Opportunistic Networks (OppNets) are a sub-part of DTN but are quite different from DTN. OppNets do not need any end to end path between the nodes, as they forward the message in the store-carry-forward method. Due to the absence of an end to end path, nodes forwarding messages may face a delay. The nodes have zero or partial information about its neighboring node due to which uncertainty arises. OppNets are infrastructure-less networks, so they can be implemented in such situation where human reach is not possible. Due to these features, they have gained much interest in researches. OppNets work well when designed in an e cient manner. The two parameters to design a good OppNet are Mobility Model and Routing Protocols. Mobility models de ne the pattern of movement of the nodes and the routing protocols de ne the best, shortest, and e cient path so that a node can reach its destination in the least time possible. Proper selection of the mobility model and routing protocol is crucial in designing an e cient routing protocol. There have been many mobility models and routing protocols de ned till date, and this paper has designed and identi ed the e cient models.DTN in the opportunistic network in Sec. 3. Next secs. 4 and 5 will discuss the taxonomy of mobility models and routing protocols, respectively in detail. Sec. 6 describes how to evaluate the performance of mobility models and routing protocols. Sec. 7 will nally provide the conclusion.
Related WorksOppNets [4], a sub-part of DTN, use a store-carry and forward approach to continue the successful communication between two nodes. A continuous connection can be interrupted due to a variety of factors such as limited battery lifetime, lack of capacity, abrupt change of network topologies and routes, and lack of mobility pattern. In DTN, there are a lot of routing protocols [2] and mobility models [4] that will be discussed in this research paper more elaborately. This paper contains all the proposed and implemented strategies for mobility models discussed in Section 4 and for routing protocols in Section 5. The stochastic mobility models are random walk, random waypoint, and random direction. In the traditional random walk (RW) [6] model, the nodes use random speeds for random direction to move in. Random waypoint model [6] uses pause time for newly considered destination. In a random direction model [6], nodes travel randomly until gets an edge. Levy walk model uses power-law distribution to select the pause time used in the random waypoint. The synthetic models, called map-constrained mobility models use the random pattern within the map. Among these models, the most commonly used is the shortest path-based map mobility model [6] where Dijkstra's shortest path strategy is used. In map-based mobility, nodes move with random speed and directions depending on the map. Route-based map mobility uses prede ned paths on the map. Routing protocols can be classi ed based on different categories, geographic, link-state aware, context-aware, probabilistic, optimization,...