Danladi A. Models and methods of traffic generation wireless networks and assess their performance

The work is devoted to constructing mathematical models and methods of traffic generation wireless networks and assess their performance. Improved multilevel ON-OFF-model for aggregated data stream, which has a Pareto distribution. The numerical simulation shows that the total traffic of wireless networks has substantial self-similar properties, is estimated Hurst exponent and a lot of dependent random process. The regularities that allow to determine the degree of self-similarity of traffic, which in turn allows us to estimate the parameters of the incoming flow of real computer systems. A simulation of existing routing protocols in wireless networks, such as DSR, GRP, OLSR. The efficiency of routing protocols based on criteria such as the number of sent and received traffic, bandwidth utilization, throughput, delay, lost data access latency to the environment. The results obtained allow to use wireless network resources by proper selection of the routing protocol. For the first time developed a mathematical model of the performance of a wireless network, as the main parameter that uses the share of refusals of requests. The analytical expressions can facilitate the construction of labor-intensive simulations for large-scale wireless networks. The results of the research put into practice in the learning process.