Clustering Optimized Genetic Algorithm-Based 5G Communication base Station Site Selection

Abstract

To minimize the negative impact of the base station construction, and at the same time make the base station cover as many users as possible, the Genetic Algorithm optimized Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method is proposed. Initially, the number of points is fixed and encoded into a gene sequence with 4 different features at each point. After that, the process of the DBSCAN is stated and then it constructs the cluster adaptation function. Then, with the combination of the cluster adaptation function and the coverage function, it can select the individual, cross the gene sequence, and change the slight value of the sequence. Ultimately, by iteration, it can output the optimal point location and the result shows that the optimal coverage is 95%.