ExpSFROA-Based DRN: Exponential Sunflower Rider Optimization Algorithm-Driven Deep Residual Network for the Intrusion Detection in IOT-Based Plant Disease Monitoring

Abstract

Internet of Things (IoT) significantly gained attraction nowadays, as it assists in numerous purposes. Several computerized technologies are employed for monitoring the plant disease in the IoT paradigm, and the major challenging lies in detecting the intrusions while monitoring the plant disease. This paper proposes an Exponential Sun Flower Rider Optimization Algorithm-driven Deep Residual Network (ExpSFROA-based DRN) for achieving effective intrusion detection results in the IoT. The proposed ExpSFROA is devised by incorporating Exponential Sun Flower Optimization (Exponential-SFO) and Rider Optimization Algorithm (ROA). Meanwhile, Exponential-SFO is designed by combining the Exponential Weighted Moving Average (EWMA) and Sunflower Optimization (SFO) algorithm. Here, the information is collected from the simulated IoT nodes based on the Cluster Head (CH), and performs the process of routing in order to predict the leaf disease more effectively. Based on the disease prediction process, the intrusion detection process is achieved by the devised ExpSFROA-based DRN. The performance of the newly developed ExpSFROA-based DRN is evaluated using four metrics such as accuracy, throughput, energy, and True Positive Rate (TPR). The developed method attained better results than the existing methods, such as SecTrust-RPL+DRN, OSEAP+IBFO+DRN, LASeR++DRN with a maximum accuracy of 0.950, higher throughput of 7533350 bps, minimal energy of 49.74 J, and higher TPR of 0.956. The proposed method is used in the agricultural land, which helps the farmers for monitoring the conditions of the fields anywhere and anytime, thereby minimizing the manpower and time.