G Sunil Kumar
Engineering
March 2023
A wireless sensor network (WSN) is consist of more number of sensor nodes that lack energy, storage, and computational capability. One of the primary functions of the sensor nodes is data gathering and transmission to the base station (BS). As a result, network longevity becomes the primary criterion for effective design of data collection systems in WSN. Clustering is one of the process which helps to identify the cluster head (CH), but the perfect solution for identifying the CH is still challenging. The selection of energy-efficient cluster head (CH) technique is proposed in this study employing improved manta ray foraging optimization (IMRFO). An improved manta ray foraging method based on Latin hypercube sampling and group learning is presented to address the drawbacks of the manta ray foraging optimization (MRFO) algorithm, such as sluggish speed of convergence and difficulty escaping from the local optimal solution. To begin, the Latin hypercube sampling (LHS) method is used to populate the population. Second, to avoid early convergence, the Levy flying strategy is implemented during the exploratory stage of cyclone foraging. The adaptive t-distribution technique is added prior to the somersault foraging stage to update the population in order to maximize variety and avoid slipping into the local optimal solution. The suggested technique is put to the test by performing several simulations with the base station in various places. The comparison of results with existing methods like GA-Leach, PSO-Leach, SSA-Leach and kmGOA-Leach is made using the matlab simulation and proposed model i.e. IMRFO-LEACH having better performance in terms of all the parameters evaluated.
191- 215
© Airo Journals 2021
