A survey on trust based routing in manet

INTERNATIONAL JOURNAL OF COMPUTER ENGINEERING &
International Journal of Computer Engineering and Technology (IJCET), ISSN 0976 – 6367(Print),
ISSN 0976 – 6375(Online) Volume 3, Issue 3, October-December (2012), © IAEME
TECHNOLOGY (IJCET)
ISSN 0976 – 6367(Print)
ISSN 0976 – 6375(Online)
Volume 3, Issue 3, October - December (2012), pp. 213-222
IJCET
© IAEME: www.iaeme.com/ijcet.asp
Journal Impact Factor (2012): 3.9580 (Calculated by GISI) ©IAEME
www.jifactor.com

A SURVEY ON TRUST BASED ROUTING IN MANET

S.Sridhar P.Chandrasekar
Dept. of Computer Applicatons, Dept. of Computer Applications,
S.A.Engineering College, S.A.Engineering College,
Thiruverkadu P.O., Chennai-77, Thiruverkadu P.O., Chennai-77,
ssridharmca@yahoo.co.in mail2chandruu@yahoo.co.in

ABSTRACT

A mobile ad-hoc network (MANET) is a peer-to-peer wireless network where nodes can
communicate with each other without the use of infrastructure such as access points or base stations.
Nodes can join and leave the network at anytime and are free to move randomly and organize
themselves arbitrarily. In MANETs, each node should not only work for itself, but should be
cooperative with other nodes. Working in such environment, MANETs are vulnerable to attacks by
malicious and misbehaving nodes that try to compromise the routing protocol functionality. Neighbor
based communication without any trust worthiness creates a major vulnerability in security related
aspects of this network. In this type of situations, trust value plays a crucial role in all of the network
activities. Continuous evaluation of node’s performance and collection of neighbor node’s opinion
value about the node are used to calculate the trust relationship of this node with other nodes. Hence a
mechanism to formalize trustworthiness among these anonymous nodes is needed to make sure
resources are shared among trusted nodes only. Managing trust in a Mobile Ad Hoc Network is
challenging when collaboration or cooperation is critical to achieving mission and system goals such
as reliability, availability and scalability. This article extensively studies the various trust management
approaches and presents an analyses of various trust based protocols which have provided reliability
in MANET routing and increases Qos metrics.

Keywords
MANET, Trust, Misbehaving node, Malicious node, QoS metrics.
1. INTRODUCTION

Mobile ad-hoc networks [1], are dynamically configured, multi-hop wireless networks characterized
by absence of any infrastructure, dynamic topology and wireless links. MANET composed only of
nodes and these nodes do not have fixed infrastructure or any centralized controller such as access
point or server to determine the route of the paths. Thus, each node in an ad hoc network has to rely
on each other in order to forward packets and there is a need to use a specific cooperation mechanism
to forward packet from hop to hop before it reaches a required destination by using routing protocol.

213

International Journal of Computer Engineering and Technology (IJCET), ISSN 0976 –
6367(Print), ISSN 0976 – 6375(Online) Volume 3, Issue 3, October-December (2012), © IAEME

As nodes may not aware to which nodes it is connected with or which nodes connected to them,
therefore access to resources or information can be shared among both trusted and non trusted nodes,
unfavorable situation which makes private transactions impossible. The inherent freedom in self-
organized mobile ad-hoc networks introduces challenges for trust management; particularly when
nodes do not have any prior knowledge of each other. Hence, to assure that access to resources is
given only to trusted nodes, the trustworthiness among anonymous nodes needs to be formalized.
The concept of trust originally derives from social sciences and is defined as the degree of subjective
belief about the behaviors of a particular entity [2]. Battle field, emergency and disaster environments
require immediate network .formation and trusted route establishment for their communication.
MANET is the suitable network for such type of application areas. A trusted system[3] is defined as
an entity whose security mechanisms are isolated from and are uncircumventable by unauthorized
users; the system can be identified, content controlled and secure, and managed by a competent
authority. All the existing MANET protocols simply trust their neighbors and make a route through
the neighbors. This kind of neighbor based routing is disturbed by intruders and internal attackers or
malicious nodes. In spite of malicious nodes presence in the network, the network must provide its
services without any problem. This is achieved only by the trust based protocols. Trust establishment
and management between entities (nodes or agents) can be done through a central trusted authority or
in a distributed fashion by nodes [4], by a combination of both. Trust is a passive entity in the
network.
The term ”Trust Management” was introduced by [5] and identified it as a separate component of
security services in networks and clarified that ”Trust management provides a unified approach for
specifying and interpreting security policies, credentials, and relationships.” According to [6, 7], there
are four major properties of Trust and they are, Context Dependence:- The trust relationships are only
meaningful in the specific contexts. Function of Uncertainty:- Trust is an evaluation of probability of
if an entity will perform the action. Quantitative Values:- Trust can be represented by numeric either
continuous or discrete values. Asymmetric Relationship:- Trust is the opinion of one entity for another
entity. That is, if A trusts B, it is unnecessary to hold that B trusts A.
Although many trust management schemes have been proposed to evaluate trust values, no work
clearly addresses what should be measured to evaluate network trust. Most trust-based protocols for
secure routing calculated trust values based on the characteristics of nodes behaving properly at the
network layer. Trust measurement can be application dependent and will be different based on the
design goals of proposed schemes [8]. The metrics include overhead (e.g., control packet overheads),
throughput, good put, packet dropping rate, and delay. ”Route usage” refers to the number of routes
selected particularly when the purpose is for secure routing. ”Trust level” is a recently used system
metric. Example metrics using the trust level include confidence level of the trust value,
trustworthiness, opinion values about other nodes, and trust level per session. In the following section
we will discuss about the various trusts based schemes which have deployed routing in mobile ad hoc
network based on some form of trust.

2. ANALYSES OF TRUST MANAGEMENT IN MANETS
2.1 Forming Trust Using Direct and Indirect Approach [9]
A trust schema for MANET is build to allow two nodes to use trust value computed to determine the
access control action. The value computed based on formation of trust will either block or allow
access to be shared between these nodes. The following definitions to be used In the schema.
Definition 1: A node that services/information to be shared is known as Servicing Node (SN).
Definition 2: A node that request for information/services sharing is known as Request Node (RN).
Definition 3: A peer node (PN) is node which is peer to servicing node, which gives recommendation
about Request Node (RN).
In this model, 0 represent complete distrust and 1 represent completely trust. An assumption is also
made where authentication between nodes is already handled by some certification or credential
passing among nodes when nodes start to collaborate. As such the initial trust value 0.5 is given to RN
once both collaborate. An initial trust value of 0.5 is assumed. After the initial collaboration, during

214


the interaction, the new trust value is calculated to determine the access control action. In calculating
the trust value, The initial trust value plus or minus is combined with the trust value calculated during
the formation of trust .The value can be increased to >0.5 or be decreased which is <0.5 based on
input gather during interaction or nodes behavior.
2.1.1 Direct Approach.
There are two cases where direct approach can be used. Case I: RN request to access from SN and SN
already know RN. In this case, SN has the information about RN based on past knowledge via
interaction. However since the new request for resource sharing could be different from the previous
or it is a totally new item thus, there is a need to calculate new trust value between these nodes. Each
time nodes collaborate, the initial trust value is obtained (0.5), and therefore through observation of
SN do on RN, the trust value can be increased (or decreased) based on RN behavior. Both facts are
supported by context specific data such as place/situation and also time. Case II: RN is new to SN and
no recommendation about RN from PN. For the second case, when two nodes never met each other
and no recommendation can be gather from peer nodes, therefore SN needs to evaluate RN based on
observation only and SN will have to depend solely on RN’s behavior to calculate the trust value. The
context specific data together with time is used to support the evaluation.
2.1.2 Indirect Approach
As for indirect approach, an assumption that there is an existence of peer node (PN) when RN is
request for resource sharing from SN is made and also that SN has never interacted with RN before is
assumed. Hence the trust value calculated will be based on SN’s own observation ((O)SN(c,t)) with
peer recommendation plus/minus the initial trust value . The context data and time will be used as in
the previous formula. Evidences are collected via past knowledge of PN have on RN, reputation of
RN and SN own observation. Reputation used as it can strengthen the recommendation made by PN.
Reputation which can be defined as ‘some idea or report of its tendency to fulfill the trust placed in it
during a particular condition; its reputation is created through feedback from individuals who have
formerly interacted with the entity ‘[11]. For example, if RN is a reputable person i.e. well known
tailor who make dresses for many famous people, therefore in this case the tailoring context will be
very good and most importantly can be trusted.
2.1.3 Trust Mapping
The mapping between the trust value and access control action for resource and information sharing is
as follows. The initial trust value 0.5 for collaboration is given to RN each time the nodes collaborate.
If the value is less than 0.5 than no access (block) is given otherwise the value will increase gradually
if SN belief that RN is behaving well. To allow flexibility to SN, SN is allowed decide how much
resource can be share with RN during the interaction.
Thus a trust schema to form trust among anonymous nodes in the network is framed. The schema is
constructed via direct and indirect approach. The trust schema that is build is used to allow resource to
be shared among trusted nodes. The trust value calculated is then mapping with the access control
privileges to allow resource/information sharing in MANET.
2.2 Hybrid Trust Management Framework (HTMF) [10]
Here the focus is on trust management framework, which is intended to cope with misbehavior
problem of nodes and stimulate nodes to cooperate with each other. However, there are still many
intrinsic problems with the existing trust management frameworks, which remain unsolved, such as
frangibility under possible attacks. To design a robust trust management framework these intrinsic
problems are investigated with the existing frameworks, and propose a hybrid trust management
framework (HTMF) to construct trust environment for MANETs. The proposed HTMF is more robust
and reliable than the existing frameworks which face problems like Selective Misbehavior Attack and
Location-dependent Attack
Step S1: Update ITF through Direct Information. Each node in the network monitors the behavior of
its neighbors using watchdog mechanism. Step S2: Distribute and process second-hand information.
Here, the nodes receiving these information perform deviation test and another check to inhibit bad
mouthing attack and conflicting behavior attack. Step S3: Evaluate trust value and confidence value.
In this step, these two values are formed based on ITF obtained through steps S1 and S2. Step S4:

215


Evaluate trustworthiness. Since two parameters are difficult for comparison two parameters formed in
S3, namely trust value and confidence value, are combined into a whole trust metric, trustworthiness.
This HTMF holds objective characteristic by which trust for a node is evaluated based on not only
direct observations but second-hand information. It makes HTMF robust under selective misbehavior
attack and location-dependent attack in contrast to the trust establishment framework.

2.3 AMLET: Adaptive Multi Level Trust Framework [11]
Adaptive Multi Level Trust model called AMLeT is proposed using two complementary trust levels.
AMLeT calculates trust in one level of hard or soft and updates calculated trust considering network
field, modifications of context and situations with the passing of the time. AMLeT is extensible for
different fields of network, applications and situations. AMLeT based AODV routing was improved.
and the results indicated overall improvement in the network performance without imposing
particular overhead time in network operations.
2.3.1 Trust Criteria
Three new criteria introduced to improve more accurate trust model for MANETs. The proposed
criteria can lead to develop an adaptive trust model. The criteria are introduced in the following:
Process time: The first one is elapsed time to evaluate and deploy trust value to target node and
transmit the value to the others. It is obvious that this factor depends on the amount of process and
being early or late (proactive or reactive). Optimism or pessimism: This factor firmly depends on
security level of the application. The context and situation of the application have an important effect
on the default trust value of an entity. Thus, the initial trust values to others are supposed to be low.
Trust slope: This criterion depends on two factors. First, the elapsed time in which the trust reaches to
highest value. That refers to the manner of the trust function to increase the trust value observing the
positive evidences. Second, the required subsequence of negative behaviors that cause the trust value
becomes lowest level. That means the manner of the trust function to decrease the trust value
observing negative evidences.
2.3.2 Hard trust and Soft trust
Using both the criteria, defined by the others and the criteria proposed above, two levels of trust
which take into account these criteria differently are stated below. Hard trust is a level of trust that the
manner of its trust function is slow in rising and sharp in falling. In Hard trust, trust function increases
trust value gradually, but decreases it keenly. This level of trust is well suited to the network
applications which require operating in a high security mode. The transmitting data in this networks is
worthy to insider and outsider entities. So there are many incentives to express malicious behaviors.
Soft trust is a level of trust that the manner of trust function is sharp in rising and slow in falling. Soft
trust is appropriate for network applications with stable conditions. The proportion of needed security
to the amount of the transmitting data is low. Therefore, there are not many incentives to express
malicious behaviors in these networks. Continuity of life in these network applications is more
important than security. It means that network will continue to operate correctly despite of light injury
in some data and negative behaviors of some nodes.
Hardness parameter is a numeric parameter that indicates required security level of a MANET
application in an instant of network life. This parameter is computed regarding type of MANET
application, optimism or pessimism, and trust slope criteria. Considering hardness parameter in every
interaction, trustworthiness will be estimated using the computed Hard trust and Soft trust. Therefore
in this trust model, the required level of security for the application and dynamic situations are
considered to evaluate trustworthiness in each interaction between nodes. In Figure 3, a schematic
view of AMLeT framework is presented. This framework consists of six operating units to evaluate
trust adaptively to situation and type of the network, using two defined level of trust
2.3.3 An AMLeT based routing
AODV is one of the most important reactive routing protocols in MANETs. This improvement is
done by integrating AMLeT modules into AODV agent in order to boost some security and efficiency
factors. Threshold value for trustworthiness is 0.5 which determines the border of trustworthiness and
untrustworthiness. After packet forwarding, the AODV agents evaluate the behavior of their adjacent

216


nodes. Hardness parameter is determined base on N recent interactions. Finally, these values are
passed to mixture trust function to calculate the trustworthiness of the target node. Trustworthiness of
each node is used in routing process to punish distrusted nodes and to improve routing efficiency by
isolating established routs from them and having more trusted routes. Thus two complementary levels
of trust have been introduced for mobile ad-hoc networks naming Hard trust and Soft trust. These
levels can be foundation to develop trust models for each application. Each of defined trust levels is
desirable and usable for the applications with same security requirements. In fact, AMLeT adapts
itself to context and situations changes of network. Functionality and feasibility of AMLeT were
proved by AMLeT based AODV routing.

2.3.4 An AMLeT based routing
AODV is one of the most important reactive routing protocols in MANETs. This improvement is
done by integrating AMLeT modules into AODV agent in order to boost some security and efficiency
factors. Threshold value for trustworthiness is 0.5 which determines the border of trustworthiness and
untrustworthiness. After packet forwarding, the AODV agents evaluate the behavior of their adjacent
nodes. Hardness parameter is determined base on N recent interactions. Finally, these values are
passed to mixture trust function to calculate the trustworthiness of the target node. Trustworthiness of
each node is used in routing process to punish distrusted nodes and to improve routing efficiency by
isolating established routs from them and having more trusted routes. Thus two complementary levels
of trust have been introduced for mobile ad-hoc networks naming Hard trust and Soft trust. These
levels can be foundation to develop trust models for each application. Each of defined trust levels is
desirable and usable for the applications with same security requirements. In fact, AMLeT adapts
itself to context and situations changes of network. Functionality and feasibility of AMLeT were
proved by AMLeT based AODV routing.

2.4 Trust Based Security Approach Using Trust Counter [12]
A trust based packet forwarding scheme is designed for detecting and isolating the malicious nodes
using the routing layer information. It uses trust values to favor packet forwarding by maintaining a
trust counter for each node. A node will be punished or rewarded by decreasing or increasing the trust
counter. If the trust counter value falls below a trust threshold, the corresponding intermediate node is
marked as malicious.
In this approach, by dynamically calculating the nodes trust counter values, the source node can be
able to select the more trusted routes rather than selecting the shorter routes. Changes are done to the
AODV routing protocol. An additional data structure called Neighbors' Trust Counter Table (NI T) is
maintained by each network node. The routing process can be summarized into the following steps: 1)
Discovery of routes: it is just like the route discovery in DSR. Suppose A starts this process to
communicate with D. At the end, A collects all the available routes to D; 2) Validation of routes:
Node A check the trust values of the intermediate nodes along the path. Assuming node B's trust value
is missing in A 's trust table or its trust values is below a certain threshold, put B into a set X. 3)
During the transmission, node A updates its trust table based on the observations. When some
malicious behavior is found, A will discard this path and find another candidate path or restart a new
discovery. 4) Compute trust values for every node in X based on the trust graph. 5) Among all paths,
A chooses the one with the max ( in= 1 pi) where n is the number of nodes along with path.
A trust based security protocol which attains confidentiality and authentication of packets in both
routing and link layers of MANETs is developed. It uses trust values to favor packet forwarding by
maintaining a trust counter for each node. The perfect security solution is hard to reach. But the
average security level (for a node) can be achieved as expectation based on accumulated knowledge
and as well as the trust relationship built and adjusted.
2.5 Trust Based Secure Routing Using NTC And RTC [13]
Continuous evaluation of node’s performance and collection of neighbor node’s opinion value about
the node are used to calculate the trust relationship of this node between source and estimations

217


without any intruders or malicious nodes can be established with the proposed trust based routing
protocol that equally concentrates both in node trust and route trust.
2.5.1 Node Trust Calculation Process
Each node has opinion about other node’s (neighbor) trustworthiness. Node X has an opinion about
trustworthiness of one of its neighbor node Y based on Y’s previous and current behaviors. All the
nodes in such environment already maintain Routing Table. Additionally added Neighbor Table
should be maintaining in all the nodes for keep tracks the dynamically changing neighbor list and its
corresponding node trust value.Node trust is calculated by the collective opinion of node’s neighbors.
The resultant trust value is placed in Trust Value field of Neighbor Table. The node trust is computed
based upon the information that one node can collect about the other nodes.
2.5.2 Route Trust Calculation Process
Route trust is computed by every node for each route in its routing table. Existing Routing Table
extended with one more field; Route Trust. In this approach, source node selects the route which is
having the highest Route Trust value. Route Trust field of every Routing Table entry is updated at
some regular interval. In this method, only one additional field is enough to monitor the route trust
worthiness. Destination node in each entry originates R_ACK message packet to node which one is
maintaining this routing Table. R_ACK is the modified version of RREP_ACK message packet. Each
route already having the entry for number of packets sent to the corresponding destination. Usually
destination node originates R_ACK message packets. This packet moves backward direction from the
destination. Node which one is receiving R_ACK packet uses the entry Number of Packets Received
for route trust calculation.
2.5.3 Route Establishment Process
Source initiates route establishment process by broadcasting RREQ message to all of its neighbors.
Each node maintains two main table; Route Table and Neighbor Table. Each node updates its
Neighbor Table by broadcasting HELLO packets in the regular interval. Neighbor Table consists of
two fields; Neighbor_ ID and Trust Value. Another table in every node is route Table. It maintains the
route detail information like Destination IP Address, Destination Sequence Number. Valid
Destination Sequence Number, Next Hop, Hop Count and Route Trust etc. for all the routes those are
valid from this node. At the time of route establishment process or packet forwarding process, this
table is updated. Neighbor nodes check this routing table whether they are having any route to the
desired destination or not. If it exists then nodes can sent a RREP message to source in the backward
path. Source waits for more than one RREP from its neighbors. In this method, Route Trust value of
the RREP message plays a major role in RREP selection process. Based on the highest Route Trust
level, the RREP is selected for further route establishment.
Proposed modifications are in acceptable limit. With this minimum overhead, the malicious nodes are
eliminated as well as a best trusted route between source and Destination is established and also it
creates a secure communication in this environment without any internal attackers.

2.6 TBAODV - Trust Based AODV [14]
The performance of Ad-hoc On Demand Vector protocol is modified by including the source route
accumulation feature. As low transmission power of each ad-hoc node limits its communication
range, the nodes must assist and trust each other in forwarding packets from one node to another.
However, this implied trust relationship can be threatened by malicious nodes that may fabricate,
modify or disrupt the orderly exchange of packets. Security demands that all packets be authenticated
before being used. A particularly hard problem is to provide efficient broadcast authentication, which
is important mechanism for MANET. Here, a routing algorithm is proposed which adds a field in
request packet which stores trust value indicating node trust on neighbor. Based on level of trust
factor, the routing information will be transmitted depending upon highest trust value among all. This
not only saves the node’s power by avoiding unnecessary transmitting control information but also in
terms of bandwidth (channel utilization), which is very important in case of MANET. Here, trusted
path is used irrespective of shortest or longest path which can be used for communication in the
network. Route trust value is calculated for the complete reply path which can be utilized by source
node for next forthcoming communication in the network.

218


Algorithm for different functions used in packet transmission and reception routine as follows: Step
1:-Initially trust value 100 is assigned to all nodes in the network using assign trust ( ) function. Step
2:- Trust value are printed using Printtrust ( ) function. Step 3:- Source node broadcast request to all
its neighbouring node using Send_Request( ) function.In this function hop count is initialized.
Schedular class is invoked to run the simulation. Step 4:- Neighbouring node receive the request then
it will check whether it is destination or not.If it is Destination then it will Send_Reply( ) function
otherwise forward request to its neighbouring node. This will check in Receive_Request( ) function.
Step 5:- After confirming that it is not destination, it will further forward request to all its
neighbouring node using Forward_Request( ) function.Hopcount is increased at each node. Step 6:-If
it is destination then it will send reply using Send_Reply( ) function. Trust value 200 is assigned to all
nodes in the path from destination to source node. Now, Source becomes destination for the current
node. Step 7:-After receiving the reply then the decision will take whether the index node is
destination or not using Receive_Reply( ) function. If it is not destination then it will forward reply.
In TBAODV an extra field is created in the route request format. This trust value is updated on every
successful communication. The forthcoming communication is based on the route selection value
calculated for each RREQ path. This route selection value is used to select most trusted path rather
than selecting shortest or longest path. This significantly improves the trust factor on the neighboring
nodes in the network. Thus the trust based routing protocol proposed here improves the security level
and also prevent malicious node attack in the network.
2.7 Trusted And Shortest Path Selection Using R-AODV [15]
A security enhanced AODV routing protocol called R-AODV (Reliant Ad hoc On-demand Distance
Vector Routing) is defined. The implementation of this work is done by modifying a trust mechanism
known as direct and recommendations trust model and then incorporating it inside AODV which will
allow AODV to not just find the shortest path, but instead to find a short path that can be trusted. This
enhances security by ensuring that data does not go through malicious nodes that have been known to
misbehave. R-AODV does provide a more reliable data transfer compared to the normal AODV if
there are malicious nodes in the MANET.
2.7.1 Reliant On-Demand Distance Vector Routing Protocol(R-AODV)
AODV can be modified to select better path (best path (Bp)) during the route discovery cycle based
on the trust and number of hops (trusted and shortest). When the route request and route reply (R-
RREQ and R-RREP) messages in Reliant R-AODV are generated or forwarded by the nodes in the
network, each node appends its own trust to the trust accumulator (trust summation accumulator S[t])
on these route discovery messages. Each node also updates its routing table with all the information
contained in the control messages. As the R-RREQ messages are broadcast, each intermediate node
that does not have a route to the destination forwards the R-RREQ packet after appending its trust to
the trust accumulator in the packet. Hence, at any point, the R-RREQ packet contains a list of all the
nodes visited with their trust value added to trust summation accumulator S[t]. Whenever a node
receives a R-RREQ packet, it will check the updates of the route to the source node. It then checks for
better path (best path (Bp)) for intermediate nodes. The hop count included in the request message. A
new entry is made in the routing table for any of the intermediate nodes and assigns full trust to them,
if one did not already exist. If a route entry for a node does exist, and if best path (Bp) to any of the
intermediate nodes is greater than the previously known best path (Bp) to that node, the routing table
entry is updated for that node and assigns new trust value. New trust value will be updated in routing
table. The entry is updated by retaining the previously known sequence number for that node.
2.7.2 Route Discovery in R-AODV
The goal of this protocol is for source node to select the secure route with less hop count to a
destination node. The source node, S, broadcasts a route discovery message (R-RREQ) to its
neighbours which contains: S broadcasts R-RREQ <Source_Addr, Source_Seq#, Broadcast_ID,
Dest_Addr, Dest_Seq#, Hop_Count, S[t], Bp> As RREQ messages in AODV, for R-AODV, when a
node receives R-RREQ message, it sets up a reverse path back to the source by recording the
neighbour from which it received the R-RREQ. Meanwhile, when the node receives the R-RREQ, it
will check whether it is the destination or not, if so, it will updates the routing table for that node and
generate R-RREP. But if the receiver node was intermediate node, it attaches the trust value in its

219


routing table to the trust summation accumulator S[t] in the message. Upon receiving the message, a
node verifies the Best path in the routing table with the new best path value attached in the message, if
the new best path greater than the one in the routing table, the node then update the routing table
2.7.3 Route Reply in R_AODV
After receiving the R-RREQ, the destination node creates a route reply message (R-RREP), signs it
and unicasts the reply massage back to the source over the reverse path. The destination node, D,
creates the R-RREP, and sends it back to its neighbour. Route Reply message contains: D unicasts R-
RREP: <Source_Addr, Dest_Addr, Dest_Seq#, Hop_Count, Lifetime, S[t], Bp> With the inclusion of
trust mechanism.
It is expected that using R-AODV would result in a higher percentage of successful data delivery as
compared to AODV. However, it is also expected that due to the extra processing done and the
possibility that the packets may take a longer route, it is also expected that the normalized routing
load and end-to-end delay would increase. The use of R-AODV does provide a higher percentage of
successful data delivery. It has also shown that the impact to normalized routing load and end-to-end
delay is very minimal.

2.8 Reputation-Based Trust Model [16]
The main objective here is to highlight critical issues that impinge upon trust management and to
propose a reputation based approach for establishing trust that dynamically assesses the
trustworthiness of the participating nodes in the MANET environment.

2.8.1 Critical Issues in Trusted Systems
Initially certain security issues are discussed here. Identify a set attributes necessary for a mobile node
to support critical infrastructure related activities in a MANET. To be deemed as trustworthy, a node
must make a reasonable effort to perform its generic functions and duties in the network in a
dependable manner, broadly categorized under headings of Routing/Forwarding, Quality of Service,
and Security.

The routing and forwarding functions are governed by routing protocols. A robust routing protocol
reduces packet loss rates, eliminates the possibility of having frequent route failures, and is able to
cope quickly against topological changes. Quality of Service (QoS) is the ability of the underlying
network to provide preferential treatment to certain network traffic flows over others. In Security
Related Functions Trust on a system is enhanced if the system has security mechanisms to safeguard
its assets and actions. However, trust can also be established in a situation where “trustee will act in
the interests of a trustor without a guarantee” and in the absence of security. What this implies is that
if an entity is convinced or if enough assurance evidence exists that a target has acted in a consistent
and predictable manner over an extended period of time and its actions adhere to a moral code or a
standard as prescribed by a high level management policy then this target entity is perceived to be
trustworthy. Certainly the presence of security services in the system such as confidentiality, integrity
and access control and the ability to enforce them using a policy is likely to enhance trust on this
system.
The design principles of the trust model and then proposes trusted-platform architecture to support it
are discussed below. The process of computing trust is based on the reputation of mobile nodes and
independent of the underlying security, routing and QoS protocols, mechanisms, and algorithms being
used.
2.8.2 Design Principles of Reputation-Based Trust Model
The framework is built upon the following principles; Individualistic Model: This framework
provides a means for node to build its trust on a target independent of other nodes. It may also use the
recommendation of a trustworthy peer. Notion of reputation: The sense of trust builds upon the notion
of reputation. The reputation of a target node is determined on the basis of first hand information that
results through direct experience of interacting with that node, or through observation, or through
recommendation / references of peers. Trust is context dependent: This means that entities can be
trusted for only certain functions. In this model, trust is also a function of reputation that is
220


continuously changing with each new interaction. It is also time sensitive meaning an entity can only
be trusted for a certain period time. Trust is reflexive, asymmetrical, non-transitive: There is a
reflexive need to be able to protect one’s own resources. This model also perceives trust between two
nodes as an instance of asymmetrical reciprocity. Also the non transitive property of trust is
preserved. Trust depends on device capability: Each participating entity in a MANET must be aware
of the device capability (hardware, OS, trusted platform) of every other node or at least its
neighboring nodes.

2.8.3 Protocol Stack and Trusted Platform
The trust framework proposed is supported by a trusted architecture and a protocol stack. The trusted
platform (TP) acts as an enforcement point of various trustworthy services in the node. It provides a
tamper resistant hardware supported by a Trusted Platform Module (TPM) that provides some
primitive cryptographic functions such as random number generation, RSA key generation and has a
mechanism for protecting data by never releasing the root key outside the confines of TPM.

Thus a decentralized framework is presented for building a trust model for MANETs. The model of
trust is independent of underlying cryptographic schemes and also takes into account the hardware
constraints of the devices. In this scheme the notion of trust is bound to a function the node performs
in the network rather than to the node itself.

3. CONCLUSION

The study covers a hand full of works related to trust management in MANETS. The routing protocol
improved with enhancements and new strategies implemented to bring in reliability in MANET
routing. The objective of all works listed here focuses on providing trustworthy routing in MANET,
eliminating the misbehaving nodes and the nodes which bring down the performance of the network.
It also focuses to improve the performance of the network by increasing the Qos metrics. Based on the
observations the future work can be coined in such a way that the routing in MANET should be
implemented with a trust scheme which guarantees trustworthy routing, scheme to identify and isolate
misbehaving node and also to improve the QOS metrics.

4. REFERENCES

[1] Corson, S. and Macker, J. 1999 Mobile Ad Hoc Networking (MANET):Routing Protocol
Performance Issues and Evaluation Considerations. In RFC 2501.
[2] Cook, K. S. (editor), Trust in Society, vol. 2, Feb. 2003, Russell Sage Foundation Series on Trust,
New York.
[3] IBM Corporation. Enterprise Wide Security Architecture and Solutions Presentation Guide. 1st
Edition, November 1995.
[4] Rahman. A. and Hailes, S. 1997 A Distributed Trust Model. New Security Paradigms Workshop
1997 ACM.
[5] Blaze, M., Feigenbaum, J. and Lacy, Z. 1996 Decentralized Trust Management. In Proceedings of
the IEEE Symposium on Security and Privacy.
[6] Farooq Anjum, Dhanant Subhadrabandhu and Saswati Sarkar, 2003 Signature based Intrusion
Detection for Wireless Ad-Hoc Networks: A Comparative study of various routing protocols. In
the of proceedings of IEEE 58th Conference on Vehicular Technology.
[7] Marc Branchaud, Scott Flinn, x Trust: A Scalable Trust Management Infrastructure.
[8] Jin-Hee Cho, Ananthram Swami, and Ing-Ray Chen, A. 2011 Survey on Trust Management for
Mobile Ad Hoc Networks. In IEEE communications surveys & tutorials.

221


[9] Asmidar Abu Bakar, Roslan Ismail, Jamilin Jais, 2009 Forming Trust in Mobile Ad -Hoc
Network. In IEEE International Conference on Communications and Mobile Computing.
[10] Ruidong Li , Jie Li , Peng Liu, and Jien Kato, 2009 A Novel Hybrid Trust Management
Framework for MANETs. In IEEE 29th IEEE International Conference on Distributed
Computing Systems Workshops.
[11] Hamed Samavati, Behrouz Tork Ladani, Hossein Moodi, 2011 AMLeT: Adaptive Multi Level
Trust framework for MANETs. In the International symposium on CNDS.
[12] Dr. Sanjeev sharma, Renu mishra, and Inderpreet kaur, 2010 New trust based security approach
for ad-hoc networks. In IEEE.
[13] Menaka Pushpa, A., 2009 Trust Based Secure Routing in AODV Routing Protocol. In IEEE.
[14] Mangrulkar, R. S., and Dr. Mohammad Atique. 2010 Trust Based Secured Adhoc on Demand
Distance Vector Routing Protocol for Mobile Adhoc Network. In IEEE.
[15] Hothefa Sh.Jassim, Salman Yussof, Tiong Sieh Kiong, Koh, Roslan Ismail, S. P. 2009 A Routing
Protocol based on Trusted and shortest Path Selection for Mobile Ad hoc Network. In the
proceedings of 9th Malaysia International Conference on Communications, IEEE.
[16] Rajan Shankaran, Vijay Varadharajan, Mehmet Orgun, A., and Michael Hitchens. 2009 Critical
Issues in Trust Management for Mobile Ad-Hoc Networks. In IEEE IRI , Las Vegas, Nevada,
USA.

222

A survey on trust based routing in manet

More Related Content

What's hot (17)

Viewers also liked (9)

Similar to A survey on trust based routing in manet (20)

More from IAEME Publication (20)

A survey on trust based routing in manet