Research On Neighbor Discovery In Ad Hoc Networks With Directional Antennas

Directional antenna has a lot of application in the wireless ad hoc networks topology control and topology management process.By configuring the directional antenna in the ad hoc networks,it can not only improve the transmission performance of the network as a whole,but also bring great anti-jamming and anti-interception performance,and it can also make the networks get more optimal performance in the military communication,the emergency communication,the personal communication and some temporary occasions.However,directional antenna is a kind of physical layer device,to give full play to its various advantages,it must solve the complex question of the neighbor discovery and network topology control.Therefore,this paper studies the neighbor discovery and topology control of wireless ad hoc networks with directional antennas.By studying the mechanism of improving the neighbor discovery efficiency,we can reduce the time of neighbor discovery and improve the network communication performance.In this paper,we first construct and analyze the model of neighbor discovery in directional wireless ad hoc networks.We propose a scheme that performs a different search process in every main lobe of the directional antenna,and sets different lookup probabilities for different main lobe beam so as to fast convergence of discovery time.The discovery probability and average time of each process are analyzed,and the theoretical model of discovery probabilities and the average time is given for directional wireless ad hoc networks.Secondly,this paper proposes a neighbor discovery mechanism for wireless ad hoc networks with directional antenna,that is,a directional neighbor discovery method based on probability optimization.Firstly,by setting the probability of beam selection for each node and adjusting the value periodically in the process of neighbor search,it can ensure that the neighbor scanning is performed at the most needed beam,and the number of the neighbor searching direction is reduced.Therefore,the protocol reduces the overall neighbor search time and improves the search efficiency.Secondly,by changing the slot length of the sending and receiving process in a neighbor searching period,this protocol can reduce the information collision and improve the probability of successful reception of the information,so the search efficiency is further improved.In addition,this protocol transmits a special format of the control information,so that there is no need to establish a handshake between nodes by sending a special kind of HELLO information,thus reducing the type of control information and the information interaction and making the search process more simple and efficient.Finally,this paper designs and implements the simulation model of the directional neighbor discovery of wireless ad hoc networks using OPNET software.And we validate the performance of the proposed neighbor discovery scheme.The results show that the scheme can effectively improve the efficiency of neighbor discovery in different network scenarios.At the same time,this paper compares with the random neighbor discovery protocol and the NDSA neighbor discovery protocol,finding that the scheme can get better neighbor discovery performance.