Research Of MANET Broadcasting Algorithm Based On Packet And Neighbor Node Set

A Mobile Ad Hoc Network (MANET) is composed by a set of nodes communicating with each other via multihop wireless links. There are no centre nodes in MANET. Nodes can form a network composed of groups freely. Intermediate nodes forward messages to the nodes which are out of the range in one hop distance. People can communicate with each other by MANET whenever and wherever they want. There are several fields in the research of MANET, such as routing protocol, clustering structure and broadcast protocol. Especially the broadcast protocol that send message to other nodes in MANET is more important than others.Broadcasting is much important in Mobil Ad Hoc Network (MANET). The operation of broadcasting is widely used in the routing protocol of MANET, especially in finding the path from the source node to the destination node. It also could apply in the clustering protocol for selecting the cluster head and maintain the structure of the cluster. Blind flooding is the most simply algorithm of broadcasting. But it could cause a lot of redundant packets and lower the efficiency of the broadcast which will lead to the broadcast storm. There are serial algorithms that were elected to solve that problem. Such as Probabilistic Based Schemes, Area Based Schemes and Neighbor Knowledge Schemes. Those schemes could improve the influence of the broadcast storm. But they also have many disadvantages. They are either complicated or not well in function. So it is very important to design a new broadcast protocol for Ad Hoc network. In this article, we do deeply research on the traditional broadcast algorithm, and find out how those disadvantage acts. Then we provide two algorithms to solve these problems.In sparser maps of MANET, many collisions will occur because of the nodes which adjacent to each other rebroadcast in the same time. That makes the reachability of the network lower. Present a message-partition test broadcasting algorithm. This algorithm focuses on how to raise the reachability. This thought is completely contrary to the traditional broadcast algorithms which try to increase the reachability by lower the redundancy. Thereby reducing the collision between neighboring nodes .This algorithm can take the long message apart to short ones. Short message broadcasting could reduce the collision probability. So we can raise the reachability of every short message in order to increase the reachability of the whole one. That could also lower the average latency at the same time. At last, the algorithm can check the integrity by using inspection mechanism which can also raise the reachability. This algorithm can work with any existing broadcasting algorithm in combination. But the effect in denser maps is not obvious, because the reachability is high enough. Nodes in denser maps could receive many other successful rebroadcast messages after some message collision.In those traditional broadcast algorithms, nodes which succeed in receiving rebroadcast message do not send corresponding message to the source node in order to ensure the channel utilization. If node sends corresponding message, other nodes may also occupy the channel to send data message. The nodes in the range of the message can not receive both of those messages in the same time. General broadcasting algorithms are only focus on how to ensure the coverage of one message. This algorithms work on how to send out a group of messages successfully. Nodes in MANET could check the integrity of the messages in one group after rebroadcast. It can reduce the probability of the collision between the data messages and the corresponding messages. Nodes could also apply rebroadcast messages for several missing messages in one application message. Although the test schema can raise the reachability, the average latency may last longer than the traditional algorithm. This algorithm is suitable for the condition which needs high reachability but does not care the latency.In sparse Ad Hoc Network, the performances of many traditional algorithms are unsatisfactory excepting the Neighbor Knowledge Schemes. The reason is that those schemes can not select the suitable rebroadcast sub nodes. And the cost in maintain the knowledge of the neighbour nodes is high. Present a neighborhood-based adaptive broadcasting algorithm. This algorithm can figure out the best sub nodes to rebroadcast message in the neighbour nodes set. It also could monitor vary of the rebroadcast nodes. So it can adapt the change of MANET.It is difficult to monitor the information of the neighbor nodes and select the best rebroadcast nodes every moment. If the source node does notice the variation of neighbour nodes immediately, it can not send message effectively. Find out the proper rebroadcast nodes can rise up the coverage of the broadcast. The old algorithms of the neighborhood-based broadcasting pay much attention on how to organize the topology and select the rebroadcasting nodes. The focus of this algorithm is on how to maintain the structure of the neighbour nodes. When few nodes move, center node can adjust the topology of the network. There are three methods which could be used in different circumstance. One is the method that collecting the information of the neighborhood. The other is how to select the suitable rebroadcast nodes. The last one is how to adjust the neighbor knowledge. This algorithm could improve the performance of the neighborhood-based broadcasting. It can make the efficiency of broadcasting in MANET higher than before.At last, we use the NS2 software to simulate the message-partition test broadcasting algorithm and neighborhood-based adaptive broadcasting algorithm. Compared with the existing algorithm, new schemes have better coverage and save more time in the same experimental conditions. The new broadcast algorithm could improve the performance of the broadcast in MANET.