedings of the Second Annual Internatinal Conference
on Mobileand Ubiquitous Systems: Networking and
Services(MobiQuitous’05).
[9] L Wang, Y Shu, M Dong, L Zhang, OWW Yang ”Multipath Source
Routing in Wireless Sensornetwork ,“ Proceedings of IEEE ICC'01,
2001.
[10] P. Michiardi and R. Molva, “Simulation based analysis of security
exposures in mobile sensornetworks,” Proceedings of the European
Wireless Conference, February 2002.
[11] K.Mandalas, D.Flitzenis, G.F Marias and P Georgiadis, “A Survey of
several cooperation schemes for MANETs,” Proceedings of
International Symposiam on Signal processing and Information
Technology, 2005.
[12] Eliabeth M.Royer, Santa Barbara Chai-Keong Toh “ A Review of
Current Routing Protocols For SensorMobile Wireless
Networks,”Proceedings of IEEE Personael Communications, April 1999.
[13] S. Zhong, J. Chen and R. Yang, "Sprite: A Simple, Cheat-proof,
Creditbased System for Mobile Ad-hoc Networks", in Proc. IEEE
INFOCOM2003, Apr. ’03.
[14] “The ns Manual (formerly ns Notes and Documentation)”, The VINT
Project Collaboration between researchers at UC Berkeley, LBL,
USC/ISI, and Xerox PARC, January 6 –
2010.http://www.isi.edu/nsnam/ns/doc/ns_doc.pdf.
AUTHORS PROFILE
Mr. Praveen Kaushik is Assistant Professor in Maulana
Azad National Institute of Technology (MANIT) Bhopal,
India. He is pursing PhD in WSN. His general research
interests include WSNs, Ad hoc Network and Wireless
Communication.
156 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
Related docs
Other docs by ijcsiseditor
Digital Images Encryption in Spatial Domain Based on Singular Value Decomposition and Cellular Automata
Views: 0 | Downloads: 0
Agent Behavior in Multiagent Systems: Issues and Challenges in Design, Development and Implementation
Views: 1 | Downloads: 0
Optimizing Cost, Delay, Packet Loss and Network Load in AODV Routing Protocols
Views: 2 | Downloads: 0
