Impact of MAC Protocol on the Performance of Routing in MANETs by Allied Journals

International Journal of Engineering, Management & Sciences (IJEMS) ISSN-2348 –3733, Volume -2, Issue-3, March 2015

Impact of MAC Protocol on the Performance of Routing in MANETs Ghanesh Jain, Ambika Gupta, Satpal Singh Kushwaha

 Abstract— Mobile Ad hoc Network is an autonomous system of mobile nodes that utilize multi-hop radio relaying and are capable of operating with any support from fixed centralized infrastructure. Since nodes in such networks can leave or enter any time development of a routing protocol that can respond quickly to underlying dynamic topology and utilizing the available bandwidth efficiently is a research challenge. There have been many research works comparing the performance of routing protocols under varying conditions and constraints. The majority of the research is focused towards coping up with the node mobility, network density, energy reduction, routing and control overhead minimization and increasing the throughput. Most of them are unaware of MAC Protocols, which affect the relative performance of routing protocols considered in different network scenarios. As routing and MAC protocol runs in coordination, the choice of MAC is an important criterion. This paper investigates effect of MAC protocols on the relative performance of AdHoc routing under various scenarios. Two routing protocols are simulated in NS2 with two different MAC protocol and the results show that the performance of routing protocols of ad hoc networks will suffer when run over different MAC Layer protocols. Index Terms— MANET, Routing, MAC, DSR, AODV, Overheads, Performance

I. INTRODUCTION Mobile Ad-Hoc network (MANET) is a collection of nodes, which are capable of communicating with each other utilizing multihop relaying without any centralized infrastructure by forming an arbitrary and dynamic wireless network. The dynamic topology of MANET makes it necessary to route the traffic through multihop paths to make communication possible between two nodes. So the routing protocol must have the capability to respond in a whiz to the changes in underlying topology. Available bandwidth is also an important constraint that should be dealt optimistically by incorporating the methodologies to minimize the routing and control overheads. Several routing protocols exist, addressing the problems of routing in mobile ad-hoc networks. Depending on the route Manuscript received March 24, 2015 Ghanesh Jain ,PG Scholar Department of Computer Science and Engineering Modern Institute of Technology & Research Centre, Alwar,Rajasthan-India Ambika Gupta, Assistant Professor, Department of Computer Science and Engineering Modern Institute of Technology & Research Centre, Alwar, Rajasthan-India Satpal Singh Kushwaha, Assistant Professor, Department of Computer Science and Engineering Modern Institute of Technology & Research Centre, Alwar, Rajasthan-India

acquisition time the protocol may be Proactive or Reactive. Reactive protocols are characterized by nodes acquiring and maintaining routes on-demand. In general, when a node has some data to send to some unknown destination a query is flooded onto the network and replies, containing possible routes to the destination, are returned. There are various reactive routing protocols among which Ad Hoc On Demand Distance Vector Routing Protocol (AODV) and Dynamic Source Routing (DSR) have caught a great attention. The routing protocol runs over a MAC layer protocol so the selection of MAC protocol plays a vital role in the efficiency of a routing protocol. The nodes in MANETs share a common broadcast radio channel. Due to limited available radio spectrum there are limitations on the bandwidth for communication. Access to this shared medium should be controlled by the underlying protocol in such a way that at the same time it should maintain the fairness of share of the available bandwidth towards the nodes and utilize it efficiently. The basic aim of any communication system shall be to transport the data, from source to destination successfully, in a reliable way and in minimum time. If a good routing protocol is selected to run on an inferior performance MAC frequent link breaks will result in increased end to end delay and choice of robust MAC for a substandard routing mechanism may result in increased routing overheads and packet drops thereby increasing the end to end delay resulting in bandwidth consumption. Although substantial attention have been devoted towards the design and development of routing protocols there is relatively less understanding of interaction of routing protocol with other network layers and effect of other layers on routing and that of routing on other layers. There exist many MAC protocols for MANETs which can be broadly categorized as contention based, contention based with reservation mechanism and contention based with scheduling mechanisms. In this paper the effect of two different MAC protocols on two popular routing protocols is presented to demonstrate the effects of MAC on the performance of routing. Rest of the paper is organized as follows. Section 2 discusses the related work. Routing and Media access control protocols considered in this study are explained in brief in Section 3. Simulation environment is explained and the results with the performance metric are presented in Section 4. Last section concludes the paper with discussion on results and future scope.

www.alliedjournals.com

Impact of MAC Protocol on the Performance of Routing in MANETs II. RELATED WORK In this section the previous studies concerning routing performance in AdHoc network and effect of MAC on routing are summarized. Most of the previous studies compare the different routing on the basis of various performance metrics which take into account the factors involved in routing activity only. Previously very less work has been carried out to investigate the effect of MAC on the performance of routing. MAC and routing both are very crucial functions in any type of communication. In [1] author compare four routing protocols based on node pause time. AODV, DSR and STAR protocols are compared for their routing performance [2]. There are few studies investigating the choice of MAC-Routing combination suitability. Authors in [7,8] used statistical ANOVA method to analyze the interaction but they used only two connections and compared the performance of AODV, DSR and LAR1 with IEEE802.11DCF, CSMA and MACA with latency, total number of packets received and long term fairness of the protocols as the performance measures. In [9] four routing protocols DSR, AODV, DSDV and WRP are investigated with IEEE802.11DCF, CSMA and E-TDMA with 25 nodes over a large area of size 1000m X 600m and considered average end to end delay as the major performance metric which includes queuing delays and retransmission delays along with Packet delivery ratio and control overhead as other metrics. But it is not clarified how end to end delay take into account the MAC behavior as it is the average time taken by a data packet to reach from source to destination. This paper includes number of MAC packets transmitted as an additional performance metric to understand the relative behavior of underlying MAC protocol and 50 nodes are simulated over a relatively small area of size 500m X 500m to increase the node density.

III. OVERVIEW OF PROTOCOLS This section gives brief introduction of the protocols used in this study. DSR and AODV are two popular demand driven routing protocols. Performance of these routing protocols is studied with IEEE802.11DCF and TDMA media access control protocols in this paper.

A. Dynamic Source Routing (DSR) DSR [4] is a beaconless demand driven routing protocol designed for controlling the bandwidth consumed by control packets in infrastructureless multihop wireless networks with elimination of the periodic table update messages required in table driven approaches. In this protocol when a mobile (source) node has a packet to send to some destination, it first checks its route cache to see if there exists already a route to the destination. If there exist an unexpired route to the destination it uses this route for data transfer otherwise it initiates a route discovery by broadcasting a route request packet containing the destination address and the unique identification number. The node receiving the route request rebroadcast the packet to its neighboring nodes by adding its own address to the associated route record if it is not

forwarded already or if it is not a destination node itself. A reply to route request is generated by the destination or an intermediate node of which route cache contains an unexpired route to the destination. Route cache is populated with routes that can be extracted from the information contained in data packets that get forwarded. The route maintenance is accomplished with the use of route error packets and acknowledgments. While operating in promiscuous mode the intermediate nodes learn about the link breaks. Information so gained is used by the route cache to deactivate maintained route.

B. Ad Hoc On-demand Distance Vector (AODV) AODV [5] is also a demand driven routing protocol similar to DSDV [6] algorithm which employs destination sequence numbers to identify the most recent path. When a source node has a packet to send to some destination it initiates a path discovery process by broadcasting a route request packet to all the neighboring nodes in its range which then forward the request to their neighbors. This process is repeated by all the intermediate nodes receiving the route request till the intermediate node is either a destination or has a fresh enough path to the destination. The freshness of the path available at the intermediate node is checked by comparing the sequence number in the request packet to the number available at the node. While forwarding the route request the intermediate nodes, discarding the duplicate copies, record the address of the neighboring node from which it has received the first copy of the route request along with its broadcast identity. Upon receipt of route request the destination or an intermediate node (with a fresh enough path to the destination) sends a route reply along the reverse path. A route discovery process is reinitiated by the source if it receives a link failure error message propagated by the node in the upstream of the path break. C. IEEE 802.11 DCF The IEEE 802.11 specifies two modes of MAC protocol: Distributed Coordination Function (DCF) mode (for ad hoc networks) and Point Coordination Function (PCF) mode (for centrally coordinated infrastructure-based networks). The DCF in IEEE 802.11 is based on CSMA with Collision Avoidance (CSMA/CA), which can be seen as a combination of the CSMA and MACA schemes. The protocol uses the RTS–CTS–DATA–ACK sequence for data transmission. Time slots are divided into multiple frames and there are several types of inter frame spacing (IFS) slots. The node waits for the medium to be free for a combination of these different times before it actually transmits. Different types of packets can require the medium to be free for a different number or type of IFS. For instance, in ad hoc mode, if the medium is free after a node has waited for DIFS, it can transmit a queued packet. Otherwise, if the medium is still busy, a backoff timer is initiated. The initial backoff value of the timer is chosen randomly from between 0 and CW-1 where CW is the width of the contention window, in terms of time-slots. After an unsuccessful transmission attempt, another backoff is performed with a doubled size of CW as decided by Binary Exponential Backoff (BEB) algorithm. Each time the medium is idle after DIFS, the timer is decremented. When the timer expires, the packet is

www.alliedjournals.com

Number of MAC packets

International Journal of Engineering, Management & Sciences (IJEMS) ISSN-2348 â&#x20AC;&#x201C;3733, Volume -2, Issue-3, March 2015 transmitted. After each successful transmission, another Average end to end delay: It is the ratio of total delay to random backoff (known as post-backoff) is performed by the number of packets received successfully. transmission completing node. A control packet such as RTS, Normalized routing overhead: Ratio of number of routing CTS or ACK is transmitted after the medium has been free for packets transmitted to number of data packets successfully SIFS. received. MAC overhead: Number of MAC packets transmitted. D. TDMA Unlike contention based IEEE802.11MAC protocol a TDMA C. Results MAC protocol allocates different time slots for nodes to send Following figure shows the effect of MAC on the and receive packets. The superset of these time slots is called performance of routing protocols. Fig.1 shows the number of a TDMA frame which contains preamble besides the data MAC packets transmitted over the simulation time of 500 transmission slots. Within the preamble, every node has a seconds. dedicated subslot and uses it to broadcast the destination node identity of outgoing packet. Other nodes listen in the 30000 preamble and record the time slots to receive packets. Each 25000 node turns its radio on and off explicitly to avoid unnecessary 20000 802.11-AODV power consumption. The radio only needs to be on when it is 802.11-Dsr in the preamble phase and there is a packet to send and 15000 Tdma-Aodv receive. 10000 IV. SIMULATION AND RESULTS

Tdma-Dsr

5000 0

A. Simulation model

Table 1. Simulation parameters

Simulation Parameters Network size 50 nodes Area 500m x 500m Traffic model CBR Traffic sources 30% Packet size 512 bytes Packet rate 4 packets/s Max. speed 20m/s Transmission range 250m Bandwidth 2 Mb/s Node movement model Random way point Simulation Time 100 s Pause time 0, 20, 40, 60, 80, 100 sec.

100

Pause time (sec)

Fig. 1 - Number of MAC packets transmitted It can be observed that number of MAC packets transmitted is far less for DSR as compared to AODV with both TDMA and IEEE802.11DCF. It is also clear for both the routing protocols TDMA requires less number of MAC packets as compared to IEEE802.11DCF. The normalized routing overload is shown in Fig.2. The normalized routing overhead is more for AODV as compared to DSR with both IEEE802.11DCF and TDMA. TDMA MAC has slightly lower normalized routing overhead. 6

Normalised routing load

Network Simulator2 [10] is used for the simulations. The traffic sources are CBR. The source-destination pairs are spread randomly over the network. The node movement generator of ns-2 is used to generate node movement scenarios. The movement generator takes the number of nodes, pause time, maximum speed, field configuration and simulation time as input parameters. The propagation model is the two ray ground model. The simulation parameters used are shown in table 1. Several runs of each scenario are simulated for enough time to reach and collect the desired data at steady state to obtain statistically confident averages.

5 4

802.11-AODV 802.11-Dsr

Tdma-Aodv Tdma-Dsr

2 1 0 0

100

Pause tim e (sec)

Fig.2 Normalized routing overhead

B. Performance Metric Following performance metrics are considered for studying the effect of MAC on routing. Packet delivery fraction (PDF): The Packet delivery is defined as the ratio of number of data packets successfully received to number of total data packets transmitted.

Fig.3 (a) and Fig.3 (b) show that the average end to end delay with TDMA is very large as compared with IEEE802.11DCF but still AODV performs better to DSR. It is obvious because DSR packet carries the entire route it has to go through where as AODV packet carries the information of only next hop in the route.

www.alliedjournals.com

Impact of MAC Protocol on the Performance of Routing in MANETs

1.2

0.07 0.06 0.05

802.11-AODV

0.04

802.11-Dsr

0.03 0.02 0.01 0 0

100

0.8 802.11-Dsr

0.6

Tdma-Dsr

0.4 0.2

Fig.3 (a) Average end to end delay (IEEE802.11DCF) Tdma-Aodv

100

Pause tim e (sec)

25 Average end to end delay

Pause time

Fig.4 (b) packet delivery fraction (DSR)

Tdma-Dsr

IV. CONCLUSION AND FUTURE SCOPE

10 5 0 0

100

Pause time (sec)

Fig.3 (b) Average end to end delay (TDMA) Fig.4 (a) and Fig.4 (b) show the effect of choice of MAC on the packet delivery fraction in AODV and DSR respectively. The number of packets dropped is a very serious issue in any type of communication. The communication system is supposed to transport the packets successfully from source to destination in real time. TDMA performance is very poor. With TDMA MAC the average end to end delay is very high and hence it is very prone to drop the packets. The variation in the packet delivery ratio with different MAC protocol indicates that the performance of routing varies significantly under various media access control protocols.

Packet deleivery fraction

Average end to end delay

0.08

1.2 1.1 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

802.11-Aodv Tdma-Aodv

100

Pause time (sec)

Fig.4 (a) Packet delivery fraction (AODV)

The routing and MAC both are very crucial in wireless communication and the choice of MAC affects the performance of routing. In this paper two routing protocols are investigated to observe the effect of MAC with only two MAC protocols and it is shown that the routing performance is MAC dependent. So it is necessary to understand the performance of MAC and routing together instead of analyzing the routing performance in isolation. The detailed investigation of interaction of MAC and routing protocols needs to be carried out to understand the behavior of the system when it is scaled up in terms of node density and traffic both. REFERENCES [1] Broch, D. A. Maltz, D. B. Johnson, Y. C. Hu, and J.Jetcheva “A Performance Comparison of Multi-Hop Wireless Ad Hoc Network Routing Protocols”, In Proc. of the ACM/IEEE MobiCom, October 1998. [2] C Hong Jiang, J. J. Garcia-Luna-Aceves, “Performance Comparison of Three Routing Protocols for Ad Hoc Networks”, In Proc. of the IEEE ICCN 2001,2001. [3] S. R. Das, C. E. Perkins, and E. M. Royer. “Performance Comparison of Two On-demand Routing Protocols for Ad Hoc Networks”, Proceedings of the IEEE Conference on Computer Communications (INFOCOM), Tel Aviv, Israel, March 2000, pp. 3-12, [4] D.Johnson, D.Maltz, Y.Ho, “The dynamic source routing protocol for mobile ad hoc networks”, IETF Manet working group, Internet RFC 2026. [5] C. E. Perkins and E. M. Royer, “Ad Hoc On-Demand Distance Vector Routing,” Proceedings of IEEE Workshop on Mobile Computing Systems and Applications 1999, February 1999, pp. 90-100. [6] C. E. Perkins and P. Bhagwat, “Highly Dynamic Destination-Sequenced Distance-Vector Routing (DSDV) for Mobile Computers,” in Proceedings of ACM SIGCOMM 1994, August 1994, pp. 234-244. [7] C.L.Barett, M.Drozda, A Marathe, M.V. Marathe,”Analyzing interaction between network protocols, toplogy and traffic in wireless radio networks,”, IEEE 2003 [8] C.L.Barett, M.Drozda, A Marathe, M.V. Marathe,”characterizing the interaction betweem routing and Mac protocols inAd-hoc networks”, MobiHoc02, June 9-11, 2002

www.alliedjournals.com

International Journal of Engineering, Management & Sciences (IJEMS) ISSN-2348 –3733, Volume -2, Issue-3, March 2015

[9} T.G.Basavaraju, Subir Kumar Sarkar and C. Puttamadappa, “ Impact of MAC layer on the performance of Routing Protocols in Mobil Ad hoc Networks”, International Journal of Information Technology, Vol.3, No.4, 2006 [10] K. Fall and K. Vardhan, The Network Simulator (ns-2). Available: http://www.isi.edu/nsnam/ns1.