Research And Design On Obstacle-aware Avoidance Mobility Models In Mobile Ad Hoc Networks

A Mobile Ad hoc Network (MANET) is a collection of wireless mobile nodesforming a temporary network without using any existing infrastructure. The networkis self-organized, which is very useful in such applicants where temporary, fast andon-demand communication needed to be built, and has a broad developmentprospects.Because of the difficulty in deploying real mobile ad hoc networks formeaningful performance evaluation, simulation is a primary tool for studying thesenetworks and their protocols. One of the important parts of mobile ad hoc networksimulations is mobility mode which determines the mobility patterns of the nodes inthe network. It is clear that the mobility patterns have a significant impact on thenetwork performance and has a decisive impact on the reliability of simulations.Whether the mobility model can reflect the real characteristics of the nodes’movements is critical for building a comprehensive and reasonable simulationenvironment.In real world obstacles are almost ubiquitous, but obstacles are not taken intoaccount by most of the existing mobility models. They hinder the movements of nodesand affect the propagation of the signals between nodes. What’s more, differentperception about obstacles can cause different movement traces of nodes. Hence, it isimportant to consider the impact of obstacles in different circumstance whendesigning mobility models. The main work and contributions of this paper are asfollows:(1) Analyze the mobility characteristics of nodes in obstacle environment.(2) We present Obstacle-aware Avoidance Mobility (OAM) models for ad hocnetworks. These models are discriminated by the different ways of detecting obstaclesby nodes. They can be applied to simulating the movements in emergency situationslike battlefields and disaster area. In order to imitate the human motion in obstacleenvironment, we draw the idea of motion planning into the OAM models.(3) In the use of several topology connectivity metrics, we analyze the mobilitycharacteristics of OAM models, compared with other obstacle mobility models.Furthermore, by variable parameter settings, the relationship between mobility modelsand topology changes is shown, which measures the effectiveness of the OAMmodels. (4) Building a network platform for mobility models and integrate obstacle into itto impact the communication between nodes. We evaluate the performance of ad hocnetwork protocols by data packet delivery ratio, end-to-end delay and controloverhead. This may offer some references for selecting mobility models. Theexperimental results show that the three OAM models have different impact on thenetwork topology and the performance of routing protocols with other models.