Secure Trust Management Scheme over the Detection of ON/OFF Attacks to Predict an Efficient Crop Yield Production in Wireless Sensor Network

Agriculture is the process of growing plants and animals. Grains, vegetables, fruits, and other agricultural goods are divided into different categories. Farmers can increase yield while using the less number of resources, such as water and fertilizer by using smart farming, this is also known as precision farming. Farmers can use sensor nodes in the field to understand their crops on a microscopic size, which saves resources and decreases environmental impact. Stem borers, bollworms, thrips, and other pests destroy crops. For instance, Stem borer is an attack that causes plant leaves to wilt, the base of the plant to exhibit holes, and the color of the plant to shift from green to orange-yellow. If the stem borer spreads to other crops, the plant’s growth tips, or the youngest leaves, will be killed. These types of issues are identified in this paper employing sensor nodes. Using the WSN, the affected area of the crop is first determined. The remaining parts of the field will be protected from attacks by detecting the malicious part. Second, the discovered afflicted sections of the crop are analyzed by using the Machine Learning (ML) algorithm in order to prove that the crop has been attacked by some disease. Finally, both the outcomes are examined to determine the disease spread. This research provides On-Off attack mitigation for wireless sensor networks’ trust systems, which assists in the prevention of various crop diseases and attacks. As a result, a variety of sensing technologies are developed. The Machine Learning (ML) algorithm used here is Random Forest (RF). The WSN On-Off attack plays a significant role as it investigates the challenges in Trust Management (TM) procedures.