Research On Fast Spectrum Allocation Scheme Based On Graph Theory In High Density Network

With the explosive growth of the number of terminals in the network and the continuous improvement of users' requirements,the 5G have put forward higher requirements on data rate and system capacity.As we all know,the spectrum resources are limited and the cost of spreading spectrum resources is huge.Meanwhile the technology of the physical layer is relatively mature and it can approach the limitation of the Shannon in communication link.So in this way,the system capacity can be increased with little benefit.In order to effectively improve the signal coverage and the quality of service,5G operators deploys a great deal of small cells and hotspots based on original scenario.A new type of network structure has emerged.So it should reuse the spectrume resource which will seriously course the inter-cell interference in dense scenario to ensure the spectrum utilization and improve the quality of service(QoS).Therefore,it is necessary to design a fast and efficient spectrum resource allocation strategy to mitigate the interference problem between cells and effectively improve the system performance.This dissertation mainly studies the spectrum resource allocation problem in high-density network scenarios,and proposes a spectrum resource allocation scheme based on graph shading theory and the maximnal independent set theory.This shceme can optimize the system capacity and reduce time complexity,and at the same time,teh fairness of the system can be guaranteed.Graph coloring theory is widely used in the research of resource allocation.In this dissertation,we first explore the optimal allocation of spectrum resources by the simulated annealing algorithm to solve the graph coloring problem and optimize the system capacity.In order to ensure system fairness,we introduce a fairness factor,Search based on optimal utility function.Due to the high computational complexity of the simulated annealing algorithm,computational time is very high in a high-density network scenario,which is not applicable in actual scenarios.Therefore,we simulate a large number of random scenarios,summarize the simulation results,and design a more rough spectrum resource allocation scheme.Compared with the simulated annealing algorithm,the distribution scheme has a certain gap with the optimal distribution result,but the calculation speed is very fast.Based on the above research,this dissertation farther proposed a spectrum resource allocation scheme based on maximal independent set.First,by referring to the maximal independent set theory in graph theory,the small cells are grouped by designing reasonable distance interference thresholds,and the interference between any two cells in the group is as small as possible.In order to improve system performance,we design the empirical fromula of spectrum resource allocation based on a large number of simulation results,and allocate resources directly to different groups.The simulation results show that this scheme is very close to the approximate optimal solution found by the simulated annealing algorithm,which is superior to the classic Max K-cut scheme in terms of system capacity and fairness.At the same time,it shows great advantages in time complexity,this enables rapid spectrum resource allocation in high-density network scenarios.