基于k-Means聚类算法的医院隐蔽性网络抗干扰通信OA

Due to the large number of radio equipment and medical devices in hospitals,a large amount of electromagnetic interference is generated,causing serious interference to the communication quality.In order to improve the communication performance of hospital networks,an anti interference communication method for hospital covert networks based on unsupervised learning is proposed.Through preprocessing the interference signal,the time domain moment kurtosis coefficient,frequency domain moment kurtosis coefficient,single frequency energy aggregation degree,and average spectrum flatness coefficient are selected as the characteristic parameters of the interference signal.The unsupervised learning algorithm-k-means clustering algorithm is introduced,the characteristics of the interference signal is extracted,time domain and frequency domain interference signal suppression algorithms is developed,and the interference signal in network communication is suppressed.Experimental results show that the bit error probability of the proposed method reaches a stable state of 2.4％,and the minimum proportion of interference signals is 1.29％,which meets the application requirements of interference signal suppression.