# Introduction ireless networking grows rapidly because of the human desires for mobility and for freedom from limitation, i.e., from physical connections to communication networks [10]. Recent advances in wireless technology have equipped portable computers, such as notebook computers and Personal Digital Assistants (PDA's) with wireless interfaces that allow networked communication even while a user is mobile [4]. A particular kind of wireless network called mobile adhoc network is presently under development, which is the subject of this. A mobile adhoc network is a selforganizing and rapidly deployable network in which neither a wired backbone nor a centralized control exists. The network nodes communicate with one another over scarce wireless channels in a multi-hop fashion. In this, we propose an algorithm for topology management for the Adhoc wireless networks is: A power management algorithm to reduce the consumption of the power of each node in the adhoc wireless networks, by the introduction of MARI (Mobile Agents with Routing Intelligence) topology (a topology having MARI nodes) and management. The absence of a central infrastructure implies that an adhoc wireless network does not have an associated fixed topology. Hence, a most important task of an adhoc wireless network consisting of geographically dispersed nodes is to determine (in real time) an appropriate topology over which high-level routing protocols can be implemented [12]. Some of the properties of adhoc networks that make them difficult to manage are: # Mari topology formation and management Minimizing energy consumption is an important as well as a difficult challenge in mobile networking. The requirement of cooperation between power saving [7] and routing protocols is particularly important in the case of multi-hop adhoc wireless networks, where nodes must forward packets from one to another [3,8,9]. This thesis proposes a novel topology management scheme for adhoc wireless networks for power management called, MARI Topology. The nodes in this scheme are classified into three categories based on their power level. They are: a) MARI Nodes MARI nodes are selected in such a way that they have the maximum power level among their onehop neighbors and all non-MARI nodes in the one-hope neighborhood are within the transmission range of MARI nodes. These MARI nodes [16] have the routing intelligence i.e. they make decisions related to routing, such as path finding. Every MARI node has a group of member nodes connected to it, usually in its one-hop neighborhood. The responsibility of every MARI node is W to make necessary communication with any member node connected to itself or with other MARI nodes (through Gateways) within the network, for both transmission and reception of the packets. MARI nodes are selected or formed by a procedure that is explained later in this chapter. # b) Gateway Nodes The Gateway nodes [16] having sufficient power level are selected by the MARI nodes such that they can be used to forward packets between MARI nodes. Any two adjacent MARI nodes (within two-hop distance usually) in the network are connected through the concerned Gateway node only. Gateway nodes do not have routing intelligence. The MARI nodes select these Gateway nodes, according to the procedure outlined later. The MARI and the Gateway nodes stay continuously awake to route the packets of other member nodes. # c) Member Nodes A member node is a non-MARI and non-Gateway node. These are the nodes, which want to communicate with each other [16,17]. Every member node is connected to one of the MARI nodes (some kind of belonging or bonding) through which it transmits or receives the packets. The member nodes wake up only at certain specified time epochs, and for very short periods, during any given beacon period T . When a member node wakes up and if it does not have to transmit or receive data, then it goes to sleep mode again, after a brief period. This is the main principle behind the power-efficient operation of the network. The wake-up time epochs of each member node are determined apriority (pre-determined). In our simulation of the operation of an adhoc wireless network, this is accomplished with the help of pseudo-random number generator. Also, these wake-up time epochs of a member node are known to its corresponding MARI node and its one-hop neighbor nodes, through the WAKEUP messages that are exchanged at the beginning of a beacon period. Thus, the member node can remain in power saving sleep mode for most of the time [6,11], especially when it is not actively sending or receiving packets. The packets are routed over the virtual backbone consisting of MARI nodes and Gateway nodes, which are awake continuously. This is the main power-saving advantage of the topology that is suggested and nurtured in this thesis. From the fig. 2 it is clear that the number of nodes in the virtual backbone is 20% to 25% only. For an example, the list of MARI nodes, Gateway nodes and member nodes for the given MARI Topology network is shown in the table.1. From the table it is clear that the number of MARI nodes are: 09, number of Gateway nodes are: 12, and the number of member nodes are: 79. # Table.1: List of MARI nodes, Gateway nodes and member nodes In our simulation, we have considered that the nodes are operating in one of the four modes and their power consumptions are listed in table-2. # Mode Transmit Receive Idle Sleep Power Consumption 1400mW 1000mW 830mW 130mW Table. # III. Results We have used MATLAB 7.0 for the simulation of results. In this section we have shown the analysis of adhoc wire less network with MARI topology. The results are compared with the existing flat topology. The parameters we considered for analysis and evolution, and their respective results are given below: # IV. Conclussions The performance evaluation of the implemented wireless network is carried out in order (i) to demonstrate the successful operation of the MARI topology concept, (ii) to compare its performance to that of an equivalent (in size) flat-topology network, (iii) to prove the that the MARI topology performs a way better than an equivalent flat-topology. In our illustration and demonstration, the performance measures that we have considered for simulation study are, (i) the number of backbone nodes as a proportion of the total number of nodes in the network, (ii) overhead messages, and (iii) average power consumption in the network. Analysis is also done in the following cases: 1. Keeping the data packet size and beacon period constant. 2. Variable data packet size with constant beacon period. 3. Fixed data packet size with variable beacon period. V. Further scope ) is another important topic. When the multi-access level increases, the power dissipation and the throughput increase. Therefore, there may exist an 'optimal' multi-access level in a given context and under certain conditions. Finding that optimal multiaccess level can be a very good topic for R&D. 12![Fig.1: Random distribution of 100 nodes in the adhoc wireless network](image-2.png "Fig. 1 :Fig. 2 :") 2![Shows the amount of power consumption value by each node based on their mode of operation © 2012 Global Journals Inc. (US) Global Journal of Computer Science and Technology Volume XII Issue X Version I](image-3.png "2 :") 35![Fig. 3: Power consumption of the network as number of nodes in the network increases](image-4.png "Fig. 3 :Fig. 5 :") 1![Management protocols for multicasting (one source to Many number of destinations simultaneously) can be designed upon MARI topology or with some further and appropriate modifications to the MARI topology. Load distribution and load balancing among the paths in the network can be considered.(2) Performance study for multi-access level ( 1 > l](image-5.png "( 1 )") 3![Security levels on each packet, path and node can be implemented, with analysis of security by simulation or bench marking.](image-6.png "( 3 )") Power Saving Mechanism with Less Number of Nodes in the Routing Path in Adhoc Wireless Networks Using MARI Algorithm * Baker Data/Voice Communication over a Multihop, Mobile, High Frequency Network MartiBaker Giuli Lai proceedings of the IEEE Military Communications Conference (MILCOM'97) the IEEE Military Communications Conference (MILCOM'97) Nov1997 * Packet dropping in the MANET based on their malicious property of the nodes HabauxButtyan Proceedings of the 4th ACM International Symposium on Mobile Adhoc Networking and Computing, (MobiHoc2003) the 4th ACM International Symposium on Mobile Adhoc Networking and Computing, (MobiHoc2003) June 2003 * Minimum-energy mo-bile wireless networks revisited LiJHalpem IEEE Inter-national Conference on Communications June2001 * Guest Editorial: Wireless Adhoc Networks .ZHass IEEE Journal on Selected Areas of Communications 17 8 1329 august 1999 * The Dynamic Source Routing Protocol for Mobile Adhoc Networks YCHu DJohnson JJetcheva DAMaltz Mobile Ad-hoc Network (MANET) Working Group, IETF February 2002 * A power control MAC protocol for ad-hoc networks JungEVaidya N ACM MOBICOM 2002 * Power Control in Ad-Hoc Networks: Theory, Architecture, Algorithm and Implementation of the COMPOW Protocol VKawadia PRKumar RSSreenivas Narayanaswamy Proceedings of European Wireless European Wireless Feb. 2002 * Algorithmic Aspects of Topology Control Problems for Adhoc Networks LloydE L RLiu MaratheVRamanathan .R RaviS S June 2002 Mobihoc, EPFL Lausanne, Switzerland * Minimum energy mobile wireless networks THMeng Rodoplu IEEE Journal on Selected Area on Communications 17 8 August 1999 * CEPerkins Adhoc 2001 Addison-Wesley Boston * S "Power efficient MAC protocol for multihop radio networks CSRaghavendra Singh The Ninth IEEE International Symposium on Personal, Indoor and Mobile Radio Communications 1998 * Topology control of multihop wireless networks using transmit power adjustment Ramanathan Rosales -Hain Proceedings of INFOCOM INFOCOM 2000 * A performance comparison of multihop wireless adhoc network routing protocols JBroch Proceedings of MOBICOM 1998 MOBICOM 1998 1998 * J Nodes Bearing Grudges: Towards Routing Security, Fairness, and Robustness in Mobile Adhoc Networks .BSonja YvesLeBoudec Proceedings of the Tenth Euro micro Workshop on Parallel, Distributed and Networkbased Processing the Tenth Euro micro Workshop on Parallel, Distributed and Networkbased ProcessingCanary Islands, Spain IEEE Computer Society 2002 * J Cooperative routing in mobile adhoc networks: Current efforts against malice and selfishness .BSonja YvesLeBoudec Proceedings of Mobile Internet Workshop. Informatik Mobile Internet Workshop. InformatikDortmund, Germany 2002. October 2002