基于改进神经网络的医院通信安全态势感知方法  

Hospital communication security situation awareness method based on improved neural networks

作　　者：邓从香[1] DENG Congxiang(Beijing Rehabilitation Hospital Affiliated to Capital Medical University,Beijing 100144,China)

机构地区：[1]首都医科大学附属北京康复医院,北京100144

出　　处：《电子设计工程》2025年第1期166-170,175,共6页Electronic Design Engineering

摘　　要：针对医院通信安全态势感知不及时,易导致医院信息系统重要信息受到损害的问题,提出基于改进神经网络的医院通信安全态势感知方法。使用基于小波消噪的通信信号去除噪声并保留关键信息,输入基于改进RBF神经网络的医院通信安全态势感知模型。利用花朵授粉算法完成改进RBF神经网络训练。通过径向基函数对输入数据进行非线性变换,将得到的权值进行加权求和,得到当前通信网络信号的安全态势预测结果。实验结果显示,应用该文方法的医院通信网络异常信息可在1 s内完成感知。Aiming at the problem of untimely awareness of hospital communication security situation,which can easily lead to damage of important information in hospital information systems,a hospital communication security situation awareness method based on improved neural networks is proposed.Using wavelet denoising based communication signals to remove noise and preserve key information,input into a hospital communication security situational awareness model based on an improved RBF neural network.Utilizing the flower pollination algorithm to train an improved RBF neural network.By using radial basis functions to perform nonlinear transformations on input data,the weights obtained are weighted and summed to obtain the predicted security situation of the current communication network signal.The experimental results show that the abnormal information in the hospital communication network using the method proposed in this article can be perceived within 1 second.

关 键 词：改进神经网络 医院通信 安全态势 小波消噪 信号去噪 花朵授粉算法 

分 类 号：TN915.08[电子电信—通信与信息系统]