Detection Model Of Innate Immune Mechanism Of Android Malware

Android malware technology is constantly evolving and its malicious behavior is becoming more and more covert,so it is of great importance to study the effective detection methods of android malware.Among the existing detection methods,the static analysis method based on static characteristics has high accuracy,but it is easy to be interfered by such technologies as confusion and string encryption,and cannot detect unknown malware.The dynamic analysis method based on running characteristics has good detection effect,but it is not effective and has poor adaptability to latent malware.In this paper,in-depth analysis of the shortage of the existing android malware detection method and its reason,draw lessons from innate immune mechanism in Natural killer cells(NK)found that latent virus in the body and Dendritic cells(DC)antigen presenting the principle of innate immune mechanism to detect model is put forward to realize adaptive find the unknown,the latent android malware.The work of this paper is as follows:The innate immune mechanism detection model of android malware was constructed.Referring to the behavior principle of NK cells discovering latent virus and DC cell antigen presentation in innate immune mechanism,the artificial NK cell model was applied to the detection field of android malware,and the detection model of innate immune mechanism was constructed to realize anti-latency.Artificial NK cells release stimulants that amplify antigens by detecting traces left by malware.In this paper,Dendritic cell algorithm(DCA)was introduced to work in collaboration with the artificial NK cell model.The stimulating factors released by the artificial NK cells were fused with the danger signals.After collecting antigens and enhanced danger signals,the DC population realized the adaptive recognition of unknown and latent malicious software to improve the accuracy and recall rate.The immune coordination mechanism was introduced to optimize.DCA has the characteristics of fast detection speed and no training,but when the malicious signal significantly decreased,the antigen presentation effect decreased significantly.In this paper,the synergistic mechanism of NK cells and DC cells in innate immunity was studied.After the NK cells complete antigen processing,the activated NK cells fuse the stimulator with the DCA's danger signal to enhance the significance of the malicious signal,while the inhibited NK cells fuse the stimulator with the safety signal to enhance the detection ability of DCA against the latent malicious software.Optimization algorithm,parameter self-tuning.The migration threshold adjustment of DC cells in the model depends on manual experience and needs manual adjustment.In this paper,the threshold interval is automatically initialized according to the logic relationship of the characteristic data,and then the interval endpoint is adjusted at a fixed length to determine the optimal solution of the threshold range,which enhances the adaptability of the algorithm.Verify the feasibility and effectiveness of the model.In this paper,three control experiments were designed with CICInves And Mal2019 and Virtus Total data sets as data sources.The amplification effect of artificial NK cell model on malicious signal was verified.Compared with the model before optimization,the model after optimization does not need to manually adjust the migration threshold parameters,which improves the model's adaptability.Compared with DCA,K-means methods,the improved model can of the unknown,the latent ability of android malware detection ability stronger,with a higher degree of accuracy and recall rate of the experimental results show that this article build mechanism of innate immune detection model improves the adaptability,and can effectively detect unknown latent in the android malware,feasible and effective.