Research On Entity Mobility Models Of Ad Hoc Network

Mobility model is the foundation for simulation of all kinds of protocols in Ad Hoc network. It used to describe the character of mobile node. In order to set up a effective simulation framework and analyze the simulation results accuratly,it is very important to research the mobility model.Entity mobility model is an important branch of mobility model.Because of simple and implement easily,it is widely used in simulation of Ad Hoc network.The MRP model is mobility pattern with the characteristic of memory.Compared to other Entity mobility models,it has significant advantages in stability and realistic.Starting with MRP mobility model,the entity mobility model is studied deeply in this paper.The main results obtained are as follows:(1) The spatial probabilistic distribution of MRP model is presented. Firstly, According to describe the pattern of moving and mathematical derivation ,the spatial probabilistic distribution is analyzed both in situations of 1-dimension and 2-dimensi- ons. Secondly, impacts of the parameter q as well as transfer steps of the spatial probabilistic distribution are studied. Finally, according to simulation, each theoretical result on MRP model obtained in this paper is validated.(2) According to simulation,how different entity mobility models impact on the performance of Ad Hoc network protocol is studied . Firstly,with the help of the NS-2 simulator, how different entity mobility models impact on the performance of Ad Hoc network protocol when they set same speed are analyzed. Finally, how different entity mobility models impact on the performance of Ad Hoc network protocol when they set same pause time are analyzed.(3) A new entity mobility model is proposed.For the weakness of the MRP mobility model, i.e., there is no pause time and has only single speed in simulation, a new entity mobility model named TSMRP (Time-Speed Markov Random Path mobility model) is proposed. The TSMRP model is based on Markov chain and can reflect the realistic node movement more closely. After detaily describe the algorithm of this mobility model, a simulation comparison with the MRP model and the RWP model is presented.