Research On Green Deployment Of Micro Base Stations For Heterogeneous Cellular Networks

Ultra-dense heterogeneous cellular network is one of the key technologies of 5G,which has attracted extensive attention from scholars due to its characteristics of improving system capacity and service quality of marginal users.Research shows that deploying a certain number of tiny base stations in the network can improve the energy efficiency of the network,but the electromagnetic radiation intensity is also enhanced,which has a huge impact on the environment.Therefore,under the condition of ensuring the service quality of users,it is of practical significance to realize green communication by rationally deploying small base stations.Considering that the deployment of micro-base stations in heterogeneous cellular networks is a constrained multi-objective optimization problem,the constrained multi-objective algorithm is adopted in this paper to optimize the problem,while the convergence and distribution of the existing constrained multi-objective algorithms need to be improved.A large number of experiments have proved that the constrained multi-objective algorithm framework and its evolutionary strategy directly affect the performance of the constrained multi-objective algorithm.In order to better solve the heterogeneous micro base station green deployment problem in cellular networks,this paper will study from the following three aspects.1)In order to improve the convergence speed of the constrained multi-objective algorithm and avoid the population falling into local optimum,the Jumping Dolphin Swarm Algorithm(JDSA)is proposed.The jump step is added,more excellent solutions are retained,and the convergence is accelerated.The strategy of adaptively changing the length of the acoustic wave with iteration and the dynamic position update strategy with the variation factor are proposed to satisfy the evolutionary requirements of the algorithm in different periods.and the premature convergence mechanism is added.Reduce the chance of falling into local optimum.Finally,the influence of parameters on the performance of the algorithm is analyzed and compared with the four algorithms.Experiments show that it has obvious advantages in convergence speed,convergence precision and robustness.2)In order to improve the convergence and constraining distribution of multi-objective algorithms,the Constrained Multi-objective Dolphin Swarm Algorithm(CMO-DSA)is proposed.Firstly,a Harmonic distance-based circular deletion strategy is proposed,which can reflect the distribution of the population more accurately,but makes the screening of individuals at the same level more reasonable.Secondly,a new infeasible solution dominance relationship is proposed,which makes the better feasible solution participate in evolution.Finally,the position update method of the search hunting stage is adjusted,and the selection probability(constant)is changed to a function that gradually decreases with the iteration,which accelerates the convergence speed of the algorithm later.By comparing with the three algorithms in the CTP1-7 series test function,the results show that the constrained multi-target dolphin group algorithm has good convergence and distribution.3)A green deployment strategy for small base stations in heterogeneous cellular networks based on CMO-DSA is proposed.Firstly,the two-layer heterogeneous cellular network topology conforming to the actual application scenario is constructed,and the calculation model of user rate,network energy efficiency and electromagnetic radiation intensity is given.According to the dynamic characteristics of user distribution,the problem model is simplified by using probability weighting method.Secondly,a green deployment algorithm for small base stations in heterogeneous cellular networks based on CMO-DSA is proposed.Finally,simulation experiments were performed in the MATLAB environment.The optimal number of small base stations is determined by analyzing the relationship between the number of base stations and the energy efficiency of the network and the electromagnetic radiation intensity.Under this number,compared with HEEDA algorithm and two-population differential evolution algorithm,the simulation results on MATALB show that the proposed CMO-DSA algorithm can improve the network energy efficiency by up to 18.19% and the electromagnetic radiation intensity is within the safe range compared with the two-population differential evolution algorithm under the condition of guaranteeing the average user rate.