Queuing Theory Research On Service Quality Of Mobile Ad Hoc Networks

The Mobile Ad hoc Networks consist of a large number of nodes that deliver information and are not controlled by the central device.The information transmitting time among the nodes is usually required to be short,which is realized by the high quality of service,and the energy of mobile nodes mostly comes from the battery.Therefore,the energy consumption of the nodes is necessary to be considered when we design the self-organizing network.Queuing theory is a special subject to study random service systems.We can optimize the system design and control by analyzing the probability characteristics of queuing systems.And it is an effective way to improve service quality.The single-hop network and multi-hop network are studied in this thesis.Based on the idea of queuing theory,cost models were established for the joint optimization problem of network node delay and energy consumption.Some system performance measures were obtained,and an optimal design of the system servers was proposed.First,we study two different strategies for single-hop networks in a packet transmission process.Experiments show that thresholds can reduce the system energy consumption.Second,we discussed the joint optimization of the delay and energy consumption for a mobile node with two types of data packets.A method to optimize the number of servers was proposed.Lastly,for the multi-hop network,the Jackson queuing network model is established to analyze the system performance measure.We get the cost function and study the optimal server number of each node when the cost of the entire network system is the lowest.The numerical results show that the overall optimal server number of the whole network and the local optimal server number of the single-hop network are different.The local optimization does not mean overall optimization.