Algorithm-based adaptive channel allocation and power splitting strategy for SWIPT systems

In practical mobile communication systems, terminal users face challenges such as insufficient computational power, stringent battery life constraints, and scarcity of wireless resources, exacerbated by limitations in communication environments and battery technology. This paper proposes an Elitism-based Genetic Algorithm for Resource Configuration and Splitting (EGA-RCS) strategy for Simultaneous Wireless Information and Power Transfer (SWIPT) systems. The fundamental novelty of this work is the "Joint Configuration-and-Splitting" framework, which couples discrete association decisions with a closed-form derivation of optimal power-splitting boundaries to solve the Mixed-Integer Nonlinear Programming (MINLP) problem efficiently. The simulation is implemented and verified on the MATLAB R2024b platform. Results demonstrate that EGA-RCS achieves a 15% energy harvesting gain over standard GAs and a 12% gain over Particle Swarm Optimization (PSO), showcasing superior stability in handling discrete network constraints.