Answers are grounded in this paper and cite their sources.
Driven by the exponential surge on high data rate services, network operators are facing the challenges of how to enhance the capacity and optimize the coverage in a cost-efficient approach. However, traditional network optimization technologies passively adjust the network configurations based on network's congestion ratio, drop-off rate, coverage holes, etc., leading to suboptimum user experiences. Therefore, the objective of this paper is to optimize the network configurations by obtaining the accurate network status, user demand, and application request distribution based on the real-time data. The data mining technique is introduced to predict the resource margin based on historical measurement statistics. To explore the dynamic distribution of user demand and application request, a weighted k-nearest neighbors model is proposed to predict periodic characteristics of network traffics, denoting different temporal and spatial patterns of radio resource margins. In contrast to the traditional passive network optimization approaches, the radio resources can be reconfigured actively to meet the dynamic patterns of traffic loads by using the proposed optimization algorithm. Results prove that the proposed data mining model can capture the dynamics of traffic loads to optimize the traffic load balance and increase the efficiency of radio resource utilization using the network statistic data.
No takes yet. Share an insight, caveat, or question.
Feng et al. (2017) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: