Research On Mobility Model Based On Human Mobility Feature For Mobile Ad Hoc Network

Mobile Ad Hoc Network (MANET) is a collection of wireless mobile devicesforming a multi-hop and self-organized wireless network without using any existinginfrastructure. MANET has the merits of flexibility, fast networking andanti-destruction. It is applied to military affairs and disaster-saving fields broadly.The research of MANET relies almost exclusively on simulation. Mobility Model(MM) is an important component of simulator, which is used to describe themovement patterns of nodes. The reality of MM can heavily affect the reliability andvalidity of the evaluation of the routing protocol.In reality, mobile devices are usually carried by humans, so the movement of such devices has the mobility features of human movement. However, the majority ofexisting mobility models for MANET which are widely used nowadays didn’t take in consideration the human mobility features, leading to its lack of accuracy. To solve this problem this paper consider to analyze the real trace date to detect the humanmobility feature. We have found the feature about flights. This article based on the analysis of trace data from real scenarios, proposes a Weight Way Point Mobility Model (WWP) based on human movement features. The main contribution of this paperis built around the four following aspects:i. By summarizing the available trace date, we choose HAGGLE andNCSU-GPS these two trace date sets to analysis, we proved that networknodes has the character of human movement in three aspects: contact time、inter contact time and flights, discovering the feature of the distribution offlights, and this discovery as the basis to design our mobility model.ii. By studying the crucial characteristics of human movement, this paperproposes the WWP mobility model, which is based on RWP mobility model,combined with human movement characters.iii. Using Java and MATLAB we have realized RWP and WWP, designing aMobility Model simulation platform that can generate the trace file which isneeded by NS-2networking simulator.iv. By analyzing the characteristics of WWP mobility model, comparing themobility features of nodes with RWP mobility model and reality, testing thehuman-like movement of WWP mobility model mobile nodes,the resultsproves that WWP mobility model works better in reflecting the nodes’ real movement.