Data Mining Research And Application In Wireless Network Planning

The Ministry of Industry and Information of China issued4G licenses to China Mobile, China Telecom, China Unicom on December2013, which means China has started a new round of wireless communication network’s construction and upgrade. One of the most important keys of the network development is radio network planning. The traditional radio network planning is limited by the poor condition of data gathering, it usually uses the road test data and incomplete geographic information to do the research, so the results are inaccurate and cannot meet the requirement of the network service with high qualities. Because the radio environment is becoming more and more complicated, the dated network planning cannot satisfy the requirement, it is imperative to improve the wireless network planning methods.In this paper, in order to improve the existing wireless network planning, we have studied the application of data mining technology in the network planning. We use the data mining tool SPSS to analyze and design the UNet wireless network planning scheme. We use SPSS to analyze the output data of each subsystem of the UNet, then feedback the analysis results to the UNet to improve the depth and accuracy of data analysis of the network planning. The accuracy of the propagation model system’s wireless signal transmission link loss prediction is important to the effectiveness of the whole wireless network planning, so this paper utilizes cluster analysis and correlation analysis to deal with the data of the transmission model system to improve the calculation accuracy of propagation model system.The paper has designed a propagation environment clustering analysis solution based on covering power. The solution can efficiently extract the geographic information and divide the wireless network coverage area into different sectors. It designs an analysis algorithm for sector cluster, and uses SPSS software to complete the clustering analysis to create the sector cluster with the similar propagation environment characteristics. Thus we can choose the appropriate propagation model for each sector, and guarantee the accuracy of link loss calculation.The paper has designed a correlation analysis algorithm of transmitting factors and link loss budget error, which can help get the influence of transmitting factors, including propagation distance, antenna height and signal frequency, on the calculation of the link loss. The propagation model based on the correlation analysis algorithm of transmitting factors can provide accurate guide to the correction of the propagation model’s coefficients, which can provide users with right correction opinions.In the end, a conclusion is summarized and the future directions in design and implementation are pointed out.