At present,the intelligent network environment is complex and dynamic.There are a large number of unknown access requests and there is no predefined reputation information,which increases the difficulty of security early warning.Therefore,a security early warning method for large-scale unknown access sources in intelligent net-works was proposed.At first,the improved wavelet transform method was adopted to perform the noise reduction on packets of unknown access sources and construct automatic immune antibodies,thus extracting packet features of un-known access sources and identifying the anomalies.Then,a fuzzy comprehensive evaluation method was used to eval-uate the security of abnormal access sources,and thus to determine the risk level.Finally,the security early warning was achieved.Simulation results show that the proposed method has good noise reduction effect and high early-warn-ing precision,so it accurately obtains the risk change of unknown access source.
维普
万方数据