Joint optimization of energy efficiency and operational cost in wireless sensor networks using a Pareto-frontier framework

Abstract

Wireless sensor networks face a fundamental trade-off between energy efficiency and operational cost, where improving one typically degrades the other. Existing approaches treat these objectives in isolation, lacking unified frameworks for systematic trade-off analysis. This paper presents a Pareto-frontier framework that integrates a unified mathematical model coupling node-level energy dynamics with full-lifecycle operational costs, an enhanced non-dominated sorting genetic algorithm II with adaptive parameter control and problem-specific genetic operators, and a decision-support toolkit for knee-point identification and marginal trade-off quantification. Experiments across three scenarios achieve hypervolume of 0.6168, outperforming the weighted-sum sweep by 182.5% and multi-objective evolutionary algorithm based on decomposition by 104.2%, with hypervolume comparable to non-dominated sorting genetic algorithm III within 0.68% while achieving better inverted generational distance (0.0041 vs. 0.0048) and spread metric (0.933 vs. 0.990); knee-point configurations raise energy efficiency by 96.2% over the cost-minimal configuration at an 88% cost increase, delivering the majority of achievable efficiency at moderate expenditure.