Application of Wireless Return Topology Planning Based on K-Means

With the development of communication technology and the general trend of interconnection and intercommunication, people’s demand for high-quality information and communication is increasing day by day. Therefore, the intensive deployment of base stations has become the development trend of the construction of a new generation of communication network. In this paper, in order to solve the urban and rural electric scenarios such as traditional transport unreachable problem, on the basis of the operator site cost at the same time, keep the quality of signal transmission, give full play to the advantages of the traditional planning concept and artificial intelligence algorithm, integrated “pieces”, “step by step optimization”, “local optimal”, “the results of reverse”, such as ideas, set up a wireless back mathematical analysis model of network topology programming problem.

The purpose of this paper is to reduce the cost and reduce the return path loss. Aiming at the lower cost problem, a local optimal model is constructed. First division, through the K - Means algorithm is divided into K, for each region is also based on K - Means algorithm is further split into its n tribal groups, limited each tribal group one and only one host station, recently to tribal group of perimeter centroid position of butterfly stand stand as a host, the rest of the site as a child, according to the type of tribal group of judge whether the number of sub station meet the qualification, if not satisfied, change the K value and the adjustable radius, respectively, until meet the qualification, the best solution is calculated. For the lower return path loss problem, if only the loss problem is considered, the loss result is obtained by changing the distance between the first jump and the second jump based on the model. Through the comparison of the results, the best scheme is obtained when only loss is considered.