1.2.7.2 Demand-driven (on-demand)

In the demand driven, When needed to send packets only it began to prepare to send the routing table. When a wireless node needs to send data to another wireless node, the source client node will call a path discovery process, and stored in the registers of this path. The path is not valid until the expiration or the occurrence of conditions of the agreement with the first phase of a ratio of such agreements in each node. A smaller amount of data needed, and do not need to save the entire network environment and the routing information. The main benefit of this agreement is that the use of a lower bandwidth, but the drawback is that not every wireless node that sends packets can always quickly find the path. The path discovery procedure can cause delays and the average delay time is longer (Liu and Chang, 2009).

### 1.2.7.3 Hybrid

Hybrid is an improvement of the above mentioned two, or the combination of other equipment, such as Global Positioning System (GPS) and other equipment, participate in the study of mechanisms to facilitate the routing of the quick search, and data transmission (Pandey et al., 2005) (Johnson and Maltz, 1999.). However, there are already more than 13 kinds of the above routing protocol have been proposed, following the more representative for several separate presentations, and to compare their individual differences lie.

#### **1.2.8 Compare between proactive versus reactive and clustering versus hierarchical**

#### 1.2.8.1 Proactive versus Reactive Approaches

Ad hoc routing protocols can be classified into two types; proactive and On-Demand (reactive) base on each own strategy (Perkins, 2001). Proactive protocols demand nodes in a wireless ad hoc network to keep track of routes to all possible destinations. This is important because, whenever a packet requests to be forwarded, the route is beforehand identified and can be used straight away. Whenever there's modification in the topology, it will be disseminated throughout the entire network. Instances include "destination-sequenced distance-vector" (DSDV) routing (Perkins and Bhagwat, 1994), "wireless routing protocol" (WRP) (Murthy and Garcia-Luna-Aceves, 1996), "global state routing" (GSR) (Chen and Gerla, 1998), and "fisheye state routing" (FSR) (Iwata et al., 2002)and in next section will discuss about everyone.

On-demand (reactive) protocols will build the routes when required by the source node, in order for the network topology to be detected as needed (on-demand). When a node needs to send packets to several destinations but has no routes to the destination, it will start a route detection process within the network. When a route is recognized, it will be sustained by a route maintenance procedure until the destination becomes unreachable or till the route is not wanted anymore. Instances include "ad hoc on-demand distance vector routing" (AODV) (Perkings et al., 2003), "dynamic source routing" (DSR) (J.Broch et al., 2004), and "Cluster Based Routing protocol" (CBRP) (Jiang et al., 1999). Proactive protocols comprise the benefit that new communications with arbitrary destinations experience minimal delay, but experience the disadvantage of the extra control overhead to update routing information at all nodes. To overcome with this limitation, reactive protocols take on the opposite method by tracking down route to a destination only when required. Reactive protocols regularly utilize less bandwidth compared to proactive protocols, however it is a time consuming process for any route tracking activity to a destination proceeding to the authentic communication. Whenever reactive routing protocols must relay route requests,it will create unnecessary traffic if route discovery is required regularly.

#### 1.2.8.2 Clustering versus hierarchical approaches

14 Will-be-set-by-IN-TECH

when needed. However, such agreements must be periodically to broadcast messages, so a considerable waste of wireless bandwidth and wireless node power, but if you want to reduce the broadcast bandwidth consumption caused by a large number, we should lengthen the interval between each broadcast time, which in turn will result in the path table does not

In the demand driven, When needed to send packets only it began to prepare to send the routing table. When a wireless node needs to send data to another wireless node, the source client node will call a path discovery process, and stored in the registers of this path. The path is not valid until the expiration or the occurrence of conditions of the agreement with the first phase of a ratio of such agreements in each node. A smaller amount of data needed, and do not need to save the entire network environment and the routing information. The main benefit of this agreement is that the use of a lower bandwidth, but the drawback is that not every wireless node that sends packets can always quickly find the path. The path discovery procedure can cause delays and the average delay time is longer (Liu and Chang, 2009).

Hybrid is an improvement of the above mentioned two, or the combination of other equipment, such as Global Positioning System (GPS) and other equipment, participate in the study of mechanisms to facilitate the routing of the quick search, and data transmission (Pandey et al., 2005) (Johnson and Maltz, 1999.). However, there are already more than 13 kinds of the above routing protocol have been proposed, following the more representative

for several separate presentations, and to compare their individual differences lie.

**1.2.8 Compare between proactive versus reactive and clustering versus hierarchical**

Ad hoc routing protocols can be classified into two types; proactive and On-Demand (reactive) base on each own strategy (Perkins, 2001). Proactive protocols demand nodes in a wireless ad hoc network to keep track of routes to all possible destinations. This is important because, whenever a packet requests to be forwarded, the route is beforehand identified and can be used straight away. Whenever there's modification in the topology, it will be disseminated throughout the entire network. Instances include "destination-sequenced distance-vector" (DSDV) routing (Perkins and Bhagwat, 1994), "wireless routing protocol" (WRP) (Murthy and Garcia-Luna-Aceves, 1996), "global state routing" (GSR) (Chen and Gerla, 1998), and "fisheye state routing" (FSR) (Iwata et al., 2002)and in next section will discuss about everyone.

On-demand (reactive) protocols will build the routes when required by the source node, in order for the network topology to be detected as needed (on-demand). When a node needs to send packets to several destinations but has no routes to the destination, it will start a route detection process within the network. When a route is recognized, it will be sustained by a route maintenance procedure until the destination becomes unreachable or till the route is not wanted anymore. Instances include "ad hoc on-demand distance vector routing" (AODV) (Perkings et al., 2003), "dynamic source routing" (DSR) (J.Broch et al., 2004), and "Cluster Based Routing protocol" (CBRP) (Jiang et al., 1999). Proactive protocols comprise the benefit that new communications with arbitrary destinations experience minimal delay, but experience the disadvantage of the extra control overhead to update routing information at all nodes. To overcome with this limitation, reactive protocols take on the opposite method by tracking down route to a destination only when required. Reactive protocols regularly utilize less

accurately reflect network topology changes.

1.2.8.1 Proactive versus Reactive Approaches

1.2.7.2 Demand-driven (on-demand)

1.2.7.3 Hybrid

Scalability is one of the major tribulations in ad hoc networking. The term scalability in ad hoc networks can be defined as the network's capability to provide an acceptable level of service to packets even in the presence of a great number of nodes in the network. If the number of nodes in the network multiply for proactive routing protocols, the number of topology control messages will increases nonlinearly and it will use up a large fraction of the available bandwidth. While in reactive routing protocols, if there are a large numbers of route requests propagated to the entire network, it may eventually become packet broadcast storms. Normally, whenever the network size expands beyond certain thresholds, the computation and storage requirements become infeasible. At a time whenever mobility is being taken into consideration, the regularity of routing information updates may be extensively enhanced, and will deteriorate the scalability issues. In order to overcome these obstacles and to generate scalable and resourceful solutions, the solution is to use hierarchical routing. Wireless hierarchical routing is based on the idea of systematizing nodes in groups and then assigns the nodes with different task within and outside a group. Both the routing table size and update packet size are decreased by comprising only a fraction of the network. For reactive protocols, restricting the scope of route request broadcasts can assists in improving the competency. The best method of building hierarchy is to gather all nodes geographically near to each other into groups. Every cluster has a principal node (cluster head) that corresponds with other nodes. Instances of hierarchical ad hoc routing protocols include "zone routing protocol" (ZRP) (Haas and Pearlman, 2000).

#### **1.2.9 Existing ad hoc protocols**

In the ad hoc network there are more than 13 kinds of the above routing protocol have been proposed such as DSDV, GSR, CGSR, WRP, FSR, AODV, DSR, TORA, CBRP, ABR, SSR, CEDAR and ZRP, for more dilates about existing ad hoc network protocols (Saleh Alomari and Putra Sumari, 2010. Further explination for understanding some of the existing mobile ad hoc network are provided in Appendix A figure 10. The comparison between Table Driven, Demand Driven and Hybrid are shown in Table 2,and then show in Table 3 the Table Driven for three kind of protocols such as WRP, CGSR, DSDV and comparison between them, Demand Driven (On-Demand) with six type of protocols such as TORA, DSR, AODV, ABR, CEDAR and SSR and comparison between them shows in Table 4. Finally, shows compare the main characteristics of existing multipath routing protocols in Table 5.

\* CEDAR, TORA itself, although it can not also be used in multicasting, but there have been constructed in the two above the multicast routing protocol was proposed.

#### **1.2.10 Challenges and issues of MANETs**

For ad hoc networking design and implementation, there lots of factors and challenges which are:

**Scalability:** in some applications, a MANET can grow to thousands of nodes, such as, battlefield deployments, urban vehicle grids and large environmental sensor fabrics. It is extremely hard to have the scalability handled in a MANET due to the random and unlimited mobility (Perkins et al., 2002).


Table 2. Illustrates the comparison between Table Driven, Demand Driven and Hybrid


Table 3. Shows the Table-Driven for the three kinds of protocols and comparison between them

**Mobility** is at most the first designer's enemy of MANET (Murthy and Mano, 2004).


The main key issues that affect the design, deployment, and performance of an ad hoc wireless system are summarized as following: scalability, security, energy management, QoS provisioning, deployment considerations, self organization, multicasting, pricing scheme, medium access scheme, routing, transport layer protocols, addressing and service discovery.


Table 4. Shows the Demand Driven (On-Demand) with six types of protocols and comparison between them


Table 5. Shows the comparison of the main characteristics of existing multipath routing protocols

The four important issues significant in MANET are Mobility, QoS Provisioning, Multicasting and Security.

#### 1.2.10.1 Mobility

16 Will-be-set-by-IN-TECH

**Routing Protocols** DSDV,CGSR,WRP AODV,DSR,TORA,ABR,SSR **ZRP**

**Control overhead** High Low Medium **Power requirement** High Low Medium

**requirement** High Low Medium

Table 2. Illustrates the comparison between Table Driven, Demand Driven and Hybrid

**Table Driven CGSR WRP DSDV Routing philosophy** Hierarchical Flat Flat **Loop-free** Yes Yes, but not

**Number of required tables** 242

**Utilize hello messages** No Yes Yes **Critical nodes** Cluster head No No **Communication complexity** O(x = N) O(x = N) O(x = N)

Table 3. Shows the Table-Driven for the three kinds of protocols and comparison between

**Energy conservation** most ad hoc nodes, such as Personal Digital Assistants (PDAs), sensors and Laptops are often power supplied using batteries which have limited power. Therefore, for MANET, energy conservation is considered to be an enormous challenge. **Application/Market penetration:** multi-hop technology is not commercial at present. More clearly, the short coverage area's limitation of the wireless products can be justified in its

**Design/Implementation:** manageable, secure, reliable and survivable implementation and design must act for MANET since a bandwidth-constrained operation and a limited

The main key issues that affect the design, deployment, and performance of an ad hoc wireless system are summarized as following: scalability, security, energy management, QoS provisioning, deployment considerations, self organization, multicasting, pricing scheme, medium access scheme, routing, transport layer protocols, addressing and service discovery.

**Limited wireless transmission range** depends on the wireless technology's capabilities. **Operational/Business-related** how to have the network managed and how to bill for

**Mobility** is at most the first designer's enemy of MANET (Murthy and Mano, 2004).

**transmissions** Periodically Periodically and as

**Driven(Proactive) Demand Driven(Reactive) Hybrid**

instantaneous Yes

needed

head Neighbors Neighbors

Lower for Intra-zone; Higher for Inter-zone

Periodically and as

needed

**Table**

**delay** Lower Higher

**Updates transmitted to** Neighbors and cluster

belonging to the standard of IEEE 802.11.

physical security are contained in MANETs.

**Route acquisition**

**Frequency of update**

**Bandwidth**

them

services.

The mobile user can freely move anywhere and are free to join and move away from the network at anytime. The mobile client can explore the area and can form groups or teams to create a taskforce. In the ad hoc network, the mobile client can have individual random and group mobility and the mobility model can have major impact on the selection of a routing scheme and this directly influences the performance. The mobile clients in MANETs have no physical boundary and their location changes as they move around. This movement of mobile nodes makes the network topology highly dynamic as well as causing the intercommunication patterns between nodes to change frequently in an unpredictable manner (Frodigh et al., 2000), (Satyanarayanan, 2001). Thus, an ongoing communication session suffers frequent path breaks. As a result, broadcasting protocols for MANETs must handle mobility management efficiently (Basagni et al., 1998).
