International Journal of Latest Research in Science and Technology Vol.1,Issue 4 :Page No.379-382 ,November-December (2012) http://www.mnkjournals.com/ijlrst.

htm

ISSN (Online):2278-5299

A QUALITY OF SERVICE BASED ADMISSION CONTROL SCHEME FOR MANETs

Annitha N 1, Pushpavalli M 2 PG Student 1,sarnitya@gmail.com Asst.Prof(Sr.G) of ECE Department 2,pshpvll@gmail.com Bannari Amman Institute of technology ,Sathyamangalam
Abstract : Effective Mobile Ad hoc Networks (MANETs) require QoS capabilities that provide the fault tolerance and fast recovery mechanism when there is a link failure.since MANET is a dynamic topology in nature,the protocols used in multi-hop networks maintain the routing stability. In this paper,an efficient call admission control scheme is used to maintain the flows in the network,The consumption of bandwidth by a flow and the availability of resources to any wireless node depends upon the neighboring nodes within its transmission range. In this project, we proposed a scalable and an efficient call admission control scheme for maintaining the bandwidth which is incorporated with AODV(Ad hoc ondemand) protocol to support the QoS requirements in multi-hop ad hoc networks Keywords: Mobile Adhoc network, Admission control, Bandwidth, QoS, AODV protocol

INTRODUCTION The routing in Mobile ad hoc networks is organized with the cooperation among the nodes of the network, the nodes must cooperate with each other to provide QoS requirements.Such methodology includes the cooperation of implementing certain policies at the end points of flows and an admission control policy.Many solutions have been proposed for QoS support in wired networks and in wireless networks like TDMA or CDMA. But, the physical Characteristics of single-channel wireless networks like IEEE 802.11 networks [1] do not adhere to such solutions.In multi-hop ad hoc networks, the nodes cannot communicate directly, and may lead to some contention. Such a contention, is referred as c-neighborhood which affect resource allocation at individual nodes of the network in two different ways : (i) resource allocation at an individual node needs information about the allocated bandwidth along the entire route, existing beyond its transmission range; (ii) contention along a route may involve multiple nodes. RELATED WORK A number of admission control protocols are proposed for mobile ad hoc networks have been proposed.Several approaches incorporating TDMA-based protocols have been proposed to support QoS requirements in wireless ad hoc networks [5] those are oriented around an effective synchronization among all nodes in the network and implementing a slot allocation algorithm, which is vulnerable to mobility in the network. R. Rozovsky et al [12] incorporates a single

Channel MAC layer scheduling algorithms have for resource allocations in wireless ad hoc networks, which implement a single channel, shared by all the nodes of the network, where QoS requirements are maintained by a coordination of transmission schedules of packets among the nodes. Y. Hwang et al [7] uses Adaptive QoS Routing Algorithm(ADQR) which uses signal strength to predict route failures and obtains estimated bandwidth from lower layers in which Destination initiates reverse QoS-aware route discovery. D. Nguyen et al [10] uses Optimized Link State Routing (OLSR) protocol which always finds a path with larger available bandwidth. the mechanism followed is a proactive mechanism which automatically provides new route if one exists. G.-S. Ahn et al [1] used SWAN protocol in which it does not pay attention to the resources of the c-neighbors of a node during admission control, rather focus on only the local resourcesin which the packets sessions experiencing congestion, notify source which attempts to re-admit, else terminates. S.-B. Lee et al [9] proposed an In-band signaling protocol (INSIGNIA) provides an guaranteed throughput which is an connectionless routing architecture deal with only high level issues and do not consider the available bandwidth at the c-neighbors which results in attempting a local re-route, then increase scope of route repair process. Binod Kumar Patanayak et al [2] proposed a Multihop Bandwidth Management Protocol (MBMP) which is similar to DSR protocol in which they have considered the

ISSN: 2278-5299

379

Annitha et.al, International Journal of Latest Research in Science and Technology.

available bandwith at the c-neighbors , which attempts to re-admit the new cached route incase of link failure or packet drop and route repair process is not recovered. The goal of admission control is to determine if the available resources can suffice to the requirements of a new flow to be admitted, not disrupting the bandwidth levels of the existing flows by two challenges(12) :perception of available bandwidth and bandwidth consumption. PERCEPTION OF AVAILABLE BANDWIDTH A node flow in the shared medium of wireless network is starting to transmit, it consumes bandwidth of its c-neighbors. c-neighborhood available bandwidth: The maximum amount of bandwidth that a node can use without using the reserved bandwidth. local available bandwidth:The amount of unused bandwidth at a node at any point of time. To admit a flow successfully a node requires an sufficient local and c-neighborhood available bandwidth. ! Calculation of Local Available Bandwidth: MBMP uses idle channel time for estimation of local available bandwidth.After monitoring the amount of idle channel time, Tidle, during every period of time, Tp, the local available Blocal, for a given node can be calculated using a weighted average equation. (1) Blocal= Blocal+(1- )(Tidle / Tp) Bchannel ---(1) Where,Blocal is the local available bandwidth of the node in the preceding period, (initially zero),Bchannel is the channel capacity in bits per sec and the weight as [0, 1]. ! Calculation of c-Neighborhood Available Bandwidth Here,we considered two active approaches and one passive approach for estimation of bandwidth at c-neighbors of a node. In active approaches, cneighbors actively exchange the information about bandwidth between each other node. In passive approaches, a node has to monitor the channel passively, to obtain the c-neighborhood available bandwidth.

consequently, all nodes, that are one hop away from the querying node, receives the query message and broadcast it to other nodes. since the carrier-sensing range in IEEE 802.11 is twice that of the transmission range, we considered a hop count as 2. When a node senses the carrier strength is smaller than the Neighbor-carrier-sensing Threshold, then all c-neighbors of the node sense the idle channels. With the amount of time the channel is in idle state T idle for every period of time, Tp the c-neighborhood available bandwidth, Bneighbor can be derived using the equation. (2) Bneighbor= Bneighbor+(1- )Tidle/Tp Bchannel ----(2) Where,Bneighbor is the c-neighborhood available bandwidth in the preceding period of time, initially zero. PERCEPTION OF BANDWIDTH CONSUMPTION The second challenge represents the estimation of bandwidth to be consumed by a requirements of new flow. For this we considered MAC layer implementation of RTS-CTS-DATA-ACK handshake which is shown in eqtn (3) Tdata=Tdifs+Trts+Tcts+ ((L+H)/Bchannel) + Tack + 3Tsifs ----- (3) For R packets, with an average packet size of L, the channel bandwidth requirement, W is calculated by eqtn (4) W = R x Tdata x Bchannel ------ (4) PROPOSED WORK MBMP (Multi-hop Bandwidth Management Protocol) Design incorporated with AODV protocol MBMP design constitutes four components namely, route discovery, full and partial admission control, and building cneighbor sets. ! Route discovery MBMP implements on-demand route discovery with source routing, similar to AODV protocol. The effectiveness of using source routing based protocol, is to determine the route, along which the packets of the flow move, by admission control and it maintains enough reserved bandwidth for initiation of the flow. MBMP implements partial admission control in the process of route discovery in order to reduce the message overhead.In this approach, prior to sending data, a source node broadcasts the route request message to all its nearby neighbors. The sequence of hops, referred as Partial Route (PRoute)used to to eliminate the circular routes and to determine the lower bound of contention count along the entire route. ! MBMP admission control algorithm In the process of route discovery, multiple possible routes can be determined by the source node to reach a destination. MBMP implements admission control in two phases of route discovery. (1) Partial admission control is performed during route request, when each node along the route receives route request message; (2) Full

Fig. 1 Multi Hop Approach In Fig. 1, The cost of a query in this approach is directly related to the number of hop count required to reach the c-neighbors. If the querying node sends a message, and

ISSN: 2278-5299

380

Annitha et.al, International Journal of Latest Research in Science and Technology.

admission control is performed during route reply, when each node along the route receives a route reply message. ! Partial Admission Control In MBMP, after receiving a route request message, a node implements partial admission control by comparing its available bandwidth with probably underestimated bandwidth consumption. The three versions of MBMP includes MBMP-multi-hop, MBMP-power and MBMP-CS.In MBMP-multi-hop and MBMP-CS approaches, estimation of cneighborhood available bandwidth involves identification of c-neighbors is an expensive operation. For this reason, querying c-neighbors should be avoided for the nodes. ! Full Admission Control: Full admission control is performed by a node in route reply phase.. At first the local available bandwidth of the node is compared with calculated bandwidth of the flow at the nodes location. If the local available bandwidth of the node is found to be higher than the bandwidth consumption of the flow, the c-neighborhood available bandwidth of the node is compared with the bandwidth consumption of the flow. If the local available bandwidth at the node is found to be smaller than the bandwidth consumption of the flow, then the node sends an admission rejection message back to the initiator and the admission control in c-neighborhood of the node fails. In case an admission rejection message is not received by the initiator in a stipulated period of time, it times out and the initiator assumes that the full admission control succeeds. The stipulated period of timeout is determined by the propagation delay, transmission time of admission rejection message and computation time. ! Building c-Neighbor Set: The accuracy of admission control strictly depends on the completeness and accuracy of cneighbor set. Information about c-neighbors in MBMP can be gathered by monitoring control and data messages incorporated in c-neighbor set. After the accurate estimation of bandwidth consumption of the flow only, a new flow should be admitted.When the route request messages are flooded across the entire network in the route request phase, and the route request message carrying the partial route it has traversed, reaches the node, the node can cache the last two hops of the partial route as its c-neighbor set and collect information about its c-neighbors. Again, as admission request messages in MBMP-multi-hop and MBMP-power approaches are sent to reach c-neighbors, the c-neighbor set can further be updated by caching the senders and forwarding nodes of these messages in the c-neighbor set. After receiving the route reply message, a node can add the last two forwarding nodes of this message to its c-neighbor set. IV. SIMULATION ENVIRONMENT The performance of Ad Hoc routing protocol AODV with varying the number of mobile nodes with various

parametric includes packet received, end to end latency, throughput which is simulated using network simulator NS-2.34 [13]. The network simulator ns-2 is discrete event simulation software for network simulations which means it simulates events such as sending, receiving, forwarding and dropping packets. Table 1 shows the simulation parameters used in this evaluation. End-to-End latency It is defined as the average time taken by the packet to reach the server node from the client node. Fig. 6 depicts the mobility Vs delay graph. The delay is obtained by varying the mobility speed. The delay for AODV is less compared to DSR. TABLE 1: PARAMETERS VALUES FOR AODV SIMULATION

Fig. 6 No. of Nodes Vs Delay(ms) For 30 nodes ,the delay of proposed method is 2.05 ms which is lesser than the existing method MBMP similar to DSR protocol which has 2.55 ms.Thus the proposed method performs better than the existing method for low & high traffic loads. The delay is decreased by 3% than that of existing method Throughput Throughput is the number of useful bits per unit of time forwarded by the network from a certain source address to a certain destination, excluding protocol overhead, and excluding retransmitted data packets. Fig. 7 is drawn between Nodes and throughput. As the mobility of node increases, the throughput of the network increases in the proposed method which is much higher than that of the existing method. The throughput is raised to 3.1% due to bandwidth management and route failure process.

ISSN: 2278-5299

381

Annitha et.al, International Journal of Latest Research in Science and Technology.

6.

7.

8. 9.

10.

11.

Fig. 7 No. of Nodes Vs Throughput(bytes) CONCLUSION AND FUTURE WORK In order to provide higher quality services, it is required to enhance wireless multihop infrastructures with social features to achieve bandwidth management in multi-hop mobile ad hoc networks with admission control. Major contribution of this paper is the inclusion of information from nodes within the carrier-sensing range and outside transmission range during the process of admission control. As demonstrated through simulations, AODV protocol effectively manages requests for bandwidth even beyond the capacity of the network, and consequently reduces the control message overhead on the network, which performs only the admission control part of QoS protocol stack in which Quality of Service is achieved. In future, the infrastructure, ad hoc configurations and the factors such as overhead, security and reliability will be considered. a fully distributed contention window adaptation (CWA) mechanism,is to be used which adjusts the channel access probability depending on the difference between the incoming and outgoing traffic at each node, in order to equate the traffic forwarding capabilities among all the nodes in the path to improve the throughput of multihop wireless networks
REFERENCES
1. Ahn G.S., Campbell A., Veres A. and. Sun L.-H., SWAN : Service Differentiation in Stateless Wireless AdHoc Networks, Proc. Infocom, 2007 Binod Kumar Patanayak, Alok Kumar Jagadev, Manoj Kumar Mishra, Dr Manojranjan Nayak,Multi-Hop Bandwidth Management Protocol for Mobile Ad Hoc Networks, International Journal of Managing Information Technology (IJMIT), Vol.2 No.4, November 2010 Biswas S. and Morris R.,ExOR: Opportunistic Multi-Hop Routing for wireless networks, ACM SIGCOMM Computer Communication Review,2008 Chen T.W., Tsai J.T. and Gerla M., QoS Routing Performance in Multihop Multimedia Wireless Networks, Proc. IEEE Intl Conf. Universal personal comm.(ICUPC), IEEE TRANSACTIONS ON MOBILE COMPUTING,August 2007 Gao D.,Cai J. and K. Ngan, Admission Control in IEEE 802.11e Wireless LANs, Network, IEEE, vol. 19, no. 4, pp. 613, 2009

12.

13.

Hanzo L. and Tafazolli R., A Survey of QoS Routing Solutions for mobile Adhoc networks, IEEE Communication Surveys Tutorials, vol. 9, no. 2, pp.5070, 2007 Hwang Y.and Varshney P., An Adaptive QoS Routing Protocol with Dispersity for Ad-hoc Networks, Proc.. 36th Hawali IntlConf. Sys. Sci., Jan 2008 IEEE Computer Society, 802.11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications, 1999 Lee S.B.,Ahn G.S., Zhang X. and Campbell A, INSIGNIA: An IP-Based Quality of service Framework for Mobile Ad-hoc networks. Parallel Distributed computing, special issue on wireless and mobile computing and communications, vol. 60,pp.374-406,2007. Nguyen D. and Minet P., Interference-Aware QoS OLSR for Mobile Ad-hoc network Routing, in Proc. 6th Int. Conf. Software Engineering, artificial intelligence, Networking and Parallel / Distributed Computing, USA, pp. 428435, May 2009 Ratica J. and Dobos L.,Mobile Ad-Hoc Networks Connection Admission Control Protocols Overview, in Proc. 17th Int. Conf, Radio elektronika, pp.14, Apr. 2008 Rozovsky R. and Kumar P.R.,SEEDEx:A MAC Protocol for Ad Hoc Network, Proc. ACM Symp.Mobile Ad Hoc Networking & Computing, 2007. Samir R.Das. and Mahesh K. Marina., Ad hoc on-demand multipath distance vector routing, Wireless. Communication. Mobile Computing, pp.969988, 2008

2.

3.

4.

5.

ISSN: 2278-5299

382