Multi-objective modified emperor penguin optimization for resource allocation in internet of things agriculture applications

Agriculture is dwindling all across the world, which has an impact on ecosystem production capacity. Problems with resource allocation are being investigated as a result of this study. To put it another way, the test's main objective is to create an optimal allocation arrangement by proposing a novel method for overcoming resource allocation issues and challenges. This was made possible by using an intelligent optimisation technique that is both robust and distinctive. There is an urgent need to resolve the issue in the domain so that it may reclaim its vitality and resume its upward trajectory. A revolutionary two-tier approach is designed in this study endeavour to aid farmers in continual field monitoring. The proposed model is divided into two parts: (a) the land-subsystem and (b) the cloud user-subsystem. The agricultural area has been outfitted with several IoT sensor nodes for monitoring "soil PH level, water level, temperature, humidity, moisture, weeding, nutrition, variable spraying, salinity, and rainfall". These sensors are grouped together, and the Cluster Head (CH) connects the clustered nodes to the Base Station (BS). The CH is chosen using the newly proposed Multi-Objective Modified Emperor Penguin Optimization (MEPO) approach, which takes into account many factors such as remaining energy, distance, latency, and QoS. The land-based data is continually stored in the cloud server through the gateway. The cloud sub-system, on the other hand, encompasses the farmers, Physical Machine (PM), and Virtual Machine (VM). The PM assigns a certain VM to process the required work based on the request received from the farmers. Furthermore, the Lionized Golden Eagle optimization (LGEO) based sixfold goal model is projected for optimum task allocation onto the VM, with a focus on power consumption, migration cost, memory utilization, response time and server load and execution time. The proposed model, as a whole, becomes suited for end-to-end farm monitoring.