Wu, W., 2018. Ship communication network intrusion signal identification based on Hidden Markov model. In: Liu, Z.L. and Mi, C. (eds.), Advances in Sustainable Port and Ocean Engineering. Journal of Coastal Research, Special Issue No. 83, pp. 868–871. Coconut Creek (Florida), ISSN 0749-0208.

The current intrusion signal recognition method cannot effectively suppress the noise signal interference in the ship communication network environment, and is sensitive to the initial parameters, which results in low recognition accuracy. To address this problem, an identification method based on hidden Markov model is proposed for ship communication network intrusion signal in this paper. The eigenvalue of the collected network signal CSI (channel state information) is denoised and classified. The genetic algorithm is used to optimize the initial parameters of HMM and to solve the problem that the model is sensitive to the initial parameters. The classified signal is trained by hidden Markov model, and the network intrusion signal is identified by the trained HMM. Experimental results show that the maximum recognition accuracy of the proposed method is up to 97.58%, and it has a lower error rate.