IJERT-Performance Analysis of Few Replication-Based Routing Protocol in Delay Tolerant Network (original) (raw)

2014, International Journal of Engineering Research and Technology (IJERT)

https://www.ijert.org/performance-analysis-of-few-replication-based-routing-protocol-in-delay-tolerant-network https://www.ijert.org/research/performance-analysis-of-few-replication-based-routing-protocol-in-delay-tolerant-network-IJERTV3IS041940.pdf with the rise of new technologies there arise a new class of challenged wireless network derived from deep space communication known as Delay Tolerant Network or DTN characterized by intermittent connectivity, long delay, asymmetric data rate and high error rate. DTN networks lack instantaneous end-to-end communication path between the source and destination, these are opportunistic networks. Due to which routing protocols are of great concern in these class of network. In this paper we have investigated and compared the performance of five DTN routing protocols namely: EPIDEMIC, PROPHET, PROPHETv2, RAPID and SPRAY AND WAIT using two different simulation setup in the ONE simulator. One by analyzing the performance by varying the message TTL(Time to Leave) and keeping the buffer size constant, next by varying the buffer size and keeping the message TTL constant. The performance is compared based on three metrics namely: overhead ratio, average latency and delivery probability in both the scenarios. From the results obtained in both considerations it is observed that the SPRAY AND WAIT routing protocol gives the best performance. Keywords-Delay tolerant networks, EPIDEMIC, PROPHET, PROPHETv2, Spray and wait, RAPID, opportunistic network environment (ONE) Ι. INTRODUCTION DTN or disruption tolerant network is a practical class of challenged wireless network evolved from Mobile ad hoc Network. In 2002 Kelvin Fall [1] coined the term delay tolerant network and the DTN acronym by adopting some of the ideas of interplanetary network design to terrestrial networks. DTN networks are characterized by limited resources, long delay, asymmetric data rate, high error rate Intermittent connectivity and low SNR (Signal to Noise Ratio). The data transmission process of DTN networks as compared to the traditional networks is quite different, there is no fixed end-to-end path between the end hosts in a DTN network and the network topology is dynamic. All the nodes in a DTN network can act as a router with a transmission range and buffer (to store data as it adopts a store-and-forward policy).The data transmission takes place when a mobile node comes into transmission range of another mobile node until then the message is stored in its buffer. Examples of DTN include Exotic Media Networks, Vehicular Networks, Military Ad-Hoc Networks, Terrestrial Mobile Networks, and Sensor/Actuator Networks etc. In this paper we have analyzed and compared the performance of five different DTN replication-based routing protocols (EPIDEMIC; PROPHET; PROPHETv2; RAPID; Spray and Wait) using two different simulation scenario one by varying the message TTL and keeping buffer size constant and another by varying the buffer size and keeping the message TTL constant. These five protocols were analyzed on three different metrics namely Over Head Ratio, Delivery Probability and Average Latency. The details of the simulation setup along with the metrics are given in section 3. The rest of the paper is organized as follows: section 2 Briefly gives a review of routing in DTN and an abstract of the five routing protocols viz. EPIDEMIC, PROPHET, PROPHETv2, Spray and Wait and RAPID. Section 3 describes the details of the simulator used and the simulation setup for both the considered scenarios. Section 4 discusses the simulation results. Section 5 concludes this paper. II. ROUTING IN DTN As compared to the traditional routing protocol assumptions there is no end-to-end path between the source and destination in a DTN network to route data. Due to this lack of connectivity between the end hosts the main objective of routing in this challenged network is to maximize the message delivery probability, minimize delivery latency along with it also minimize the use of resources (i.e. network bandwidth, buffer space and battery energy). To fulfill these objectives many routing protocols have been devised which is basically based on store-and-forward mechanism. Routing protocols in DTN can be classified based on many characteristics. In this paper we have adopted the popular taxonomy used by Balasubramanian et al. [2] to classify a large number of DTN routing protocols based on whether the protocol is replication based or forwarding based. Protocols that create replicas of messages are known as replication based and those that do not create replicas are known as forwarding based. There is also variation in the replication