A learning-based metaheuristic administered positioning model for 3D IoT networks

Abstract

The three-dimensional (3D) IoT networks are among the most contemporary and ubiquitous networks in the real world, which consist of a wide variety of components such as smartphones, smart cars, smart watches, wireless sensors, smart flying objects, etc. The objects’ data are typically gathered, transferred, and processed in these networks for particular intentions. Evidently, the data is worthless without knowing the location of its source. The GPS is the simplest way to locate; however, in environments such as deep forests, underwater networks, underground, multi-story buildings, etc., GPS is not applicable. Likewise, recruiting GPS does not cost- and energy-efficient. Hence, a new positioning system is presented in the current paper to locate objects in 3D IoTs. The Slime Mold Algorithm (SMA) in the proposed method is initially modified and then hybridized with the Equilibrium Optimizer (EO). Moreover, a learning strategy is employed to determine and select the optimal algorithm for each iteration. Furthermore, a neighborhood search strategy is included in the proposed algorithm to enhance search efficiency. Next, the Received Signal Strength Indicator (RSSI) is integrated with the proposed algorithm. Eventually, to assess the proposed algorithm’s proficiency and competency of the contributions, the proposed algorithm is applied to fifteen 3D IoTs, and the results are compared with AEO, AO, EO, MRFO, SMA, WOA, PSO, and SSA statistically and visually. The experimental results portend the superiority of the proposed method over competitor algorithms.