Resource Scheduling Under Hybrid Broadcasting And Cellular Networks

With the rapid development of Internet and the popularity of mobile intelligent terminal,applications of users for the data will be the exponential growth trend in the next 10 years.The data grows so fast that the wireless cellular network will be caused great burden.However,according to the great relevance between traffics,we can use the principle of "90/10" or "exponential distribution" to describe this phenomenon which means that 90% of users are only interested in 10% of the hot information,and the remaining 90% of information only attracts 10% of users.Converged broadcasting network and cellular network can take advantage of this phenomenon effectively improve the transfer efficiency of the network.We make use of broadcasting network to transmit hot information,but with the cellular network to transmit the non-hot information.In order to make the performance of the converged network achieve the optimal,this paper studies key techniques of resource scheduling in wireless converged network,and analyzes and simulates its performance.Firstly,this paper redefines the evaluation standard of traditional resource scheduling,four classic resource scheduling algorithm is introduced.Then,we study the performance of converged network based on pushed and cached service on whether including real-time service.Then,we aim at the non real-time hot service pushed and cached converged network to do study.We compare the performance of throughput and fairness in converged network and in pure cellular network under the different hot distribution.Simulations show that the throughput of the model has at least six times of ascension and the user's fairness has as much as 4.52 times of ascension compared with pure cellular network,and the more concentrated the hot distribution,the more obvious improvement of throughput performance and fairness.We study converged network's performance of throughput and fairness under the condition of different pushed ratio of hot service.Simulation shows that the higher pushed hot service ratio,the more obvious the effect of shunting user in converged network compared pure cellular network;simulations also find that under the given conditions,users' energy efficiency can reach maximum when the pushed ratio is 6%?We also further research the performance of converged network under different user requests;when compared with pure cellular network in throughput and fairness,the simulation results show that the more users' requests,the more effective converged network.After that,we study cached converged network included real-time hot service,resource scheduling algorithms are mainly studied.In order to simplify the problem,we first study the performance of without non QoS guarantee real-time service cached convergence network,present a non QoS guarantee resource scheduling algorithm based on hot priority,which is revised RR,Max C/I,PF algorithm.Simulations find that when the radio frequency band resource constraints,revised algorithms have a great difference in throughput.Comparing with pure cellular network,we find that the performance of converged network in throughput and fairness is better with increasing ratio of hot real-time service pushed.Finally,in order to in conformity with the actual situation,we study the performance of QoS guarantee real-time service cached convergence network,present a QoS guarantee resource scheduling algorithm based on hot priority.Compared with pure cellular network performance,simulations show that the throughput has at least 9.5 times of ascension,the user's fairness has as much as 3.9 times of ascension and the retransmission times even is reduced by 93.8%.On this basis,the proposed algorithm is studied under different conditions,such as changing pushed ratio of real-time hot service?change proportion of non real-time hot service pushed ahead?changing proportion of real time hot service in total hot service,performance of converged network.Simulation results all indicate that the proposed QoS guarantee resource scheduling algorithm can ensure better performance of throughput and fairness when the retransmission times are 0.