**2.1 Mobile ad hoc network routing protocols**

As shown in Figure 3, routing protocols in mobile ad hoc networks are classified into two classes: table-driven and on-demand [Misra S et al, 2008]. The table-driven d, or proactive, method is used for alternate updating links and can use both the distance vectors and link statuses used in fixed networks. The word "proactive" means that this method is always active and will also react to a change in the linkage [Norouzi. A and Ustundag B. B, 2011]. A problem arises for this method when movement is low because additional work is done and the network tends toward instability [Bai F et al, 2004].

In the on-demand method with reactions, other nodes do not update the route and the routes are determined at the origin of the request. Therefore, there is the possibility of using a caching mechanism. The advantage of this method is that both energy and bandwidth are used effectively. In this chapter, the table-driven and the on-demand protocols are explained and then compared with different parameters [Dixit et al, 2005].

## **2.2 Table-driven protocols**

In this group of protocols, each node maintains one or more tables that include routing information to other nodes of the network. All nodes update their tables to preserve compatibility and to give upgraded viewpoints of the network. When the topology of the network changes, the nodes distribute update messages across the network [Misra S et al, 2008]. Some identifying aspects of this class of routing protocols include the ways in which information is distributed, the ways the topology is changed and the number of tables necessary for routing. The following sections explain some of these routing protocols [Boukerche. A et al, 2011].

Energy Consumption Analysis of Routing Protocols in Mobile Ad Hoc Networks 253

In an FSR, an updating message does not include information about all of the nodes. Instead, it exchanges information with the adjacent nodes with a higher frequency more than it does with farther nodes, leading to a decrease in the size of the updating message [Pei. G et al, 2000]. Thus, each node has accurate information about its neighbours, and the details and accuracy of the information decrease when the distance between two nodes increases. Figure 5 defines the area of a fisheye for a central node that has been indicated in

The central node should know more detail about the nodes that are located inside the white circle. FSR is suitable for massive networks, because in this method, overload is controlled

In comparison with table-driven routing protocols, all updated routes are not maintained in each node in this group of protocols; instead, routes are constructed only when it is necessary. When an origin node wants to send something to a destination, it makes a request to the destination for the route detection mechanisms. For this reason, this type of protocol is known as a reactive protocol. This route remains valid until the destination is accessible. This section explains some of the on-demand routing protocols [Boukerche. A et

This protocol can be regarded as an improvement of DSDV. AODV minimises the number of distributions by creating routes on-demand [Esmaili. H et al, 2011]. In contrast, DSDV maintained a list of all routes to find a route to the destination whenever the origin broadcasts a route request packet to all nodes. The neighbours distribute the packets among their neighbours until the packet reaches an intermediate node with a new route to the

**2.2.3 FSR protocol (Fisheye State Routing)** 

Fig. 5. Accuracy of information in FSR protocol

**2.3.1 AODV (Ad-Hoc On-demand Distance Vector routing)** 

[Liu. Y et al, 2004].

al, 2011].

**2.3 On-demand protocols** 

destination [Song J. H et al, 2004].

red [Boukerche. A et al, 2011].
