The Allocation of Base Stations with Region Clustering and Single-Objective Nonlinear Optimization

For the problem of 5G network planning, a certain number locations should be selected to build new base stations in order solve weak coverage problems existing network. Considering construction cost and some other factors, it is impossible cover all areas so necessary consider business volume give priority with high volume. Aimed at these problems, clustering point data was carried out by using k-means algorithm. With objective function as minimization total stations, well constraints minimal distance between adjacent communication traffic, single-objective nonlinear programming models were established obtain layout macro micro illustrate impact shape station area, circular “shamrock” shaped compared this paper. station, secondary undertaken judge main directions three sector areas. Then, an improved model taking overlapping into consideration proposed correct area different sectors. Finally, optimal obtained adjusting distribution globally. The results show that planning method paper has good practicability, which also provides very reference for solving similar allocation dynamic resources.