Research On Sighting-aware Obstacle Mobility Model In MANET

In Mobile Ad Hoc network (MANET), the node’s moblie features will affect thenetwork’s performance directly. Construct a real and reasonable mobility model tosimulate node’s moving process has become an important approach for researchingMANET and also has significance for correctly assessing the performance (such asrouting protocol, transmission performance) of MANET.The assumption environments given by some existed mobility models have notconsidered obstacles in node’s moving space. However, obstacles will affect thenode’s move. So it becomes a new challenge to do research on mobility model in theenvironments with obstacles. This paper assume the sighting-aware obstacle scenariosare: each node has not known the global information of obstacles; each node clusteredand does not move randomly, usually the probability of node belongs to a certainregion (activity area) is higher than it belongs to other areas, and the nodes’ movingpatterns are also similar in the activity area.This paper constructed sighting-aware obstacle mobility models, according to thedifferent movement characteristics of individual nodes and cluster nodes. The mainworks are as follows: At first, based most mobile devices are carried and controlled bypeople in MANET, through analyzing the main features of node’s moving in theenvironments with obstacles, proposed a Sighting-aware Obstacle Mobility (SOM)model by combining with the thought of motion planning. The SOM model is tosimulate the node moving in emergency (such as military action and medical aid, etc).The second, According to the similarity of moving patterns between the same typenodes (clustered), proposed a Sighting-aware Obstacle Mobility Based Activity Area(SOMBAA), which based on the SOM model. In emergent task scenarios (such asrescue), The SOMBAA model is simulate the node’s moving pattern in the emergenttask scenarios (such as rescue) that nodes are clustered and actives in eachcorresponding area. At last, build a mobility evaluation platform using VC++6.0,which is used to analyze the network topology characters of mobility models, discussthe relationships between nodes’ moving and changes of network topology. Byanalyzing the average node density, the average number of link changes and othertopology parameters, to verify the effectiveness of SOM model and SOMBAA modelabout the mobile features of pathway selection and destination selection. Expand NS-2to simulate network performance for mobility models. Using the results ofpackets delivery rate, end to end delay and control overhead to evaluate the influencesof routing protocol performance by SOM model and SOMBAA model so as to providereference for selecting an effective mobility model.