Design And Implementaion Of Wireless Channel In Network Simulation Platform

Wireless sensor network is composed of nodes which are subject to random distribution in the monitoring area, and it can be applied to environmental monitoring, military, industrial control and some other fields.The size and complexity of the network increases along with the increase of the demand of the wireless sensor network monitoring application, and it put forward higher requirements for large-scale wireless sensor network simulation tools.Complex wireless channel leads to randomness and uncertainty of the node communication link,which directly affect the network simulation evaluation process.The most researches are focus on the protocol and energy consumption,and the channel design are mostly based on the ideal channel model.COOJA, as a new wireless sensor network simulation platform in recent years, can be efficiently simulate most wireless sensor network covering hundreds of nodes. But due to its channel model does not accurately reflect fading characteristics of wireless channel in real environment, it is particularly important to make some researches on the channel models for wireless sensor network.This article first introduced the research status and application problems of wireless channel model,and analyzed insufficient of channel model in COOJA and put forward the possibility of developing new channel model.This paper studies the existing classical channel model setted up based on statistical rule of the measured data,and analyzed their respective applicable scenario. This paper studies the channel which is composed of path loss and random component. If channels exist line-of-sight propagation, it is subject to Rice distribution. Construct Rice distribution combined Clarke model’s simplified thought with Jakes model’s simplified thought. The result of simulation verified the thought has the statistical properties of Rice distribution, good stability, Ergodic property, small amount of calculation and some other characteristics. Path loss use Plane Earth model, and gave the correction factor through by compared with the measured dual slope model in different environments, and analyzed its applicable conditions.Finally,this paper design a channel model adapted to flat cement,flat sand, flat grassland, low thickets and forest environment five different environments. But because of the last two environment will be affected by line-of-sight component, Plane Earth model cannot make universal description, thus using dual slope model design. Finally, this paper designed and implemented five kinds of channel models based on the further study on architecture and simulation mechanism of the large scale simulation platform COOJA, and completed accepting strength prediction, packet error rate calculation and some other functions. Integrate the channel model into the original channel in COOJA simulation platform, there exits large difference to the original channel prediction model prediction.We established wireless sensor network scenario covering 500 nodes. Because of the calculation complexity of the designed channel is higher; the overall simulation efficiency is slightly lower. But the channel prediction model is more rationality, and it can be one of choices of the channel models for COOJA wireless sensor network simulation.