| Ultra Dense Network(UDN)is a kind of technology to improve the system capacity through intensive deployment of various low power small base stations(SBS).Owing to the characteristics of UDN technology such as overcoming the blind zone problem,realizing seamless connection,alleviating data transmission pressure of Macrocell base station and providing a degree of configuration flexibility,a majority of operators and communications equipment manufacturers pay more attention to UDN technology,which is also listed as one of the key techniques in future 5G communication systems at the same time.Although the path loss between base station and the user equipments has been reduced under the circumstance of such dense SBS deployment,the interference signal will increase with the enhancement of beneficial signal,which can affect the performance of the communication system and the service quality of user equipments.In addition,it will inevitably result in a large amount of energy consumption under the premise of wide-scale deployment of SBS.Therefore,how to improve the system energy efficiency and coordinate inter-cell interference becomes an urgent problem under the UDN deployment.Based on the problems mentioned above,the purpose of this thesis is to find a green and environmental method for coordinating inter-cell interference in UDN.The research is of great significance for improving system capacity and spectral efficiency,reducing system energy consumption,ensuring effective communication and increasing system stability under the complex network environment.The main works of this thesis can be summed up as follows:1.The causes and effects of inter-cell interference in UDN are analyzed and discussed,and two ways of interference coordination in UDN are summarized: Random Interference Coordination(RIC)and Topology-based Interference Coordination(TIC).Besides,the two methods are compared and analyzed in terms of signal to interference plus noise ratio and data transmission rate under the UDN system model.Simulation results show that the performance of TIC method is better than RIC method.2.In order to further keep up with the requirements of green communication in 5G,a SBS load prediction method based on chaos theory and RBF neural network is designed and a SBS sleep strategy is correspondingly proposed on the basis of load prediction.Finally,the optimal sleep time of SBS can be obtained according to partially observable markov decision process(POMDP)model.The MATLAB results illustrate that the load prediction method in this thesis can accurately learn and predict the SBS load in the future cycles and the overall system model can also enhance the system energy efficiency.3.Aiming at solving the problem of complex interference in UDN,a subchannel allocation algorithm based on artificial bee colony coloring clustering is put forward under the energy saving system model.Coloring clustering algorithm can use the least colors to dye the cells in interference topology and then the cells dyed the same color can be divided into the same cluster.Finally,the power of active SBS is optimized and the resource is allocated based on the optimal channel gain,which can maximize energy efficiency and spectral efficiency.The interference coordination algorithm in this thesis does not color the cell in sequence,which is suitable for the actual scenario.Moreover,the efficient subchannel allocation algorithm can further improve the spectral efficiency and system throughput.4.Considering the co-layer interference and cross-layer interference under the three layers UDN coverage(Macrocell,Picrocell,Femtocell),a two-level interference coordination algorithm is proposed.Firstly,the minimum throughput is used to allocate subchannels for Macrocell users and Picrocell users,and power optimization is done for the Picrocell base stations.Finally,the improved K-Means is used to cluster the Femtocell base stations,and subchannel allocation algorithm is proposed according to the priority of Picrocell users. |
PDF Full Text Request