In order to meet the communication needs of underground space, it is necessary to build an underground communication network. Taking the data of a region as an example, considering the construction cost, the communication expense, equipment and lines carrying capacity and the characteristics of underground communication, the model of the problem was analyzed. By using the iterative judgment of position and quantity of information flow, k-means clustering algorithm, genetic algorithm and other methods, we established LANs and high-level nodes, as well as the trunk line and the branch line. The process of designing the underground network took into account many factors. And the final results meet the actual requirements and the common sense. It provides a reference for similar problems in the future.
[1]
Richard A. Notebaart,et al.
Network-level architecture and the evolutionary potential of underground metabolism
,
2014,
Proceedings of the National Academy of Sciences.
[2]
Guangzhu Chen,et al.
Design and performance analysis of wireless sensor network location node system for underground mine
,
2009
.
[3]
Xu Shi-he.
Type selection design of computer network structure
,
2003
.
[4]
Dimple Malik,et al.
Evolving limitations in K-means algorithm in data mining and their removal
,
2011
.
[5]
Daeryong Lee,et al.
Modified K-means algorithm for vector quantizer design
,
1997,
IEEE Signal Processing Letters.