Research On Performance Analysis Based On Stochastic Geometry And Interference Management For Ultra Dense Networks

With the rapid development of mobile communication,the future 5th generation(5G)mobile communication system will focus on user experience,providing higher transmission rate and connecting more terminals.By increasing the density of small base stations(SBS)per unit area and providing intensive deployment for wireless network infrastructure,ultra dense networks(UDN)can enhance the network coverage and obtain higher spectral efficiency,which has become one of the key technologies to cope with 5G challenges.However,in the actual deployment,UDN presents strong randomness and irregularity,and usually integrates with other 5G key technologies,which brings great challenges to the performance analysis before network deployment and interference management after network deployment.In view of this,for the downlink of dense heterogeneous networks,this thesis focuses on the coverage performance and interference management methods of dense heterogeneous networks based on stochastic geometry.The specific work contents are summarized as follows.1.Analysis of coverage probability in non-orthogonal multiple access heterogeneous networks(NOMA-Het Net)based on Matern cluster process(MCP).Considering the correlation between users and base stations(BS)and the irregularity of the network,integrating with non-orthogonal multiple access(NOMA)technology,the NOMA-Het Net based on MCP is modeled,and the coverage probability of various users is analyzed.In the process of analysis,firstly,the signal to interference plus noise power ratios(SINR)of users are theoretically derived.Secondly,according to the imperfect serial interference cancellation(SIC)in reality,an updating law for the upper limit of SINR is presented.Then,the association probability and the service distance of different types of users are analyzed by using mathematical tools such as spatial coordinate system,probability theory,stochastic geometry theory and order statistics.Finally,the theoretical expression of coverage probability of all kinds of users is derived.The effects of cluster radius,base station density and interference residual factors on service distance and coverage probability are verified by numerical simulations.The results are consistent with the theoretical analysis,which testifies to the correctness of the theoretical analysis.2.Joint interference alignment(IA)and power allocation(PA)interference management scheme for downlink NOMA-Het Net.For the different kinds of interference existing in NOMA-Het Net,IA is used to eliminate the inter cluster interference and interlayer interference,and PA is used to suppress the intra cluster interference.First of all,on the basis of the analysis of feasibility of IA,the two-phase type IA based on interference channel alignment algorithm to eliminate the interlayer interference of macro base station(MBS)to small users(SUE),the small base stations’ precoding vectors to eliminate inter cluster interference of SBS to SUE and the macro users’(MUE)detection vectors to eliminate interlayer interference of SBS to MUE are designed,respectively.Secondly,aiming at maximizing the sum rate of the system,mathematical modeling of the problem is carried out.By transforming the original problem into a convex optimization problem,using Lagrange multiplier method and Karush Kuhn Tucker(KKT)optimality condition,the closed form expression to the optimal solution of the problem is derived.Finally,the simulation results verify the high efficiency of the interference management scheme proposed,and prove the advantages of NOMA in improving the sum rate of the system compared with orthogonal multiple access(OMA)in dense heterogeneous network.