Research Article Security-Based Mechanism for …downloads.hindawi.com/journals/misy/2016/5653010.pdfSecurity-Based Mechanism for Proactive Routing Schema ... many problems in mobile

Research ArticleSecurity-Based Mechanism for Proactive Routing Schema UsingGame Theory Model

Hicham Amraoui1 Ahmed Habbani12 Abdelmajid Hajami3 and Essaid Bilal4

1SIME Lab MIS Team ENSIAS Mohammed V University Rabat Morocco2LEC Lab EMI Mohammed V University Rabat Morocco3LAVETE Lab FST Hassan I University Settat Morocco4Research and Development OCP Casablanca Morocco

Correspondence should be addressed to Hicham Amraoui amraouihicham1gmailcom

Received 25 August 2016 Accepted 12 October 2016

Academic Editor Jose M Barcelo-Ordinas

Copyright copy 2016 Hicham Amraoui et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Game theory may offer a useful mechanism to address many problems in mobile ad hoc networks (MANETs) One of the keyconcepts in the research field of such networks with Optimized Link State Routing Protocol (OLSR) is the security problemRelying on applying game theory to study this problem we consider two strategies during this suggested model cooperate andnot-cooperate However in such networks it is not easy to identify different actions of players In this paper we have essentiallybeen inspired from recent advances provided in game theory to propose a new model for security in MANETs Our proposalpresents a powerful tool with a large number of players where interactions are played multiple times Moreover each node keepsa cooperation rate (CR) record of other nodes to cope with the behaviors and mitigate aggregate effect of other malicious devicesAdditionally our suggested security mechanism does not only take into consideration security requirements but also take intoaccount system resources and network performances The simulation results using Network Simulator 3 are presented to illustratethe effectiveness of the proposal

1 Introduction

In our everyday life we are very interested in and dependenton wireless connection technology In addition the use ofmobile devices and applications based on wireless networksis continuously increasing day after day However this maygenerate several kinds of problems in terms of communi-cation between mobile devices in some difficult situationsThese problems can be observed especially where a networkinfrastructure is missing Therefore we need a powerfuland efficient mobile ad hoc network (MANET) to ensureand improve communication between devices in differentsituations such as military fields conferencing and sensornetworks

MANETs are a collection of wireless mobile devices thatform a temporary network without an existing infrastructureor a centralized administration Furthermore and due tolimited transmission range of wireless interfaces it may be

necessary for one node to identify other nodes to forwardtheir packets to its destinations In such networks each nodedoes not merely work as a host for transmitting and receivingdata but acts as a router or gateway for routing packets fromother nodes as well Moreover each node participates inrouting process that allows it to establish paths to reach anypossible destination inside the network In addition all nodesdynamically establish paths between themselves to createnetwork infrastructure that depends on individual behaviorof nodes Along these lines and in a spontaneous natureof MANET all nodes move randomly across the networkbecause of the nodes mobility and bandwidth-constrainedwireless connection for communicating with each otherThisnature permits infiltrating and disrupting network perfor-mances by malicious and selfish nodes Thereby maliciousbehavior represents one of the most famous challenges anddestructive routing problems that can influence networkperformances Additionally the concept of selfish node attack

Hindawi Publishing CorporationMobile Information SystemsVolume 2016 Article ID 5653010 17 pageshttpdxdoiorg10115520165653010

2 Mobile Information Systems

is based on the absorbing of significant amount of trafficdropping received packets and not cooperating during thepacket routing process However this problem arises whenexamining network topology where malicious nodes cannotbe easily detected

MANETs are infrastructureless and self-organized allnodes have to cooperate between themselves in order toprovide the best performances and offer necessary networkfunctionalities The cooperation mechanism must complywith the rules imposed by the routing protocol for trans-mitting and receiving data On the other hand the non-cooperation behavior can be produced by nodes that donot follow these rules However all nodes act as routers orgateways and contribute to discovering and maintaining therouting process Moreover each node is constrained in termsof limited resources (energy etc) and it may not always beinteresting to accept relay requests that require consumptionof most resources Therefore a cooperative system shouldbe integrated and incorporated with any network operationssuch as packet forwarding and route discovery The goal ofthis system is to prevent malicious behavior and encouragenodes to cooperate with each other However the applicationof cooperation mechanism is particularly difficult because ofMANET features that impose certain requirements

In this context many researchers have investigated selfishnodes and security problems in MANETs In this way theauthors in [1] proposed a solution to improve network per-formances when attacks are launched and mitigate aggregateeffect especially that all nodes in such networks are vulner-able to being isolated by malicious nodes Furthermore theauthors in [2] presented a powerful intrusion detection sys-tem called Enhanced Adaptive ACKnowledgment (EAACK)Compared to other intrusion detectionmechanisms EAACKshows an efficient attack detection without affecting networkperformances Likewise the authors in [3] surveyed theimpact of packet dropping attacks in MANET In their workthe authors try to demonstrate the importance of attackselaborate a new detection system and avoid maliciousnodes during communication with each other AdditionallyMANET nature makes it more attractive to many types ofattacks In this way the authors in [4] proposed a surveyin two parts the first one addresses important securitymechanisms and types of attacks that can affect networkperformances especially in the network layerThe second oneaddresses the classification of detectionmechanisms that dealwith a single or series of attacks Furthermore in the workpresented in [5] the authors suggested a powerful solutioncalled I-Watchdog protocol to detect malicious nodes inMANETs In addition the authors proved through simu-lations the effectiveness of this solution with Destination-Sequenced Distance-Vector (DSDV) routing protocol interms of packet drop ratio (PDR) throughput and end-to-end delay

Recently game theory provides a useful solution formodeling and addressing different problems in MANETs Insuch networks players (nodes) have conflicting objectivesand different profiles with each other Additionally in thegame theory a utility function represents a payoff (reward)that allows each player to evaluate a particular outcome that

reflects its objectives The utility of each player depends notonly on its actions (strategies) but also on othersrsquo actionsIn addition a security scheme must take into account pastand current strategies of different players to be successful inMANETs

In the same context our research is focused onmobile adhoc networks using Optimized Link State Routing Protocol(OLSR) which is a proactive protocol In such networksOLSR is one of the most used routing protocols Moreoverthe cooperation concept is an essential element that can resultin the evolution of network performances in MANET In thisway and in this paper the cooperation rate (CR) representsthe value which indicates howmany times a node cooperatesor not during the game (or during network lifetime)Throughthis value each node can evaluate the behavior of anothernode before sending a packet Moreover in this paper athreshold is considered as the minimum value of the CRaccepted by all rational nodes We consider that each nodewhich has aCRgt 0 is considered as a legitimate nodewhereasa node with CR lt 0 is considered as a noncooperative nodeTherefore our contribution is briefly summarized below

(i) Firstly our conduct is to put forward an enhancedalgorithm based on game theory and establishingconfident relationships between nodes In this pro-posed model each node keeps a cooperation rate(CR) record of other nodes to evaluate their behaviorsand avoid malicious nodes

(ii) Secondly the calculation of CR is based on OLSRmessages (HELLO and topology control (TC))exchanged between nodes and forwarding processes

(iii) Thirdly the cooperation rate will be shared betweennodes in addition to other network information usingHELLO and TC messages

(iv) Fourthly the key novelty of this paper is that the valueof CR will be used as a metric to construct routingtables instead of hop count metric used by OLSRstandard

The remainder of this paper is organized as follows TheOLSR routing protocol as a proactive scheme is presentedin Section 2 Some previous studies that aim at addressingcooperation and selfish behavior in MANETs are presentedin Section 3 A game model formulation is described inSection 4 Our suggested systemmodel based on cooperationbetween nodes is discussed in Section 5 A malicious detec-tion algorithm and enhanced routing table computation areintroduced in Section 6 A simulation environment used toaddress our approach is discussed in Section 7 The resultsthat concern the validation of our solution are presented inSection 8 Finally the paper is concluded in Section 9 withfuture work

2 Proactive Schema Case of OLSR

OLSR is a proactive routing protocol [6] based on MPR(multipoint relay) mechanism that is considered as the keyconcept used in this protocolTheMPRs are used tomaintain

Mobile Information Systems 3

N3

N8

N10

N7

N0N4

N2

N1

N6

N12

N5

N9

N11

TC messageHELLO message

One-hop neighbors

Two-hop neighbors

Multipoint relay (MPR)

Malicious message

Malicious node

Figure 1 Attack example in MANETs

routing tables and topology control In addition and owing tothe proactive nature ofOLSR the control of the state links andpaths is done proactively and periodically In the same waythe optimization in this protocol can be done in two stepsthe first one uses control messages with reduction in sizeThesecond one uses a reduced number of links to forward the linkstate packets In addition this reduction is made by declaringonly a subset of links in link state updates Moreover eachnode in OLSR uses HELLO messages to find its one-hopand two-hop neighbors through their repliesThe transmitternode can select its MPR based on its one-hop neighborsset that offers the best reachability to nodes belonging tothe two-hop neighbors Furthermore the transmitter usesTC (topology control) messages to declare a set of links(advertised link set) that must include at least the links to allnodes of its MPR selector set [6]

The routing inMANETs usingOLSR is amethod throughwhich each node sends information to a quite precise recip-ient The problem of routing is limited not only to howto find a path between two nodes inside the network butalso to how to find an optimal and secure routing pathHowever in such networks one of the major problems ofrouting processes is the noncooperative and selfish behaviorsas denoted in Figure 1 In these cases the noncooperativeand selfish nodes take advantage of legitimate nodes and donot cooperate with others in order to save resources for theirown communication Thereby network resources becomeunavailable for legitimate nodes

3 State of the Art

MANETs are used in a wide range of applications in variousfields For successful execution of different operations in

such networks routing processes are the most importantoperations that improve network performances Thereforemany researches have been reported in the literature toaddress routing processes In this way the authors in [7]presented an incentive solution for probabilistic routing inorder to stimulate selfish nodes to cooperate with others Inaddition the authors proved properties of this solution andextensively evaluated it using GloMoSim Furthermore theresult presents more than 758 of the gain concerning deliv-ery ratio compared to other probabilistic routing protocolswithout incentive

On the other hand and in order to address joint routingnetwork coding and scheduling problems matrix gametheoretic models which are based on a nonlinear cubicgame have been proposed in several works such as [8]The authors in this work due to necessity of the inherentmulticast gain of network proposed a new approach based ona compressed topology matrix to model routing and networkcoding problems Additionally the authors proposed a newapproach called Network Graph Soft Coloring (NGSC) tooptimize scheduling problems Furthermore the authors in[9] presented a solution based on a two-hop relaywith limitedpacket redundancy 119891 to propose a forwarding game andaddress the optimal forwarding problem in MANETs In thisgame each node (119894) chooses a strategy with probability 119879119894(119879119894 isin [0 1]) to send and forward its own traffic Additionallyeach node (119894) helps to forward other trafficwith probability119901where 119901 = 1minus119879119894 while its payoff is the attainable throughputcapacity of its own traffic

In wireless ad hoc networks the routing is the mostimportant process that needs cooperation between nodesThus some cooperation schemes and trusted models havebeen proposed in several works such as [10ndash15] The mainobjectives of these works are the following (i) to enforcecooperation between nodes and (ii) to evaluate and addressaggregate effect ofmalicious nodesMoreover and in order toreach these objectives the authors presented many solutionssuch as collaborative reputationmodel a game theoretic trustmodel collaborative caching priority coalition formationand cooperation strategies for processing requests In addi-tion the authors in [16] presented in noncooperative wirelessad hoc networks a study of collusion-resistant routing Thiswork is based on two solutions Group Strategy Proofnessand Strong Nash Equilibrium for collusion resistance ingame theory Also the authors proposed a cryptographicmechanism to avoid profit transfer among colluding players(nodes) In the same context the authors in [17] usedthe game theory to address cooperation incentive of nodesbased on reputation mechanisms price-based systems and asystem without cooperation incentive strategy Through thiswork a strategy based on a threshold to determine nodereliability and reward cooperative nodes may bemanipulatedby selfish nodes In addition the authors in [18] proposeda powerful solution built on Mean Field Game (MFG)approach with multiple players for security enhancementsin MANET Based on recent advances in MFG theory thisapproach permits enabling each node to elaborate strategicsecurity defense decisions Additionally this approach takesinto account system resources permits each node to know its


own state information and evaluate aggregate effect of othernodes However the authors studied the interactions betweennodes and only one attacker

InMANETs nodesmust cooperate between them to sendand forward packets from sources to destinations In this waythe work presented in [19] showed that node misbehaviorproblems can influence MANETs and sensor networks per-formances In addition and in order to avoid this problemthe authors proposed a solution adapted to wireless multihopnetwork in order to deal with collusive networking behaviorbased on game theory Additionally this solution is derivedfrom recent works that are based on the theory of imperfectprivatemonitoring for the dynamic BertrandOligopoly Alsothe authors showed the effectiveness of this solution undera wireless environment Along these lines and due to theimportance of the cooperation concept the authors in [20]proposed a solution called Finite-Time Reputation System(FITS) that uses a new technique named Threat To Interfere(TTI) to enforce cooperation between nodes In addition thismechanism is based on two solutions the first one calledFITS-D needs a Perceived Probability Assumption (PPA)The second one called FITS-I uses more techniques to avoidthe necessity of PPAMoreover this work showed that both ofschemes have a SubgamePerfectNash Equilibrium (SPNE) inwhich the probability of forwarding packet of nodes is closeto one

In the same context the authors in [21 22] proposeda secure routing protocol to protect nodes from anony-mous behaviors These works are based on game theorywhich provides a powerful tool to analyze formulate andaddress selfish behaviors In addition these authors usedthe Dynamic Bayesian Signaling Game (DBSG) to analyzestrategy profiles for rational and malicious nodes to findthe best strategies for each player (node) Furthermore theauthors studied the equilibrium by combining strategies andutility functions (payoff) of nodes to solve this incompleteinformation problem Moreover and to reach this goal theauthors used Perfect Bayesian Equilibrium (PBE) that offersan important solution for signaling games Likewise theauthors presented in [23] a solution to deal with selfishnessand moral hazard in noncooperative wireless networks Inaddition they proposed a solution based on several methodsthat discourage hidden actions under secret informationFurthermore some mechanisms for routing scenarios havebeen proposed for instance each malicious node tries tomaximize its utility functionwhen it sincerely declares its costand actions Also the authors proved through simulationsthat payments are larger compared to current cost incurredby all intermediate devices

Along these lines and in order to detect and isolatepacket dropping attacks efficiently the work in [24] proposeda protocol named SADEC (Stealthy Attacks in WirelessAd Hoc Networks Detection and Countermeasure) Thisprotocol is based on two techniques the first one is basedon how to keep additional information about routing pathsby neighbors The second one is based on how to add somechecking mechanisms to each neighbor This protocol canoffer a powerful solution to use local monitoring In additionthe authors showed by simulations the effectiveness of the

protocol in how to reduce the impact of packet droppingattack In the same way the authors in [25] proposed asolution based on Social Network Analysis (SNA) to developan intrusion detection mechanism (SN-IDS) in MANETsusingMAC and network layers data After that these authorsselected relevant social functionalities and constructed a setof sociomatrices Moreover these authors showed that thesemethods based on social analysis can be applied to thesematrices to detect malicious activities of mobile nodes usingmultiple rules

Similarly in the work proposed in [26] the authorspresented an intrusion detection system to detect attacksequences in MANET using MAC layer applications Thissystem can be applicable to MANET environment based onstable and efficient attack observations In such a way thesolution presented in [27] suggested an intrusion detectionand adaptive response mechanism to provide an effectivereply in case of a range of attacks in MANETs This solutionin order to offer a better security requirement proposeda flexible response scheme based on effectiveness level ofnetwork performances measured confidence and the impactof attacks In addition the authors in [28] proposed asolution called Sentinel Protocol (SP) to detect and dealwith replica attacks that can influence network perfor-mances The main objective of this attack is that maliciousnodes deploy a large number of replicas of compromisedor captured devices across the network Furthermore theauthors proved through simulations the effectiveness of thisprotocol

Due to the importance of the routing efficiency indelay tolerant networks the authors in [29] suggested anenhanced routing protocol which is based on the social linkawareness The main objective of this algorithm is to avoidthe selfish nodes and solve the problems of intermittentconnection and high latency in order to improve the routingprocess In addition the proposed algorithm used the sociallinks to construct the friendship communities of the nodesMoreover different mechanisms such as the intracommunityand intercommunity forwarding are implemented to improvenetwork performances in terms of the successful deliveryratio with low overhead and decrease the transmission delayIn the work presented in [30] the authors proposed a solutionbased on game theory and load feedback control (LFC) withprice elasticity to maximize profit benefits for distributedgenerations (DGs) for their participation in energy lossreduction In addition the proposed model can be used toreward DGs and improve their profit by using the game the-ory approach Moreover and where a distributed locationalmarginal pricing (DLMP) feedback signal is calculated bycustomer demand the proposed mechanism can be used toregulate peak-load value of multiple customers by using anLFC submodel with price elasticity In addition the authorsin [31] proposed a global punishment-based repeated gamemodel to enforce the cooperation between nodes acrossthe network Additionally when the whole network is in acooperative state the authors investigated the equilibriumconditions of packet forwarding strategies by taking intoaccount rational nodes Moreover a metamodel is used todesign forwarding strategies in order to reduce the impact


Table 1 A duality between a game approach and a MANET

Elements of agame Elements of a mobile ad hoc network

Players Nodes

StrategyAction linked to each player to evaluate its utility

In our game we consider two strategiescooperate and not-cooperate

Utilityfunction

(i) Performance metrics (throughput packetforwarded packet received and end-to-end

delay)(ii) The cooperation rate (CR) of each node

(player)

of selfish nodes on network performances and encourage thecooperation between mobile nodes

Recently the effective cooperation incentive of nodeshas become a hot issue in cooperative communication suchas mobile ad hoc networks In such a way the authors in[32] proposed a topology transform-based recommendationtrust model to stimulate the cooperation between nodesand mitigate effect of selfish behaviors Furthermore themodel is used to mitigate the aggregate of malicious effectson the accuracy of recommendation trust which resultfrom fake recommendation In addition the authors usedsome mathematical models and simulation to ensure theeffectiveness of their proposed model

To address these problems and imperfections andthrough this paper our concern is to design a new algorithmof cooperation based on relationships between nodes Thenwe will compare the proposal with the original OLSR and aselfish OLSR protocol after that we integrate it with originalOLSR Additionally we address the proposal based on amathematical model and set of simulations Furthermorethe main objective is to be fully extended to universal adhoc networks and practical MANET applications especiallyrouting processes and malicious node detection

4 Game Model Formulation

41 Modeling Ad Hoc Network as a Game In this sectionwe propose a description of a mobile ad hoc network 119866which is formed by a set of mobile nodes using the gametheory approach This formulation contains a set of nodes(players) denoted by (119873) a strategy space denoted by (119878) anda utility function denoted by (119865) Thus the network can beexpressed by 119866 = 119873 119878 119865 Table 1 presents briefly a dualitybetween a game approach and the mobile ad hoc network inour situation

In the abovementioned network 119866 each node has autility function 119865 that represents the payoff of each player(node) across the network In addition a utility functionrepresents a payoff (reward) that allows each player toevaluate a particular outcome which reflects its objectivesThe main objective of all nodes (players) is how to maximizeor minimize the utility function depending on a contextIn the same way each player acts as a relay or gateway

for routing packets from other players based on availablerouting and topology tables In addition each player (119894)chooses its strategy 119878119894 from the strategy space 119878 defined by119878 = C cooperate NC not-cooperate (cooperate meansto participate in packet forwarding and not-cooperate meanspacket dropping)

42 Static and Repeated Game Approach To analyze the out-come of the static game our two-player game is similar tothe prisoners dilemma game [33] Each player can choosedifferent strategies cooperate (C) or not-cooperate (NC) Ifone of the two players chooses to cooperate it will act as arouter or gateway for the other player However if the playerchooses the not-cooperate strategy it will forward its ownpackets and will not participate in routing packets for theother player

In this paper we consider that if a player chooses tocooperate it will be rewarded by a lot of information (ACKstopology control links update routing of packets etc) thisreward is denoted by 119881 but at the same time it will lose acost denoted by (119890) However if the two players choose not-cooperate strategy both of them will lose the informationalready mentioned above

Let us denote by (119881 minus 119890) the reward of each player thatchooses to cooperate and by (119881) the reward of the playerthat chooses not-cooperate in case the first player choosesto cooperate and by (minus119903) the punishment that each playerreceives if both choose not-cooperate strategy Therefore inthe rest of this paper we assume that 119881 gt (119881 minus 119890) gt minus119903

The only optima equilibrium if the two players arerational is the strategy profile (119881 minus 119890 119881 minus 119890) where the firststrategy denoted in the pair is that of player (1) and the secondis that of player (2) This strategy profile will be available onlyif the two players choose the cooperate strategy Moreoverthis situation cannot be realized in all static games due to aselfish behavior of some players However the profile (minus119903 minus119903)where the two players choose the not-cooperate strategy isundesirable from the network perspective

In our situation we consider that the past strategiesinfluence the payoff (utility) function in current period(stage) Thus the game can be analyzed using the repeatedgame approach [34 35] where all players face the same staticgamemany times and in every period 119905Therefore we chooseto apply the repeated game approach in our situation for thefollowing reasons

(1) The game or nodes interactions are played severaltimes In addition when a node (player) takes intoconsideration the impact of its current strategy onfuture actions of other nodes the game is calledrepeated game

(2) During this kind of games all nodes (players) canobserve different actions of other players and thischaracteristic helps to adapt their actions (strategies)to respond to other players especially that each nodekeeps track of the cooperation rate (CR) record ofother nodes


Table 2 Payoff matrix of two-player game in strategic form

Player (1) Player (2)C NC

C (119881 minus 119890 119881 minus 119890) (119881 minus 119890 119881)NC (119881 119881 minus 119890) (minus119903 minus119903)

(3) Furthermore selfish players act as routers or gatewaysonly to their interest without taking into consider-ation network performances So we can define andimpose some rules to enforce cooperation betweennodes In addition these rules can be modeled usingthe repeated game

(4) These rules can be implemented to reach a desirableresult of developed games Moreover repeated gamessupport different equilibrium solutions which areadapted for many requirements of ad hoc networks

In this paper and in order to enforce cooperation betweennodes each player keeps track of the cooperation rate (CR)record of other players as a rule in this gamemodelThemainobjective of this rule is to show the importance of cooperationpotential benefits through interactions between nodes Alsothis rule can be modeled in a repeated game

43 Problem Formulation and Nash Equilibrium

431 Pure Strategy In this section we consider a problemthat may exist in different types of networks where optimiza-tion of communication is very important In our study weconsider a flowof network traffic generated by a finite numberof nodes (players) In addition each node knows a list ofpaths that fits its strategy and its objective is to maximizeits utility function The situation where all players maximizetheir utility functions is known asNash Equilibrium (NE) [3136] In the repeated and noncooperative gamemodels theNEis used to predict the stable situation where no player (node)has nothing to gain by changing its strategy unilaterally

In the same context and in this pure strategy a NashEquilibrium is a strategic profile 119878lowast = 119878lowast1 119878lowast2 119878lowast119899 suchthat each player (119894) has its utility 119880119894 and for each strategy1198781015840119894 isin 119878119894

119880119894 (119878lowast119894 119878lowastminus119894) ge 119880119894 (1198781015840119894 119878lowastminus119894) (1)

where 119878lowast119894 is the best response of player (119894) 119878lowastminus119894 are the bestresponses of other players and 119878119894 is the set of strategiesof player (119894) In addition we are dealing with a dynamicgame with 119873 players (nodes) playing a repeated game Thepayoff of different profiles in strategic form is presented in(bimatrix) Table 2 with cooperate strategy denoted by C andnot-cooperate strategy denoted by NC

We use the strategic form because our game is consideredas a simultaneous game where both players can choose theirstrategies simultaneously

Table 3 Payoff matrix in mixed strategy of two-player game instrategic form

Player (1) Player (2)C (119901) NC (1 minus 119901)

C (119902) (119881 minus 119890 119881 minus 119890) (119881 minus 119890 119881)NC (1 minus 119902) (119881 119881 minus 119890) (minus119903 minus119903)

Based on the matrix payoff (Table 2) if one of the twoplayers chooses to cooperate and if the other player choosesnot-cooperate strategy thus the payoff of the second playeris improved from (119881minus119890) to (119881) In addition if one of the twoplayers chooses not-cooperate strategy and the other playeralso chooses the same strategy then the payoff of the secondplayer is decreased from (119881minus119890) to (minus119903) Furthermore we notethat any strategy (cooperate or not-cooperate) cannot alwaysoffer a better utility to each player in different situationsThusa dominant or dominated strategy does not exist Howeverin terms of stability this game supports two Nash Equilibria(NE) (119881 minus 119890 119881) and (119881 119881 minus 119890) In both situations of NEno player can profitably change its strategy Furthermore(119881 minus 119890 119881 minus 119890) and (minus119903 minus119903) cannot be NE because the twoplayers would have an incentive to change their strategies Inthis game the two NE are considered as situations of stabilitybut are not equitable because only one of the two playerscan be rewarded Additionally the (minus119903 minus119903) strategy profile isundesirable from the network context

432 Mixed Strategy A mixed strategy of a player (119894) is aprobability distribution 120590119894 defined upon all its pure strategiesLet us denote by sum119894 all mixed strategies of player (119894) and by120590119894 a mixed strategy of this player

A mixed strategy Nash Equilibrium is a mixed profile ofstrategies 120590lowast isin sum119894 such that for each player (119894) and for all120590119894 isin sum119894


where 120590lowast119894 is the best response of player (119894) and 120590lowastminus119894 are the bestresponses of other players

In the mixed strategy and to analyze the outcome of thestatic game each player chooses a strategy cooperate (C) withprobability 119901 (or 119902) and the other strategy not-cooperatewith probability (1 minus 119901) or (1 minus 119902) Table 3 presents the payoffmatrix of the two players in the mixed strategy

Let us denote by 1198801(C) the average utility of player (1)when it chooses cooperate strategy Thus the average utility1198801(C) can be written as

1198801 (C) = ((119881 minus 119890) times 119901) + ((119881 minus 119890) times (1 minus 119901)) = 119881 minus 119890 (3)

Let us denote by1198801(NC) the average utility of player (1)whenit chooses not-cooperate strategy Thus the average utility1198801(NC) can be written as

1198801 (NC) = (119881 times 119901) + ((minus119903) times (1 minus 119901)) (4)


At the mixed strategy Nash Equilibrium 1198801(C) = 1198801(NC)(ie (3) = (4)) Then

(119881 minus 119890) = (119881 times 119901) + ((minus119903) times (1 minus 119901)) (5)

Equation (5) can be written as

(119881 minus 119890) + 119903 = 119901 times (119881 + 119903) (6)

Therefore

119901 = ((119881 minus 119890) + 119903)(119881 + 119903) (7)

Thus (7) can be written as

119901lowast = 1 minus ( 119890(119881 + 119903)) (8)

where 119901lowast represents the probability at the mixed strategyNash Equilibrium

In this game 119903 represents a punishment that needs topenalize the players and encourage them to cooperate Inaddition if the value of 119903 is very high (see infinity) the playerswill tend to cooperate in order to avoid this punishmentTherefore we can calculate the limit of 119901lowast when 119903 approachesinfinity (119903 rarr infin)

lim119903rarrinfin

119901lowast = 1 (9)

We can follow the same operations concerning player (2)because the game is symmetrical therefore

119901lowast = 119902lowast (10)

Thus the mixed strategy (119901lowast 119902lowast) is a Nash EquilibriumHowever in case of (119873) players the situation can be

considered as the volunteerrsquos dilemma game [37 38] Inaddition we can demonstrate that in such a situation thecooperation between nodes decreases

Therefore in this case and from Table 3 we can calculatethe average utility of each player (119894) depending on actions ofother players Thus we will study two cases

Case 1 Let us denote by 119880119894(C) the average utility of player(119894) if it chooses to cooperate Then we have to study twosubcases

Case 11 If at least one of the other players chooses tocooperate

119880119894 (C) = (119881 minus 119890) times (1 minus (1 minus 119901)(119899minus1)) (11)

where (1 minus (1 minus 119901)(119899minus1)) is the probability that at least one ofthe other players chooses to cooperate

Case 12 If no player chooses to cooperate

119880119894 (C) = (119881 minus 119890) times (1 minus 119901)(119899minus1) (12)

where ((1 minus 119901)(119899minus1)) is the probability that no player choosesto cooperate

Then the average utility119880119894(C) of player (119894) can be writtenas

119880119894 (C) = (11) + (12) (13)

So

119880119894 (C) = ((119881 minus 119890) times (1 minus (1 minus 119901)(119899minus1)))+ ((119881 minus 119890) times (1 minus 119901)(119899minus1))

(14)


119880119894 (C) = (119881 minus 119890) (15)

Case 2 Let us denote by 119880119894(NC) the average utility of player(119894) if it chooses not-cooperate strategy In this case we have tostudy two subcases as well


119880119894 (NC) = (119881 times (1 minus (1 minus 119901)(119899minus1))) (16)



119880119894 (NC) = (minus119903) times (1 minus 119901)(119899minus1) (17)


Then the average utility 119880119894(NC) of player (119894) can bewritten as

119880119894 (NC) = (16) + (17) (18)

So

119880119894 (NC) = (119881 times (1 minus (1 minus 119901)(119899minus1)))+ ((minus119903) times (1 minus 119901)(119899minus1))

(19)


119880119894 (NC) = 119881 minus ((119881 + 119903) times (1 minus 119901)(119899minus1)) (20)

At the mixed strategy Nash Equilibrium 119880119894(C) = 119880119894(NC)(ie (15) = (20)) Thus

(119881 minus 119890) = 119881 minus ((119881 + 119903) times (1 minus 119901)(119899minus1)) (21)


(1 minus 119901)119899minus1 = 119890(119881 + 119903) (22)

So

(1 minus 119901) = ( 119890(119881 + 119903))

(1(119899minus1))

(23)


Therefore

119901 = 1 minus ( 119890119881 + 119903)

1(119899minus1) (24)

So

119901lowast = 1 minus (119899minus1)radic( 119890119881 + 119903) (25)

where 119901lowast represents the probability at the mixed strategyNash Equilibrium In addition we can follow the same opera-tions concerning player (2) because the game is symmetricaltherefore

119901lowast = 119902lowast (26)

Thus when the number of players is increased (ie when 119899approaches infinity) the limit of 119901lowast when 119899 rarr infin is

lim119899rarrinfin

119901lowast = 0 (27)

Therefore we notice that in such a situation where thenumber of players increases the cooperation between nodesdecreases as well and becomes more interesting to encouragenodes (players) to cooperate In addition we notice that thenoncooperative strategy can offer a selfish player to takeadvantage of a cooperating player Therefore we must takeinto account a cooperative system to deal with this behaviorand enforce cooperation between nodes In addition thiscooperative system must offer each node (player) a rewardfor cooperating and impose a punishment on each node fornot cooperating

Thus let us denote by ℎ(119905) the cooperation utility (thecooperation history) of the 119894th player in the entire reputationgame and in each stage or period 119905 The utility is the sum ofits utilities in all stages Additionally let us denote by 120573(119905) thevalue added or subtracted periodically according to playerbehavior (cooperate or not-cooperate) during the game inorder to update its ℎ(119905) The cooperation rate (CR) of eachplayer (119894) is calculated using the following equation

CR = sum(ℎ (119905) + 120573 (119905)) (28)

In the next section we propose a mathematical model wherewe formulate the calculation of the cooperation rate (CR)

5 System Model

51 Cooperation-Based Mechanism In other similar gametheory models which have been cited above in related worksection a reputation entity such as the watchdog is used todetect misbehaving nodes In addition and in every time amonitoring entity needs to monitor and verify the correctexecution of a function However this mechanism is basedon an assumptionwhich is not always true and requiredmoreenergy consumptionMoreovermany other gamemodels arenot adequate to OLSR routing protocol

Concerning our proposed model we have selected forperformance evaluation the OLSR protocol that considers

the stability of links The key novelty of this model is tostimulate the cooperation between nodes in a MANET usingthe cooperation rate (CR) in order to prevent selfish behaviorAdditionally the CR value is calculated based on varioustypes of specific OLSR messages (HELLO and topologycontrol (TC)) and different network operations (forwardingand routing) In our proposal the correct execution of afunction is according to the player behavior cooperate ornot-cooperate (cooperate means to participate in packetforwarding and the exchange of OLSRmessages (HELLO andTC) with reception of ACKs and not-cooperatemeans packetdropping)Thus this process ensures the correct execution ofan OLSR function

In this paper we propose a new strategy based on thegame theory to enforce the cooperation between nodes bycalculating a cooperation rate (CR) for each node Addi-tionally this strategy has evaluated using OLSR messages(HELLO andTC) and different network processing (forward-ing and routing) In the rest of this section we propose amathematical model where we formulate the calculation ofthe cooperation rate (CR)

In the rest of this section we propose a mathematicalmodel where we formulate the calculation of cooperation rate(CR) In this model each node (119895) can know the cooperationrate of a node (119894) inside the network

511 Cooperation Rate CR The cooperation rate (CR) of anode (119894) in relation to its neighbors set is directly calculatedfrom an observation of each node (119895) which belongs to theneighbors set 119867119894 of the node (119894) The CR at time interval 119905is calculated using a weighted average of the observationsrsquorating factors provided by nodes belonging to the neighborsset 119867119894 of the node (119894) Moreover and in order to (i) reacha better evaluation of node behaviors (ii) avoid incorrectdetections due to connection breaks (iii) and ensure thatthe nodes which are involuntary noncooperative due to theirlimited resources (energy levels etc) are not excluded fromthe network we should take into consideration a minimalimpact on the evaluation of the final cooperation value Inaddition the CR value is calculated periodically over a giventime interval (t) that depends on the default time of OLSRmessages exchanged between nodes Therefore in case ofHELLO message the time interval is 2 seconds in case ofTC message the time interval is 5 seconds and in case of aforwarding process it is directly calculated after the end of thisprocess Moreover in this paper the threshold is consideredas the minimum value of the CR accepted by all rationalnodes We consider that each node which has a (CR gt 0) isconsidered as a legitimate node whereas a node with (CR le0) is considered as a noncooperative node Moreover at thebeginning of this algorithm the cooperation rate of eachnodeis initialized by zero In addition all newly joined nodes willhave a CR which is initialized to zero as well

The equation that permits calculating the CR of node (119894)at time interval 119905 and based on a network operation 119865 is

CR (119894 119905 119865) = sum(ℎ (119905) + 120573 (119905)) (29)


Reserved Cooperation rate Htime Willingness

Link code Reserved Link message size

Neighbor interface address


0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9

middot middot middot

Figure 2 Enhanced HELLO message format

Reserved Htime Willingness




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


Figure 3 Standard HELLO message format

where 120573(119905) is the value added or subtracted periodicallyaccording to node behavior (cooperate or not-cooperate)during the game

120573 (119905)

= +1 +2 +3 +119898 if the node cooperates

minuslast value added if the node does not cooperate

(30)

(i) 119865 is the network operation (TC andHELLOmessagesprocessing and forwarding processes)

(ii) ℎ(119905) represents the cooperation rate (CR) record savedby a given node (119894) in relation to another node (119895)Also it is a time dependent function that gives higherrelevance to past values of CR Additionally (119905) isused to update the CR according to node behavior(cooperate or not-cooperate) during the game inorder to update its ℎ(119905) Also this value is influencedand depends on the observationsrsquo rating factors (othercooperation rates) provided at time interval 119905 byother nodes belonging to neighbors set119867119894 of node (119894)

512 Weighting Calculation The cooperation rate dependson different network function 119865 (HELLO and TC messagesprocessing and forwarding processes) Therefore during thecalculation of the cooperation rate wemust take into accountthe impact of each function 119865 according to its importance Inthisway and in order to calculate theweight119882 related to eachfunction 119865 we use AHP (analytic hierarchy process) method[39 40] In AHP method the decision process requires theexecution of the following stages

(1) Establish the main objective

(i) Choose a processing function

(2) Define the criteria

(i) Security routing and reliability

(3) Select options

(i) HELLO message processing(ii) TC messages processing(iii) Forwarding processing

In our case we consider that the security is the mostimportant criterion followed by routing process and reliabil-ityThe rest of theAHPprocess is very long sowe are going topresent the results directly and the CR of the node (119894) whichis presented in (29) can be written as follows

CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)

where (i)119882 = 1328 if the function 119865 is a TC processing (ii)119882 = 13060 if the function 119865 is a forwarding processing (iii)119882 = 12720 if the function119865 is a HELLOmessage processing

Moreover each node must share its correct cooperationrate (CR) with other nodes using OLSRmessagesWe presentin Figure 2 the enhanced format of HELLO message thatcontains the CR of the transmitter node and the standardformat is presented in Figure 3

We present in Figure 4 the enhanced format of TCmessage that contains the CR of the transmitter node andthe standard format is presented in Figure 5


ANSN Cooperation rate Reserved

Advertised neighbor main address



0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9

Figure 4 Enhanced TC message format

ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9

Figure 5 Standard TC message format

6 Malicious Node Detection Algorithm

In this section we propose an algorithm to detect and avoidmalicious behavior based on CR value of all nodes across thenetwork using the routing game approach during the routingtables computation

61 Routing Game The routing process requires cooperationbetween nodes for routing packets from other nodes There-fore the key novelty of this paper is to develop an algorithmbased on game theory to enforce cooperation between nodesin order to avoid selfish and malicious nodes during therouting process However the existence of malicious nodesin this area threatens cooperation and influences networkperformances (routing control lifetime etc) as denoted inFigure 6 Additionally each node (player) tries to reach thefollowing objectives it tries to maximize its utility functionminimize a path cost function or find an optimal and securerouting path In the game theory these objectives have beenaddressed inwhat is known as the routing gameMoreover ineach routing process every node chooses its path and updatesits strategy in terms of its utility function and the action it haschosen

We represent our routing game model using an undi-rected graph 119866 (119881 119864) where

(i) 119881 is the set of nodes or vertices(ii) E is the set of arcs (link) between nodes(iii) N denotes all players (nodes) where119873 = 1 2 119899(iv) any player (119899) isin 119873 is characterized by the following

information

(1) CR(119899) is utility or cooperation rate(2) A pair of vertices (119878119894 119879119894) isin (119881times119881) which repre-

sents its source and destination respectively(3) 119875119899 sub 119864 is set of the shortest paths ranging from

source 119878119894 to destination 119879119894 with cardinality119898119894

Table 4 Enhanced routing table format

Destination Next Next Cooperationaddress address interface rate

(4) A strategy space 119878 = C cooperate NCnot-cooperate indexed on the set 119875119899

(5) For each path (119894 119895) isin 119875119899

119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)

where (32) represents the utility function 119865 of the player (119899)in relation to the path (119894 119895) which belongs to 119875119899 and is basedon CR values ofM nodes belonging to this path In additionand in case of multiple choice each node chooses the pathwith greater value of this utility function 119865

The routing table computation is an essential processin OLSR protocol Therefore and in order to avoid com-munication with malicious nodes which may act as routersor gateways the calculated cooperation rate (CR) must beintegrated in the routing table as a newmetric in parallel withother information (destination address next address andnext interface) to establish secure routes between nodes Inthis way we propose a new routing table shape as mentionedin Table 4

62 Enhanced Routing Table Algorithm In this section wepresent a brief description of our enhanced routing tablealgorithm that provides the solution to avoid selfish nodes

BEGIN

(1) Based on modified HELLOmessage with the cooper-ation rate of nodes control all one-hop nodes


Malicious node

Malicious node Malicious node

Figure 6 Network routing as a game in MANET

(2) Add appropriate entries for each node to its routingtable using its one-hop table

(3) Update entries of routing table with the topology set(4) Keep recursively for each node its last address until

attaining the destination(5) Based on modified TC message with the cooperation

rate of nodes save all path information in the routingtable

(6) Delete the loop entries if any(7) For each node across the network select all paths of a

given source-destination(8) Evaluate the behavior of each node based on CR to

avoid selfish nodes on each path(9) Get the CR of each node on each path(10) For each node (119899) calculate the utility function

119865119899(119894 119895) on each selected path (119894 119895) using (32)(11) Find out the maximum utility function 119865 on each

selected path(12) Use this selected path

ENDIn the next subsection we present an example to give

a walk through example to explain our malicious nodedetection algorithm

63 Proposed Example In this example which is presentedin Figure 7 we propose a MANET with six nodes where thesource node (1) tries to send packets to its destination node(6) After (i) calculating the cooperation rate of each node asmentioned in Tables 5 6 7 8 and 9 (ii) sharing this valuebetween all nodes using HELLO and TC messages and (iii)introducing this value on routing tables the source node (1)must choose one short path among the two possibilities toavoid malicious nodes Therefore the source node (1) must

Source Destination

1

53

2 4

6

Figure 7 A sample MANET network with six nodes as routinggame

calculate its utility function in relation to the path (1 2 46) and path (1 3 5 6) based on CR values of the nodesbelonging to these paths In addition the source node willchoose the pathwith greater value of this utility function119865 Inthis example we propose that the calculating of cooperationrate is made after five iterations needed to exchange OLSRmessages (HELLO and TC) Additionally we suppose thatnode (5) is considered as a malicious node and does notcooperate with other nodes

In this example we suppose that iteration 1 means thatnode (2) and node (1) exchange the HELLO message andboth of them received a reply (the ACK) it means also thatthe link between them is symmetric Therefore the CR ofnode (2) in relation to node (1) which is initialized by zerowill be updated by adding 1 In addition these nodes willexchange the HELLO message in the second iteration andboth of them received the ACK and the CR will be updatedby adding 2 Furthermore the nodes will exchange the TCmessage in iteration 3 and the CR will be updated again byadding 3 and so on

CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)


Table 5 Cooperation rate of node (2) in relation to nodes (1) and (4)Node (1) Node (4)

Node (2) iteration 1 +1 (ACK is received) +1 (ACK is received)Node (2) iteration 2 +2 (ACK is received) +2 (ACK is received)Node (2) iteration 3 +3 (ACK is received) +3 (ACK is received)Node (2) iteration 4 +4 (ACK is received) +4 (ACK is received)Node (2) iteration 5 +5 (ACK is received) +5 (ACK is received)




Node (3) iteration 1 +1 (ACK is received) +1 (ACK is received)Node (3) iteration 2 +2 (ACK is received) +2 (ACK is received)Node (3) iteration 3 +3 (ACK is received) minus2 (ACK is not received)Node (3) iteration 4 +4 (ACK is received) minus1 (ACK is not received)Node (3) iteration 5 +5 (ACK is received) minus1 (ACK is not received)


Node (5) iteration 1 +1 (ACK is received) +1 (ACK is received)Node (5) iteration 2 +2 (ACK is received) +2 (ACK is received)Node (5) iteration 3 minus2 (ACK is not received) minus2 (ACK is not received)Node (5) iteration 4 minus1 (ACK is not received) minus1 (ACK is not received)Node (5) iteration 5 minus1 (ACK is not received) minus1 (ACK is not received)

Thus the same situation can be considered for othernodes and we can follow the same operations as mentionedin Tables 6 7 8 and 9

Calculating the cooperation rate of node (4) in relation tonodes (2) and (6)

CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)

Concerning the cooperation rate of node (5) which isconsidered in this example as malicious node we supposethat this node and other nodes will exchange the HELLO andTC messages during iterations (1) and (2) Additionally theCR of node (5) which is initialized by zero will be updated byadding 1 in the first iteration and by 2 in the second However

we suppose that node (5) chooses not-cooperate with othernodes from iteration 3 (to save its energy) it means thatthis node will not send HELLO and TC messages Thereforewhen the other nodes do not receive thesemessages (it meansthat the ACK is not received) from node (5) then theywill start by subtracting the last value added which is 2 initeration 3 and by subtracting 1 in iteration 4 Additionallywhen the CR is equal to zero it will be updated by subtracting1 and so on

Calculating of the cooperation rate of node (3) in relationto nodes (1) and (5)

CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1

then CR (5) = CR(53) + CR(56) = minus2

(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)

Thus when the cooperation rate of node (5) will beshared all nodes must detect that (CR(5) lt 0) according tothreshold proposed in this paper the nodes will handle thisnode as a noncooperative nodeTherefore the utility function119865 of node (1) in relation to the path (1 2 4 and 6) is

1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)

The utility function 119865 of the node (1) in relation to the path(1 3 5 and 6) is

1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)

Therefore the source node (1) will choose the first path withthe greater value of its utility function 119865 in order to sendpackets to its destination which is node (6)7 Simulation Environment

Our proposal is evaluated using Network Simulator 3 (NS-317) [41] that contains the OLSR module In this work weimplement our algorithm and compare it with the original

Table 10 Simulation parameters

Parameter ValueRouting protocol OLSRSimulation time 300 secondsNumber of nodes 20 40 60 80 and 100

Number of selfish nodes 50 of the number of nodes inselfish OLSR

Environment area 1000 meters times 1000 metersThe transmit power 75 dBmMAC protocol IEEE 80211Transport layer User Datagram Protocol (UDP)Pause time 0 secondsMaximum speeds 20 meterssecondNetwork Simulator NS317Mobility model RandomWayPoint

routing table algorithm described in the standard OLSR Inaddition we compare our proposal with a selfishOLSRwherenodes choose to drop packets rather than forwarding themto their destinations During all simulations the networkcontains a variable number of mobile nodes and movingin a fixed area We used the Random WayPoint (RWP)mobility model with pause time fixed to 0 seconds variousrandom seeds and max speed of 20 meterssecond Thechoice of the simulation parameters can be justified by manyscenarios of MANET applications such as military fieldsand conferencing Additionally and concerning the mobilityRWP seems to be an arduous environment to evaluate theeffectiveness of our proposal Moreover our proposedmodeloriginal OLSR and selfish OLSR protocols are evaluatedbased on the same mobility scenarios that define the nodesmovement Furthermore in this experimental study weconducted exhaustive simulations and Table 10 recapitulatesall the simulation parameters

71 Performance Metrics The main objective of the exper-iments using Network Simulator 3 (317) is to evaluateand validate our proposal by analyzing and addressing thefollowing performance metrics

(i) Energy is themetric used to quantify and evaluate thelifetime of nodes and network

(ii) Throughput is the number of messages successfullydelivered per time unit


10 20 30 40 50 60 70 100Number of nodes

Residual energy (joule)

Original OLSRSelfish OLSR

830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)

Figure 8 The residual energy in original OLSR and selfish OLSR

(iii) End-to-end delay is the time interval between thetransmission of a packet and its reception

(iv) Total packets forwarded are the total traffic andpackets received and forwarded by nodes across thenetwork

(v) Packets received are the successful packets transmit-ted to their destination

8 Analytical Results

In this sectionwe are going to compare between three variantsof protocol original OLSR enhanced OLSR and selfishOLSR

Figure 8 shows the evolution of the residual energy inrelation to the number of nodes We can observe the impactof selfish behavior on energy consumption and the differencebetween original OLSR and selfish OLSR protocols Further-more we notice that malicious nodes are able to save energywhen they refuse to cooperate for routing packets fromother nodes because these operations require most energyconsumption Therefore the rational nodes need to do morework to compensate the job of selfish nodes and then spendmore energy to complete this task

In Figure 9 we observe the evolution of throughput asfunction of the number of nodes in different variants ofOLSR It is evident that the throughput in case of originalOLSR is high compared to the selfish OLSR We interpret theresults by the existence ofmalicious nodes that choose to droppackets rather than forwarding them to their destinationsTherefore this behavior can affect the throughput by theretransmission of the lost packets by rational nodes Inanother observation we notice that the throughput in case ofenhanced OLSR is high compared to the selfish OLSR Thisimprovement can be justified because the packets discardedby the malicious nodes are decreased in enhanced OLSRusing malicious detection algorithm Furthermore throughour proposal we can get almost the same performances and

20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes

Average throughput (kbits)

Enhanced OLSRSelfish OLSR

Original OLSR

0

20

40

60

80

100

120

Figure 9 The average of throughput in enhanced OLSR originalOLSR and selfish OLSR

mitigate the aggregate effect in case of existence of maliciousnodes compared to the original protocol

In Figure 10 we observe the evolution of end-to-enddelay (ETED) in relation to the number of nodes in thethree variants of OLSR In addition we notice the impactof selfish nodes on ETED compared to the original andenhanced OLSR The results concerning ETED in originaland enhanced OLSR can be justified because the numberof nodes that participate in the routing process is increasedMoreover in OLSR routing protocol the ETED depends onthe routing process and the number of nodes involved Onthe contrary in the selfish OLSR and in our situation we areinterested only in ETED of packets which are successfullytransmitted Additionally most of the packets cannot reachtheir destinations due to the selfish nodes that choose to dropany packets that pass on instead of forwarding them to theirdestinations Therefore and owing to the large number ofselfish nodes the ETED must be less than the original andenhanced OLSR On the other hand our proposal can offernearly the same performance compared to the original OLSRFurthermore and due to the impact of some packet collisionnoise transmission and the processing time that is neededto calculate CR our solution provides an ETED which is lesseffective than the original OLSR

In Figure 11 we observe the evolution of the totalpackets forwarded (TPF) as function of the number of nodesin original OLSR selfish OLSR and enhanced OLSR Wenotice that the TPF is high in original and enhanced OLSRcompared to selfish OLSR Moreover we interpret this resultby the existence of malicious nodes that attempt to reducenetwork connectivity and undermine the network securityIn addition the impact of selfish behavior is due to themalicious nodes that choose to drop packets received insteadof forwarding them to their destinations Therefore the TPFmust be high in original and enhancedOLSR usingmaliciousnode detection mechanism Furthermore we interpret the


20 40 60 80 100Number of nodes

Average end-to-end delay (ms)


Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay

Figure 10 The average of end-to-end delay in enhanced OLSRoriginal OLSR and selfish OLSR


Average total packets forwarded


Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d

Figure 11 The average of packets forwarded in enhanced OLSRoriginal OLSR and selfish OLSR

difference between original OLSR and enhanced OLSR dueto the impact of some packet collision and noise transmissionduring the calculation of CR

In Figure 12 we observe the evolution of packets receivedin relation to the number of nodes concerning the threevariants of OLSR From this figure we notice that the numberof packets received is high in original OLSR and enhancedOLSR compared to selfish OLSR This result is due to themalicious nodes that choose to drop packets received ratherthan forwarding them to their destinations Therefore thisbehavior should influence the number of received packets Inaddition the result in this figure shows the effectiveness ofour proposal usingmalicious node detectionmechanism and


Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved

Figure 12 The average of packets received in enhanced OLSRoriginal OLSR and selfish OLSR

reinforces the result mentioned above concerning the totalpackets forwarded in Figure 11

9 Conclusion and Future Work

In this paper we have proposed a new idea based on a gametheoretic approach to enhance OLSR security mechanism inMANETs This proposal can be used to model interactionsbetween selfish nodes and a large number of legitimate nodesinside the network Contrary to some existing research onsecurity in MANETs that rely on the game theory the pro-posed solution can enable each node to evaluate behaviors ofother nodes Furthermore the rational nodes can intelligentlychoose their strategies to deal with selfish behaviorwhen eachnode keeps track of the cooperation rate (CR) record of othernodes Moreover many parameters (throughput end-to-enddelay total packets forwarded and packets received) canbe improved significantly and the aggregate effect of selfishnodes can be reduced as well In addition the simulationresults have shown that our proposed solution scheme takesinto account in addition to the security requirements thesystem resources Furthermore in this paper we have provedthat our proposal can be used as a security mechanism inorder to enforce the cooperation between nodes improvenetwork performances and prevent malicious nodes How-ever the comparison with other game theory models canenhance the efficiency of this proposed solution Thereforeand as future works

(i) we plan to study and address other game modelsespecially those treating OLSR routing protocol inorder to compare our proposed solution with thesemodels

(ii) additionally the cooperation rate is not the uniqueparameter to evaluate node behavior For this and asfuture work we plan to improve thismodel to support


other parameters like the overload of the path andthe energy consumption of the nodes constituting thepath Moreover and in order to optimize the energyconsumption we will change the interval 119905 needs tocalculate the cooperation rate Therefore instead ofusing HELLO interval which is 2 seconds we willuse 6 seconds that is after receiving three HELLOmessages and instead of using TC interval which is 5seconds we will use 10 seconds that is after receivingtwo TC messages

Competing Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

References

[1] B Kannhavong H Nakayama N Kato A Jamalipour and YNemoto ldquoA study of a routing attack in OLSR-based mobile adhoc networksrdquo International Journal of Communication Systemsvol 20 no 11 pp 1245ndash1261 2007

[2] EM Shakshuki N Kang and T R Sheltami ldquoEAACKA secureintrusion-detection system forMANETsrdquo IEEE Transactions onIndustrial Electronics vol 60 no 3 pp 1089ndash1098 2013

[3] S Djahel F Nait-Abdesselam and Z H Zhang ldquoMitigatingpacket dropping problem inmobile ad hoc networks proposalsand challengesrdquo IEEE Communications Surveys amp Tutorials vol13 no 4 pp 658ndash672 2011

[4] A Nadeem and M P Howarth ldquoA survey of manet intrusiondetection amp prevention approaches for network layer attacksrdquoIEEE Communications Surveys and Tutorials vol 15 no 4 pp2027ndash2045 2013

[5] N Lal K Shishupal S Aditya andV K Chaurasiya ldquoDetectionof malicious node behaviour via I-watchdog protocol in mobileAd Hoc network with DSDV routing schemerdquo Procedia Com-puter Science vol 49 pp 264ndash273 2015

[6] T Clausen and P Jacquet ldquoOptimized Link State Routing Pro-tocol (OLSR)rdquo IETF RFC 3626 2003

[7] F Wu T Chen S Zhong C Qiao and G Chen ldquoA game-theoretic approach to stimulate cooperation for probabilisticrouting in opportunistic networksrdquo IEEE Transactions onWire-less Communications vol 12 no 4 pp 1573ndash1583 2013

[8] E Karami and S Glisic ldquoJoint optimization of scheduling androuting in multicast wireless ad hoc networks using soft graphcoloring and nonlinear cubic gamesrdquo IEEE Transactions onVehicular Technology vol 60 no 7 pp 3350ndash3360 2011

[9] J Liu X Jiang H Nishiyama R Miura N Kato and N Kad-owaki ldquoOptimal forwarding games in mobile Ad Hoc networkswith two-hop f-cast relayrdquo IEEE Journal on Selected Areas inCommunications vol 30 no 11 pp 2169ndash2179 2012

[10] P Michiardi and R Molva ldquoCore a collaborative reputationmechanism to enforce node cooperation in mobile ad hoc net-worksrdquo in Advanced Communications and Multimedia SecurityJ B Blazic and T Klobucar Eds pp 107ndash121 Springer NewYork NY USA 2002

[11] M Mejia N Pena J L Munoz O Esparza and M A AlzateldquoA game theoretic trust model for on-line distributed evolutionof cooperation inMANETsrdquo Journal of Network and ComputerApplications vol 34 no 1 pp 39ndash51 2011

[12] M A Elfaki H Ibrahim A Mamat M Othman and HSafa ldquoCollaborative caching priority for processing requests inMANETsrdquo Journal of Network and Computer Applications vol40 no 1 pp 85ndash96 2014

[13] Y Ben Saied A Olivereau D Zeghlache and M Laurent ldquoAsurvey of collaborative services and security-related issues inmodern wireless Ad-Hoc communicationsrdquo Journal of Networkand Computer Applications vol 45 pp 215ndash227 2014

[14] H Amraoui A Habbani and A Hajami ldquoCCS a correct coop-eration strategy based on game theory for MANETSrdquo inProceedings of the IEEEACS 11th International Conference onComputer Systems and Applications (AICCSA rsquo14) pp 326ndash3322014

[15] P Michiardi and R Molva ldquoAnalysis of coalition formationand cooperation strategies in mobile ad hoc networksrdquo Ad HocNetworks vol 3 no 2 pp 193ndash219 2005

[16] S Zhong and F Wu ldquoA collusion-resistant routing schemefor noncooperative wireless Ad Hoc networksrdquo IEEEACMTransactions on Networking vol 18 no 2 pp 582ndash595 2010

[17] Z Li and H Shen ldquoGame-theoretic analysis of cooperation in-centive strategies inmobile ad hoc networksrdquo IEEETransactionson Mobile Computing vol 11 no 8 pp 1287ndash1303 2012

[18] Y Wang F R Yu H Tang and M Huang ldquoA mean field gametheoretic approach for security enhancements in mobile ad hocnetworksrdquo IEEE Transactions on Wireless Communications vol13 no 3 pp 1616ndash1627 2014

[19] S-K Ng and W K G Seah ldquoGame-theoretic approach forimproving cooperation in wireless multihop networksrdquo IEEETransactions on Systems Man and Cybernetics Part B Cyber-netics vol 40 no 3 pp 559ndash574 2010

[20] T Chen F Wu and S Zhong ldquoFITS a finite-time reputationsystem for cooperation in wireless ad hoc networksrdquo IEEETransactions on Computers vol 60 no 7 pp 1045ndash1056 2011

[21] M Kaliappan and B Paramasivan ldquoEnhancing secure routingin mobile ad hoc networks using a dynamic bayesian signallinggame modelrdquo Computers and Electrical Engineering vol 41 pp301ndash313 2015

[22] B Paramasivan M J V Prakash and M Kaliappan ldquoDevel-opment of a secure routing protocol using game theory modelin mobile ad hoc networksrdquo Journal of Communications andNetworks vol 17 no 1 pp 75ndash83 2015

[23] Y Wu S Tang P Xu and X-Y Li ldquoDealing with selfishnessand moral hazard in noncooperative wireless networksrdquo IEEETransactions on Mobile Computing vol 9 no 3 pp 420ndash4342010

[24] I Khalil and S Bagchi ldquoStealthy attacks in wireless ad hoc net-works detection and countermeasurerdquo IEEE Transactions onMobile Computing vol 10 no 8 pp 1096ndash1112 2011

[25] W Wang H Man and Y Liu ldquoA framework for intrusiondetection systems by social network analysis methods in ad hocnetworksrdquo Security and Communication Networks vol 2 no 6pp 669ndash685 2009

[26] T P Gondaliya and M Singh ldquoIntrusion detection system onMAC layer for attack prevention in MANETrdquo in Proceedingsof the 4th International Conference on Computing Communi-cations and Networking Technologies (ICCCNT rsquo13) pp 1ndash5Tiruchengode India July 2013

[27] A Nadeem andM P Howarth ldquoAn intrusion detectionamp adap-tive response mechanism for MANETsrdquo Ad Hoc Networks vol13 pp 368ndash380 2014


[28] K R Abirami M G Sumithra and J Rajasekaran ldquoAnenhanced intrusion detection system for routing attacks inMANETrdquo in Proceedings of the International Conference onAdvanced Computing and Communication Systems (ICACCSrsquo13) pp 1ndash6 Coimbatore India December 2013

[29] K Wang and H Guo ldquoAn improved routing algorithm basedon social link awareness in delay tolerant networksrdquo WirelessPersonal Communications vol 75 no 1 pp 397ndash414 2014

[30] K Wang Z Ouyang R Krishnan L Shu and L He ldquoA gametheory-based energy management system using price elasticityfor smart gridsrdquo IEEE Transactions on Industrial Informaticsvol 11 no 6 pp 1607ndash1616 2015

[31] K Wang and M Wu ldquoNash equilibrium of node cooperationbased on metamodel for MANETsrdquo Journal of InformationScience and Engineering vol 28 no 2 pp 317ndash333 2012

[32] K Wang and M Wu ldquoCooperative communications basedon trust model for mobile ad hoc networksrdquo IET InformationSecurity vol 4 no 2 pp 68ndash79 2010

[33] P K Dutta Strategies and Games Theory and Practice MITPress 2001

[34] M Le Treust and S Lasaulce ldquoA repeated game formulationof energy-efficient decentralized power controlrdquo IEEE Transac-tions on Wireless Communications vol 9 no 9 pp 2860ndash28692010

[35] Y Xiao J Park and M Van Der Schaar ldquoRepeated games withintervention theory and applications in communicationsrdquoIEEE Transactions on Communications vol 60 no 10 pp 3123ndash3132 2012

[36] S lasaulce and H Tembine Game Theory and Learning forWireless Networks Fundamentals and Applications AcademicPress New York NY USA 2011

[37] A Diekmann ldquoVolunteerrsquos dilemmardquo Journal of Conflict Reso-lution vol 29 no 4 pp 605ndash610 1985

[38] A Diekmann ldquoCooperation in an asymmetric volunteerrsquosdilemma game theory and experimental evidencerdquo Interna-tional Journal of GameTheory vol 22 no 1 pp 75ndash85 1993

[39] W L Eddie and H L Cheng ldquoAnalytic hierarchy process anapproach to determine measures for business performancerdquoMeasuring Business Excellence vol 5 no 3 pp 30ndash37 2001

[40] T L Saaty The Analytic Hierarchy Process McGraw-Hill NewYork NY USA 1980

[41] ldquoThe network simulator NS-3rdquo httpswwwnsnamorg

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Distributed Sensor Networks


Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014


ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014


Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications


Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia


Biomedical Imaging


ArtificialNeural Systems

Advances in


RoboticsJournal of



Computational Intelligence and Neuroscience

Industrial EngineeringJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in



is based on the absorbing of significant amount of trafficdropping received packets and not cooperating during thepacket routing process However this problem arises whenexamining network topology where malicious nodes cannotbe easily detected

MANETs are infrastructureless and self-organized allnodes have to cooperate between themselves in order toprovide the best performances and offer necessary networkfunctionalities The cooperation mechanism must complywith the rules imposed by the routing protocol for trans-mitting and receiving data On the other hand the non-cooperation behavior can be produced by nodes that donot follow these rules However all nodes act as routers orgateways and contribute to discovering and maintaining therouting process Moreover each node is constrained in termsof limited resources (energy etc) and it may not always beinteresting to accept relay requests that require consumptionof most resources Therefore a cooperative system shouldbe integrated and incorporated with any network operationssuch as packet forwarding and route discovery The goal ofthis system is to prevent malicious behavior and encouragenodes to cooperate with each other However the applicationof cooperation mechanism is particularly difficult because ofMANET features that impose certain requirements

In this context many researchers have investigated selfishnodes and security problems in MANETs In this way theauthors in [1] proposed a solution to improve network per-formances when attacks are launched and mitigate aggregateeffect especially that all nodes in such networks are vulner-able to being isolated by malicious nodes Furthermore theauthors in [2] presented a powerful intrusion detection sys-tem called Enhanced Adaptive ACKnowledgment (EAACK)Compared to other intrusion detectionmechanisms EAACKshows an efficient attack detection without affecting networkperformances Likewise the authors in [3] surveyed theimpact of packet dropping attacks in MANET In their workthe authors try to demonstrate the importance of attackselaborate a new detection system and avoid maliciousnodes during communication with each other AdditionallyMANET nature makes it more attractive to many types ofattacks In this way the authors in [4] proposed a surveyin two parts the first one addresses important securitymechanisms and types of attacks that can affect networkperformances especially in the network layerThe second oneaddresses the classification of detectionmechanisms that dealwith a single or series of attacks Furthermore in the workpresented in [5] the authors suggested a powerful solutioncalled I-Watchdog protocol to detect malicious nodes inMANETs In addition the authors proved through simu-lations the effectiveness of this solution with Destination-Sequenced Distance-Vector (DSDV) routing protocol interms of packet drop ratio (PDR) throughput and end-to-end delay

Recently game theory provides a useful solution formodeling and addressing different problems in MANETs Insuch networks players (nodes) have conflicting objectivesand different profiles with each other Additionally in thegame theory a utility function represents a payoff (reward)that allows each player to evaluate a particular outcome that

reflects its objectives The utility of each player depends notonly on its actions (strategies) but also on othersrsquo actionsIn addition a security scheme must take into account pastand current strategies of different players to be successful inMANETs

In the same context our research is focused onmobile adhoc networks using Optimized Link State Routing Protocol(OLSR) which is a proactive protocol In such networksOLSR is one of the most used routing protocols Moreoverthe cooperation concept is an essential element that can resultin the evolution of network performances in MANET In thisway and in this paper the cooperation rate (CR) representsthe value which indicates howmany times a node cooperatesor not during the game (or during network lifetime)Throughthis value each node can evaluate the behavior of anothernode before sending a packet Moreover in this paper athreshold is considered as the minimum value of the CRaccepted by all rational nodes We consider that each nodewhich has aCRgt 0 is considered as a legitimate nodewhereasa node with CR lt 0 is considered as a noncooperative nodeTherefore our contribution is briefly summarized below

(i) Firstly our conduct is to put forward an enhancedalgorithm based on game theory and establishingconfident relationships between nodes In this pro-posed model each node keeps a cooperation rate(CR) record of other nodes to evaluate their behaviorsand avoid malicious nodes

(ii) Secondly the calculation of CR is based on OLSRmessages (HELLO and topology control (TC))exchanged between nodes and forwarding processes

(iii) Thirdly the cooperation rate will be shared betweennodes in addition to other network information usingHELLO and TC messages

(iv) Fourthly the key novelty of this paper is that the valueof CR will be used as a metric to construct routingtables instead of hop count metric used by OLSRstandard

The remainder of this paper is organized as follows TheOLSR routing protocol as a proactive scheme is presentedin Section 2 Some previous studies that aim at addressingcooperation and selfish behavior in MANETs are presentedin Section 3 A game model formulation is described inSection 4 Our suggested systemmodel based on cooperationbetween nodes is discussed in Section 5 A malicious detec-tion algorithm and enhanced routing table computation areintroduced in Section 6 A simulation environment used toaddress our approach is discussed in Section 7 The resultsthat concern the validation of our solution are presented inSection 8 Finally the paper is concluded in Section 9 withfuture work

2 Proactive Schema Case of OLSR

OLSR is a proactive routing protocol [6] based on MPR(multipoint relay) mechanism that is considered as the keyconcept used in this protocolTheMPRs are used tomaintain


N3

N8

N10

N7

N0N4

N2

N1

N6

N12

N5

N9

N11


One-hop neighbors

Two-hop neighbors


Malicious message

Malicious node




3 State of the Art
















Players Nodes



Utilityfunction



(player)













































(119881 minus 119890) + 119903 = 119901 times (119881 + 119903) (6)

Therefore

119901 = ((119881 minus 119890) + 119903)(119881 + 119903) (7)


119901lowast = 1 minus ( 119890(119881 + 119903)) (8)



lim119903rarrinfin

119901lowast = 1 (9)


119901lowast = 119902lowast (10)












119880119894 (C) = (11) + (12) (13)

So


(14)


119880119894 (C) = (119881 minus 119890) (15)









119880119894 (NC) = (16) + (17) (18)

So


(19)






(1 minus 119901)119899minus1 = 119890(119881 + 119903) (22)

So

(1 minus 119901) = ( 119890(119881 + 119903))

(1(119899minus1))

(23)


Therefore

119901 = 1 minus ( 119890119881 + 119903)

1(119899minus1) (24)

So



119901lowast = 119902lowast (26)


lim119899rarrinfin

119901lowast = 0 (27)



CR = sum(ℎ (119905) + 120573 (119905)) (28)


5 System Model








CR (119894 119905 119865) = sum(ℎ (119905) + 120573 (119905)) (29)






0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9




120573 (119905)



(30)








(3) Select options



CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)









0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in




N3

N8

N10

N7

N0N4

N2

N1

N6

N12

N5

N9

N11


One-hop neighbors

Two-hop neighbors


Malicious message

Malicious node




3 State of the Art
















Players Nodes



Utilityfunction



(player)













































(119881 minus 119890) + 119903 = 119901 times (119881 + 119903) (6)

Therefore

119901 = ((119881 minus 119890) + 119903)(119881 + 119903) (7)


119901lowast = 1 minus ( 119890(119881 + 119903)) (8)



lim119903rarrinfin

119901lowast = 1 (9)


119901lowast = 119902lowast (10)












119880119894 (C) = (11) + (12) (13)

So


(14)


119880119894 (C) = (119881 minus 119890) (15)









119880119894 (NC) = (16) + (17) (18)

So


(19)






(1 minus 119901)119899minus1 = 119890(119881 + 119903) (22)

So

(1 minus 119901) = ( 119890(119881 + 119903))

(1(119899minus1))

(23)


Therefore

119901 = 1 minus ( 119890119881 + 119903)

1(119899minus1) (24)

So



119901lowast = 119902lowast (26)


lim119899rarrinfin

119901lowast = 0 (27)



CR = sum(ℎ (119905) + 120573 (119905)) (28)


5 System Model








CR (119894 119905 119865) = sum(ℎ (119905) + 120573 (119905)) (29)






0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9




120573 (119905)



(30)








(3) Select options



CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)









0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in














Players Nodes



Utilityfunction



(player)













































(119881 minus 119890) + 119903 = 119901 times (119881 + 119903) (6)

Therefore

119901 = ((119881 minus 119890) + 119903)(119881 + 119903) (7)


119901lowast = 1 minus ( 119890(119881 + 119903)) (8)



lim119903rarrinfin

119901lowast = 1 (9)


119901lowast = 119902lowast (10)












119880119894 (C) = (11) + (12) (13)

So


(14)


119880119894 (C) = (119881 minus 119890) (15)









119880119894 (NC) = (16) + (17) (18)

So


(19)






(1 minus 119901)119899minus1 = 119890(119881 + 119903) (22)

So

(1 minus 119901) = ( 119890(119881 + 119903))

(1(119899minus1))

(23)


Therefore

119901 = 1 minus ( 119890119881 + 119903)

1(119899minus1) (24)

So



119901lowast = 119902lowast (26)


lim119899rarrinfin

119901lowast = 0 (27)



CR = sum(ℎ (119905) + 120573 (119905)) (28)


5 System Model








CR (119894 119905 119865) = sum(ℎ (119905) + 120573 (119905)) (29)






0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9




120573 (119905)



(30)








(3) Select options



CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)









0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in






Players Nodes



Utilityfunction



(player)













































(119881 minus 119890) + 119903 = 119901 times (119881 + 119903) (6)

Therefore

119901 = ((119881 minus 119890) + 119903)(119881 + 119903) (7)


119901lowast = 1 minus ( 119890(119881 + 119903)) (8)



lim119903rarrinfin

119901lowast = 1 (9)


119901lowast = 119902lowast (10)












119880119894 (C) = (11) + (12) (13)

So


(14)


119880119894 (C) = (119881 minus 119890) (15)









119880119894 (NC) = (16) + (17) (18)

So


(19)






(1 minus 119901)119899minus1 = 119890(119881 + 119903) (22)

So

(1 minus 119901) = ( 119890(119881 + 119903))

(1(119899minus1))

(23)


Therefore

119901 = 1 minus ( 119890119881 + 119903)

1(119899minus1) (24)

So



119901lowast = 119902lowast (26)


lim119899rarrinfin

119901lowast = 0 (27)



CR = sum(ℎ (119905) + 120573 (119905)) (28)


5 System Model








CR (119894 119905 119865) = sum(ℎ (119905) + 120573 (119905)) (29)






0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9




120573 (119905)



(30)








(3) Select options



CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)









0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in

































(119881 minus 119890) + 119903 = 119901 times (119881 + 119903) (6)

Therefore

119901 = ((119881 minus 119890) + 119903)(119881 + 119903) (7)


119901lowast = 1 minus ( 119890(119881 + 119903)) (8)



lim119903rarrinfin

119901lowast = 1 (9)


119901lowast = 119902lowast (10)












119880119894 (C) = (11) + (12) (13)

So


(14)


119880119894 (C) = (119881 minus 119890) (15)









119880119894 (NC) = (16) + (17) (18)

So


(19)






(1 minus 119901)119899minus1 = 119890(119881 + 119903) (22)

So

(1 minus 119901) = ( 119890(119881 + 119903))

(1(119899minus1))

(23)


Therefore

119901 = 1 minus ( 119890119881 + 119903)

1(119899minus1) (24)

So



119901lowast = 119902lowast (26)


lim119899rarrinfin

119901lowast = 0 (27)



CR = sum(ℎ (119905) + 120573 (119905)) (28)


5 System Model








CR (119894 119905 119865) = sum(ℎ (119905) + 120573 (119905)) (29)






0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9




120573 (119905)



(30)








(3) Select options



CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)









0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in







(119881 minus 119890) + 119903 = 119901 times (119881 + 119903) (6)

Therefore

119901 = ((119881 minus 119890) + 119903)(119881 + 119903) (7)


119901lowast = 1 minus ( 119890(119881 + 119903)) (8)



lim119903rarrinfin

119901lowast = 1 (9)


119901lowast = 119902lowast (10)












119880119894 (C) = (11) + (12) (13)

So


(14)


119880119894 (C) = (119881 minus 119890) (15)









119880119894 (NC) = (16) + (17) (18)

So


(19)






(1 minus 119901)119899minus1 = 119890(119881 + 119903) (22)

So

(1 minus 119901) = ( 119890(119881 + 119903))

(1(119899minus1))

(23)


Therefore

119901 = 1 minus ( 119890119881 + 119903)

1(119899minus1) (24)

So



119901lowast = 119902lowast (26)


lim119899rarrinfin

119901lowast = 0 (27)



CR = sum(ℎ (119905) + 120573 (119905)) (28)


5 System Model








CR (119894 119905 119865) = sum(ℎ (119905) + 120573 (119905)) (29)






0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9




120573 (119905)



(30)








(3) Select options



CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)









0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in




Therefore

119901 = 1 minus ( 119890119881 + 119903)

1(119899minus1) (24)

So



119901lowast = 119902lowast (26)


lim119899rarrinfin

119901lowast = 0 (27)



CR = sum(ℎ (119905) + 120573 (119905)) (28)


5 System Model








CR (119894 119905 119865) = sum(ℎ (119905) + 120573 (119905)) (29)






0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9




120573 (119905)



(30)








(3) Select options



CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)









0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in








0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9




120573 (119905)



(30)








(3) Select options



CR (119894 119905 119865) = 119882 times sum(ℎ (119905) + 120573 (119905)) (31)









0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in








0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9


ANSN Reserved




0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 12 3 4 5 6 7 8 9







information








119865119899 (119894 119895) =119872sum119897=1

CR(119897) (32)




BEGIN



Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in




Malicious node















Source Destination

1

53

2 4

6




CR(21) = 15

CR(24) = 15

then CR (2) = CR(21) + CR(24) = 30

(33)












CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in














CR(42) = 15

CR(46) = 15

then CR (4) = CR(42) + CR(46) = 30

(34)




CR(31) = 15

CR(35) = minus1

then CR (3) = CR(31) + CR(35) = 14

(35)





CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in







CR(53) = minus1

CR(56) = minus1


(36)


CR(64) = 15

CR(65) = minus1

then CR (6) = CR(64) + CR(65) = 14

(37)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (2) + CR (4) = 30 + 30

= 60(38)


1198651 (1 6) =119872sum119897=1

CR(2) = CR (3) + CR (5) = 14 + (minus2)

= 12(39)












10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in




10 20 30 40 50 60 70 100Number of nodes



830

840

850

860

870

880

890

900

910

Resid

ual e

nerg

y (jo

ule)









20 40 60 80 100

Aver

age t

hrou

ghpu

t (kb

its)

Number of nodes



Original OLSR

0

20

40

60

80

100

120









Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in







Original OLSR

003

303

603

903

1203

1503

1803

2103

Aver

age e

nd-to

-end

del

ay





Original OLSR

0

100000

200000

300000

400000

500000

600000

Aver

age t

otal

pac

kets

forw

arde

d





Packets received


Original OLSR

0

50000

100000

150000

200000

250000

300000

350000

400000

Pack

ets r

ecei

ved









Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in





Competing Interests


References


















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in

























Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in










Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in



Documents

Research Article Security-Based Mechanism for …downloads.hindawi.com/journals/misy/2016/5653010.pdfSecurity-Based Mechanism for Proactive Routing Schema ... many problems in mobile