摘要： 根据无线网络业务数据的特点，提出一种基于模糊聚类的数据填补算法。该算法考虑数据的周期性特点，并运用沉积的历史数据，对数据进行聚类分析得出标准模板，采用标准模板对缺失数据进行填补。利用该算法对小区小时话务量数据进行填补。实验结果表明，该算法平均误差率较小，无偏性较好，在数据缺失较为严重的情况下也能完成填补。

关键词: 数据填补, 模糊聚类, GSM网络优化, 聚类分析

Abstract: According to the features of wireless network data, this paper puts forward a data complement algorithm based on fuzzy clustering. This algorithm takes data periodicity into consideration and uses deposit history data to complete missing data, clusters the deposit data to make several standard templates and uses the templates to fill in missing data. This algorithm is applied to fill in cellular hour traffic. Experimental results show that it has a small average error rate and good unbiasedness, it can still complete data under severe missing circumstance.

Key words: data complement, fuzzy clustering, GSM network optimization, clustering analysis

中图分类号: 

• TN929.53