Research on Security Risk Detection System of Intelligent Internet of Things Terminal

Aiming at the device anomaly problem in the collection of security risk detection system of intelligent Internet of Things (IOT) terminal, taking DTU power terminal equipment as the research object, and combined with the characteristics of side channel attack, a security sensing method of power intelligent Internet of Things terminal based on LSTM was proposed. Where, on the basis of analyzing the principle of side channel attack and combining the characteristics of side channel attack data as time series, the improved LSTM neural network was adopted to detect and monitor security risks. Finally, the above method was applied to the DTU of the IOT device, so as to verify the effectiveness of the system. Experimental results show that compared with the correlation-based regression algorithm Lasso, the proposed method is superior, which can effectively detect and monitor malicious intrusion of power grid terminals. Among them, the false positive rate is less than 2%, and the accuracy rate is more than 90%, which means that the proposed method can guarantee the security of power IOT terminal devices.