Incomplete Data Mining Algorithm for Multi-User Source Wireless Sensor Networks
To improve the integrity of data,an incomplete data mining algorithm for multi-user source wireless sensor networks is pro-posed.The combined generalized morphological filtering method is used to denoise the multi-user source wireless sensor network data to avoid the noise data affecting the data filling results.The integrated learning method is used to deeply mine the data,and the mined data is classified and processed.The low rank matrix filling theory is used to fill the classified data for the first time.On this basis,curve similar classification is introduced to fill the missing data for the second time to accomplish the complete mining of multi-user source wireless sensor network data.The simulation results show that the root mean square error obtained by the proposed method in different data sets is lower than 0.164％,the signal-to-noise ratio is higher than 41.8dB,the average absolute error of the data after completion is 0.023％,the average percentage error is 3.5％,and the root mean square error is 0.021％.Therefore,the proposed method has better de-noising effect and higher data filling performance.
万方数据
维普
