**3. Simulation and comparison**

256 Real-Time Systems, Architecture, Scheduling, and Application

The origin distributes a QRY packet in which destination node identifier is located. In this method, a non-circular graph is created from origin to destination. Figure 8 indicates a process of route formation in TORA. As shown in Figure 8 (a), node 5 receives the QRY packet from node 3 but it doesn't publish it because this packet has reached this node through node 2 previously. In Figure 8 (b), the origin, i. e. , node 1 can receive the UPD packet from node 2 or

In these protocols, clusters are formed by dividing the whole network into self-managed groups of nodes. These groups are dynamically rearranged when the topology of the network is changed. To form these clusters, the following algorithm is used. When a node enters the network, it enters an indefinite state. In this state, it adjusts a timer and distributes a Hello message for all other nodes. When a cluster head receives this Hello message, it replies with a Hello message immediately. When the unknown node receives this message, it changes its state to member. If the indefinite node does not receive a reply after the defined time, it introduces itself as a cluster head in the case that it has a two-sided

node 3 but it doesn't receive it from node 4 of which the height is lower.

a)Broadcast QRY

b) Distribute UDP packet

Fig. 8. Detection of route in TORA

**2.3.4 CBRP (Cluster based Routing Protocols)** 

Since energy consumption during communication is a major energy depletion parameter, the number of transmissions must be reduced as much as possible to achieve extended battery life. As a result, there is a strong need for the presence of protocols which use energy efficiently and effectively as well as technology for better management of energy. Unfortunately, battery technology doesn't grow as rapidly as CPU or memory does [Akkaya. K et al 2005]. In this section we would like to analyse comprehensively the results of the simulation for mobile ad hoc routing protocols in some aspects and different scenarios of energy consumption such as movement model, traffic model, data-sending model, environment and amount of nodes. To compare the protocols [Qasim. N et al 2009], a simulation of a MANET is required to be performed in a Network Simulator [Qun Z. A et al 2011] environment. The simulation includes motion models, a physical layer with a radio broadcast, radio network communications and the IEEE 802. 11 (MAC) protocol [Chowdhury et al 2010] with a distributed coordination equation or DCE. This model includes collisions, broadcast delays and signal damping with a bandwidth of 2 Mbps and a transmission range of 250 m. There are some protocols that have been selected for comparison, such as DSR, AODV, DSDV, TORA, FSR, CBRP and CGSR. The first four protocols have been completely simulated by NS2 [26], and the code of other protocols has been added to NS2. [Stemm. M and Katz. R 1997].

The basic model parameters that have been used in the following experiments are summarised in Table 1.


Table 1. Simulation parameters
