الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 118 |  |  |
| | | from | 118 | | |
Abstract
Computer networks have become an important technology because they provide resource sharing, computation speedup and reliability. Networks are divided according to the media into wired and wireless networks. Traditional wireless and mobile networks are based on cellular network infrastructures; mobile nodes communicate with centralized entities (i.e., the base-stations), which serve as point of wireless access to a fixed network. In contrast, ad hoc networks represent a different communication paradigm, which is not dependent on any preexisting infrastructure. Thus, ad hoc networks can be rapidly deployed, without prior planning and in unknown radio propagation conditions. A Mobile Ad hoc Network (MANET) consists of a number of mobile hosts that connect to each other through wireless communication and includes many routing mechanisms like unicast, multicast, broadcast. Many broadcasting algorithms have been developed. These algorithms are classified into three categories, these are simple flooding, 0-hop and neighbor knowledge based , ( which can be further divided into neighbor designating algorithms and self-pruning algorithms). Different types of self-pruning algorithms are introduced. These algorithms have the prosperities of overcoming flooding problems (redundancy ,overhead ,efficiency) and 0-hop category problems (reliability, accuracy). Also, they have the advantage of self determining the node status (forward or not forward) by itself. The enhanced generic self-pruning algorithm is the most powerful algorithm but it can only achieve either high reliability or high efficiency. Three algorithms are proposed in this thesis; Reduced Hello Message Interval Time (RHMIT) , 1-Hop Location Information (1-HLI) , and Hybrid Self-Pruning algorithms. These algorithms take account of both reliability and efficiency at the same time. In RHMIT algorithm, the main idea is based on the reduction of time interval between HELLO messages, the hello message and the 1-hop location information message is separated and use timer to detect link failure. The main idea of 1-HLI algorithm is based on different type of 1-hop location information message. When a node sends 1-hop location information message, it contains the exact position of this node. The hybrid self-pruning algorithm combines both RHMIT and 1-HLI algorithms. The performance of these algorithms is tested through a comparative study between them and the most recent one. The results of the comparison indicate that RHMIT algorithm achieves high reliability with acceptable efficiency while 1-HLI algorithm achieves high efficiency with acceptable reliability. While the hybrid self-pruning algorithm has efficiency like the (1-HLI) and has reliability less than the (RHMIT) and better than the (1-HLI).