Multi-objective Evolutionary Algorithms in IoT Wireless Sensor Network Performance Optimization

The IWSNs coverage, connectivity, and energy efficiency are critical elements of the IoT-based Wireless Sensor Networks (IWSNs) that need to be met with network requirements. This study examines the performance of GA-II, FA3, and CA algorithms in the placement of sensors in IWSNs so as to extract the maximum information. The relation between the quantity of nodes and the performance measures of non-dominating solution (NDS), inverted generational relativity, and set coverage is provided in the case of numerical simulations. INSGA-III has real-time applications that are delayed to the minimum, which means that convergence is quick. NSGA-II shows branch maps that are diverse and meet the criteria of solutions, which means that it has a higher number of NDS and a better hypervolume, which shows its efficiency in producing a high-quality set of solutions. MOCSA makes contributions of sincere and objective deliberations. The networks containing 60–70 nodes are highly efficient and reach the competency area of NDS and hypervolume. The high set coverage and hypervolume properties of NSGA-II result in its use in a situation where the quality of solutions is highly valued, whereas the high number of optimal solutions provided by INSGA-III provides feasibility of a wide variety of solutions. These findings highlight the significance of the choice of algorithm used to optimize IWSN, which has a direct effect on the scalability and performance of the network of IoT devices. The studies presented in this paper will offer insights into the further development and management of sensor networks to make their use more effective, as well as increase the availability and reliability of the IoT services.