ISSN: 1748-0345 (Online) www.tagajournal.

com

ENHANCED UNDERWATER WIRELESS

SENSOR NETWORKS SECURITY
WITH TIC-TAC-TOE AI-MINIMAX
ALGORITHM IN GAME THEORY.
Porkodi K.P.1*, A.M.J.Md.ZubairRahman2
1, 2
Department of Computer Science and Engineering, Al-Ameen Engineering college, Erode,
Tamilnadu, India.638 104.
*CorrespondingAuthorEmail:porkodiprabhakaran@gmail.com, alameenengg@yahoo.co.in

Abstract:

In the current era, the world is covered with 80% of water.so the communication in
underwater wireless sensor networks is essential for human to avoid electromagnetic
fields.In underwater wireless sensor networks security is the major concern to do the
communication in secure way. To improve the security and authentication in underwater
wireless sensor networks this paper approaches the game theory with tic-tac-toe AI-
MINIMAX Algorthim. In this paper the tic-tac-toe AI-MINIMAX is proposed to
implement the communication in a secure manner.

Index Terms—Under Water Wireless sensor network, Tic-Tac-Toe ,Min-Max algorithm

1. Introduction
The trend towards underwater wireless sensor networks communication plays a vital role in
research area. In the UWSN communication improving security and validation is more
important challenging task since UWSN has its own characteristics of high propagation delay
,low attenuation, low battery power, memory etc. As compared with traditional and WSN
networks the applications of UWSN is more useful in military service, health care management
etc .In UWSN nodes are dynamic and the stability of node is also very less as compare with
WSN, which leads to change the arrangement of network topology change, In addition to that in
UWSN there is no constant or predefine route for validation this because of unambiguous nodes.
The characteristics of unambiguous node, limitation of power, energy bandwidth makes the
node to behave independently that increases the malicious attacks and the risk of network
malfunctioning. Hence, these above facets make the UWSN makes the research to be more
concentrated in security schemes. To overcome the UWSN security threads many security
schemes, reputation-based techniques and principles were focused and developed.[reference
have to add]. Due to the complexity of implementation this paper will focus on Tic-Tac-Toe
game-theoretic-based protection techniques for UWSN network security.
Game theory is a very old and golden approach technique and it is implemented in various
applications. In [2], the game theory features for wireless sensor networks has been implemented
and proved. The various trends of game theory WSN has been reviewed[3,4].it tackles the

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 216 TAGA JOURNAL VOL. 14

ISSN: 1748-0345 (Online) www.tagajournal.com

problems of optimization and cost functions[1].The game theory approach is very useful in
decentralized communications[5].the decentralized communication is well suited in UWSN.
Since the nodes are dynamic and stability is very low. In the decentralized communication can
leads the highest chance of defender(s) and attacker(s) which is the main draw back in UWSNS
related to security. To overcome the above approach the game theory approach is very useful in
decentralized communications [5]. The approach mainly focuses on the interface between the
defender(s) and attacker(s) depends on it counter action. The counter-action of the other party
places game theory as a perfect fit for this security model in UWSN [6].
In this paper, we explained a different game technique which is utilized in UWSN. In
addition, an overall analysis of the security fulfillment of UWSN is also represented. The
literature topic will explained about the different types of game theory and possible techniques to
overcome the mitigation of UWSN. The different types of game theory will be based on number
of possible resources, symmetry of the game, and cooperation among nodes. Then we
implemented the min-max tic-tac toe approach for the concern about the mitigation of threads,
reduction of substance attacks, Security,etc.In view of that, we proved that the utilization of this
model is very useful in any network environment and it protect from all the internal and external
attack.At last, we present some suitable future trends and intelligent attacks [7,8,9,10] via the
evolutionary game approach [11,12,13]. The applications of wireless underground sensor
networks include acoustic signals such as Earthquake disaster observation, underwater sensor
networks (UWSNs), agriculture applications, military uses etc..has to be more concern about the
security. [23,24 ,25].
The remaining section of the paper is structured as like below mentioned topic. Chapter1
contains about the game theory description and the classification which is used for mitigation of
threats in UWSN and the import concern of security in UWSN.Chapter 2 will elaborate about the
proposed Tic-Tac-Toe Ai-Minimax Algorithm In Gametheory and their pseducode to implement
the data.Chapter 3 will talk about the application future .chapter 4 includes conclusion and future
recommendations.
CHAPTER1: GAME THEORY: OVERVIEW
Game theory is an advanced limb of intelligent rational-decision makers [19,20]. It may
applicable in wide variety of behavioral relations. It provides a better path for the human conflict
and the cooperation within a competitive situation. The concept of game theory is more useful in
interactive situation in which the independent actors(nodes) share more or less formal rules and
consequences.
Game theory is used to learn about different types of mathematical models, individual
cooperation and their conflicts. The research may involves among individual or group of
peoples. The utility payoff is the positive or negative reward to a player for the given action
denoted by Ui:A->R,which measures the outcome for players I determined by the actions which
is denoted by Cartesian product. In the focus of underwater[14].Wireless sensor networks
security based on game theory approach the min-max is best approach for finding the path and it
also more concentrated in security.es of g
1.1 Different types of game theory in UWSNs:

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 217 TAGA JOURNAL VOL. 14

ISSN: 1748-0345 (Online) www.tagajournal.com

The following are different types of games which are utilized for UWSN based on the players
involved in a game, symmetry of game and cooperation between nodes. The different types of
game theory are listed below.

Fig1: DIFFERENT TYPES OF GAME THEORY

GAMES

Cooperative and non Normal Form and SIMULTANEOUS

CONSTANT SUM SYMMETRIC
cooperative games EXTENSIVE F ORM GAMES MOVE GAMES
ZERO SUM AND AND
AND SEQUENTIAL
1. NON ZERO SUM ASSYMETRIC
MOVE GAMES
GAMES.

a. COOPERATIVE GAME AND NON COPERATIVE GAMES:

. The cooperative game is used to adapt the situation in which players can adapt the situation
through the negative and agreement between players. Such as bargaining game[16], repeated
game[18],coalition game[19]. The non-cooperative games refers the games in to produce the
maximum profit. In non-cooperative games, a deep analysis is taken much more effort.
Stackelberg Game[17], Jamming Game[18].etc.,
b. Normal form and extensive form games:
The strategy of the form is represented in the tabular form. It helps to identify the Nash
equilibrium. The sharing game methodology is used in extensive form. The extensive form can
be converted to normal form. The extensive form game is described in the form of the decision
tree.

The decision tree is look like as below:

 Nonterminal node:place where an agent chooses an action

 Edge=an available action or move
 Terminal node=a final outcome
c. SIMULTANEOUS MOVE GAMES AND SEQUENTIAL MOVE GAMES

It is the extension of extensive form games. In the simultaneous game the players do not
know the knowledge about the move of other players. Where as in sequential game the players
do not have the strategies of other players (how many options they have to in the opposite side).it
is also represented in tree structure.so it is applicable in UWSN since, nodes are dynamic.
d. Constant sum, zero sums, and non-zero sum games:

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 218 TAGA JOURNAL VOL. 14

ISSN: 1748-0345 (Online) www.tagajournal.com

Constant Sum: It Is the one In Which Sum Of The Outcome Of All The Players Remain
Constant Even If The Outcomes Are Different.
Zero-sum Game THE SUM OF THE OUTCOMES OF ALL THE PLAYERS OUTPUT IS
ZERO.: In [21], it was presented routing secured against external intruders.the utility function as
a combination of the enegry consumption,probality of malicious nodes etc..will help to prevent
malicious attacks. In [20], a Bayesian zero-sum game approach is proposed to confront the
jammers intervention in order to disrupt the acoustic signals of UWSN.

NON-ZERO SUM GAME: it is same like as zero sum. But in that one dummy player is added
with that such as cooperative games are also included in this game.

e. SYMMETRIC AND ASYMMETRIC GAMES:

Strategies adapted by all players are same. It is useful in only in short term period. Because
the player count is also short.in asymmetric game the players count is high and it is used for
long term plan.

3. The important concern in Security of UWSN:

UWSNs have been focused by many researchers because of their features and applications. Due
to these substantial reasons some requirements should be maintained for UWSNs security as
follows [22]:

 Confidentiality: The data should not understand by another node in UWSNs.

 Integrity: The integrity of data should be high it won’t reveal by any other malicious
node in UWSN.
 Authentication: An authentication makes the good relationship between the sender node
and the receiver node.
 Authorization: The authorization makes the sensor node to perform the next operation..
 Availability: The UWSN services must be available whenever the UWSN users need
them.
 Forward and backward secrecy: For securing the data it must need forward and
backward secrecy.

Chapter 2

Introduction of Tic-Tac toe AI-Minimax algorthim in game theory

2.1 Implementation of Minimax Algorithm in Game Theory for UWSN:

The main goal is of game theory is used to produce a optimal move for the playerwhen the
opposite player(attacker) is also in optimum move.it is mostly used in tic-tac –
toe,backgammon,mancala etc.
In Minimax algorthim implementation two players are used .one is called max and another one is
min. The max is the opposite of min.max is getting highest score and the min is getting lowest
score .As implementation with UWSN max is called as sender and the min is called as attacker.

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 219 TAGA JOURNAL VOL. 14

ISSN: 1748-0345 (Online) www.tagajournal.com

For the implementation of max-min algorthim in UWSN were taking a particular range[range
have to add].in the min max algorthim if the sender has reached the next level then it is denotes
with some positive signal. and if the attacker has received then it gets the negative signal.for
every type of game some heuristics are unique.

2.2 Tic-TAc-Toe AI-Minimax Algorithm in Game Theory for UWSN:

The tic-tac-toe AI-minimax algorthim is a prerequisite of minimax algorthim.Then main goal of
the communication is you have send the underwater wireless sensor networks communication in
a secured way without threats.The Tic-Tac-Toe will consider all possible scenarios and the most
optimal move in a secure way.

Fig2: The movement of Tic-Tac-Toe move in underwater wireless sensor networks

The implementation of tic-tac-toe min-max algorthim in game theory was implemented through
the following procedures

a. Finding the Best Move

b. current move better condition

c. GameOver State condition

d. Making our AI smarter

2.3 PSEDUCODE OF Tic-Tac-Toe ALGORTHIM IMPLEMENTATION:

Pseducode for finding the best move:

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 220 TAGA JOURNAL VOL. 14

ISSN: 1748-0345 (Online) www.tagajournal.com

a. Finding the Best Move:

We shall be introducing a new function called findBestMove(). This function evaluates all the
available moves using minimax() and then returns the best move the maximizer can make. The
is as follows :
Find bestmove():
PSEUDOCODE FOR BESTMOVE PROCEDURE:
Step1: let us assume bestmove=NULL
For each move in the nodes in the particular region
Step2: If current move is better than bestmove
Step3: Bestmove=current move
Step4: Return bestmove
2. Algorithm for finding the best move condition with Minimax :

b. To check the current move better condition we take the help of minmax algorthim
To check whether or not the current move is better than the best move we take the help
of minmax() function which will consider all the possible ways the nodes can go and returns the
best value for that move.instead of returning the move it will return some values.
PSEUDOCODE FOR MINMAX:
Step1: minmax(nodearea,depth,ismaxpower)
Stpe2: if current node is a terminal state:
Return value of the board
Step3:if is maxpower
Bestval=-INFINITY
For each move in nodearea
Value=minmax(nodearea,depth+1,false)
Bestval=max(bestval,value)
Return bestval
Else
Bestval=+Infinity
For eachmove in nodearea:
Value=minmax(nodearea,depth+1,true)
Bestval=max(bestval,value)
Return bestval

C. GameOver State condition:

To check the receiver destination with true or false condition the following pseudocode will be
used:

Pseducode:

ismoveleft(nodearea):
For each cell in nodearea
If current cell is empty
Then return true

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 221 TAGA JOURNAL VOL. 14

ISSN: 1748-0345 (Online) www.tagajournal.com

Else return false if current cell is empty:

d.Making our AI smarter :

Final step to make the AI a little bit smarter activities.

It will give authorization information about the node:

The pseudocode for AI smarter activities:

Step1:if max has high value

Return node reached the destination successfully

Step2: elseif min has won

returnLOOSE_SCORE+depth.

CHAPTER 3:

3.1 FUTURE TRENDS:

The game theory most probably based on the centralized authentication. It is used to handle the
protection of malicious attacks, selfish nodes using the cooperative nodes. As compare with
benefits of centralized attacks the disadvantages is very high such as resource deterioration
however, as there are many benefits of using the centralized authentication, there are also some
problems that may involve the risk of the network security such as the resources deterioration of
the centralized security node. The applicable centralized technique can be overtaken by group
authentication [31] technique. Among the different types of game theory implementation of AI
with min max algorithm will produce the good result. Consequently, we predict that AI models
will be used for the mitigation of the intelligent attacks in UWSNs leading to robust systems.

3.2 CONCLUSION:

This paper mostly focused on the important and challenging problem of assuring security and
best path routing in secured way in UWSNS. In this way, we have carried out an extensive
overview of the state-of-the-art of many game theoretic approaches that are utilized to design
defense strategies to protect sensor nodes from attacks and to guarantee a high level of
trustworthiness for sensed data. We have discussed a few most recent game theory applications
overview and how it useful for the protection. Many security algorithms proved that many

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 222 TAGA JOURNAL VOL. 14

ISSN: 1748-0345 (Online) www.tagajournal.com

researchers have to put more effort in security in underwater wireless sensor networks with
optimal path.the proposed method will give you the Artificial intelligence result with high
dominationt potential in exploring the dark yet interesting side of the oceans, provided that the
pce of research continue to rise. Almost all of the applications that were brought to light in this
effort, have agreed upon the fact that extensive research needs to be carried out in the area of
acoustic communication to enhance the range and bandwidth of the underwater communication
modems. With the advancement in the areas of sensing systems such as underwater visual sensor
networks and real time streaming applications, we will require high bandwidth communication
modules with lower latencies.

REFERENCES:
1. Kamhoua C., Pissinou N., Miller J., Makki S.K. Mitigating routing misbehavior in multi-
hop networks using evolutionary game theory; Proceedings of the Globecom Workshops
(GC Wkshps); Miami, FL, USA. 6–10 December 2010; pp. 1957–1962.
2. Tembine H., Altman E., El-Azouzi R. Delayed evolutionary game dynamics applied to
medium access control; Proceedings of the International Conference on Mobile Adhoc
and Sensor Systems (MASS); Pisa, Italy. 8–11 December 2007; pp. 1–6.
3. Zhang Y., Guizani M. Game Theory for Wireless Communications and
Networking. CRC press; Boca Raton, FL, USA: 2011.
4. Pal R., Gupta B., Cianca E., Patel A., Kaligotla S., Gogar A., Wardana S., Lam V.T.,
Ganguly B. Playing ‘games’ with human health the role of game theory in optimizing
reliability in wireless health networks; Proceedings of the 3rd International Symposium
on Applied Sciences in Biomedical and Communication Technologies (ISABEL); Rome,
Italy. 7–10 November 2010; pp. 1–5.
5. Niyato D., Hossain E., Rashid M.M., Bhargava V.K. Wireless sensor networks with
energy harvesting technologies: A game-theoretic approach to optimal energy
management. IEEE Wirel. Commun. 2007;14:90–96.
6. Osborne M.J., Rubinstein A. A Course in Game Theory. MIT Press; Cambridge, UK:
1994.
7. Čagalj M., Ganeriwal S., Aad I., Hubaux J.P. On selfish behavior in CSMA/CA
networks; Proceedings of the 24th Annual Joint Conference of the IEEE Computer and
Communications Societies (INFOCOM 2005); New York, NY, USA. 13–17 March
2005; pp. 2513–2524.
8. Khayatian H., Saadat R., Mirjalily G. Distributed power allocation based on coalitional
and noncooperative games for wireless networks; Proceedings of the 5th International
Symposium on Telecommunications (IST); Tehran, Iran. 4–6 December 2010; pp. 367–
372.
9. Vadori V., Scalabrin M., Guglielmi A.V., Badia L. Jamming in Underwater Sensor
Networks as a Bayesian Zero-Sum Game with Position Uncertainty; Proceedings of the
Global Communications Conference (GLOBECOM); San Diego, CA, USA. 6–10
December 2015; pp. 1–6.
10. Reddy Y.B. A game theory approach to detect malicious nodes in wireless sensor
networks; Proceedings of the Third International Conference on Sensor Technologies and
Applications (SENSORCOMM); Athens, Greece. 18–23 June 2009; pp. 462–468.

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 223 TAGA JOURNAL VOL. 14

ISSN: 1748-0345 (Online) www.tagajournal.com

11. Agah A., Asadi M., Das S.K. Prevention of DoS Attack in Sensor Networks using
Repeated Game Theory; Proceedings of the ICWN; Las Vegas, NV, USA. 26–29 June
2006; pp. 29–36.
12. Gavrilovska L., Krco S., Milutinovic V., Stojmenovic I., Trobec R. Application and
Multidisciplinary Aspects of Wireless Sensor Networks: Concepts, Integration, and Case
Studies. Springer Science & Business Media; London, UK: 2010.
13. Vadlamani S., Eksioglu B., Medal H., Nandi A. Jamming attacks on wireless networks: A
taxonomic survey. Int. J. Prod. Econom. 2016;172:76–94.
14. Kamath G., Shi L., Song W.Z., Lees J. Distributed travel-time seismic tomography in
large-scale sensor networks. J. Parallel Distrib. Comput. 2016;89:50–64. doi:
10.1016/j.jpdc.2015.12.002.
15. Shan X., Zhuang J. Modeling Credible Retaliation Threats in Deterring the Smuggling of
Nuclear Weapons Using Partial Inspection-A Three-Stage Game. Decis.
Anal. 2014;11:43–62. doi: 10.1287/deca.2013.0288.
16. Akkarajitsakul K., Hossain E., Niyato D., Kim D.I. Game theoretic approaches for
multiple access in wireless networks: A survey. Commun. Surv. Tutor. 2011;13:372–395.
17. Shi H.Y., Wang W.L., Kwok N.M., Chen S.Y. Game theory for wireless sensor
networks: A survey. Sensors. 2012;12:9055–9097.
18. Benmammar B., KRIEF F. Game theory applications in wireless networks: A survey;
Proceedings of the 13th International Conference on Software Engineering, Parallel and
Distributed Systems (SEPADS’14); Gdansk, Poland. 15–17 May 2014; pp. 15–17.
19. Harn L. Group authentication. IEEE Trans. Comput. 2013;62:1893–1898.
20. Shen S., Yue G., Cao Q., Yu F. A survey of game theory in wireless sensor networks
security. J. Netw. 2011;6:521–532.
21. Yu Y., Li K., Zhou W., Li P. Trust mechanisms in wireless sensor networks: Attack
analysis and countermeasures. J. Netw. Comput. Appl. 2012;35:867–880.
22. Sun Y.L., Han Z., Yu W., Liu K.R. A trust evaluation framework in distributed networks:
Vulnerability analysis and defense against attacks. INFOCOM. 2006;6:1–13.
23. Sun Y., Han Z., Liu K.R. Defense of trust management vulnerabilities in distributed
networks. Commun. Mag. 2008;46:112–119. doi: 10.1109/MCOM.2008.4473092.
24. He D., Chen C., Chan S., Bu J., Vasilakos A.V. ReTrust: Attack-resistant and
lightweight trust management for medical sensor networks. IEEE Trans. Inf. Technol.
Biomed. 2012;16:623–632.
25. Komathy K., Narayanasamy P. Trust-based evolutionary game model assisting AODV
routing against selfishness. J. Netw. Comput. Appl. 2008;31:446–471.

© 2018 SWANSEA PRINTING TECHNOLOGY LTD 224 TAGA JOURNAL VOL. 14