An improved white shark optimization algorithm for edge-cloud integrated computation offloading

Wei Li

doi:10.1007/s10586-026-06541-8

Return

An improved white shark optimization algorithm for edge-cloud integrated computation offloading

2026-09-16

PRE

Wei Li

李

李鸿健 (Hongjian Li) *

DOI：10.1007/s10586-026-06541-8

Original

Original request for help

Save

Abstract

En 中文

The rapid proliferation of Internet of Things (IoT) devices and latency-sensitive applications has made efficient task offloading in Mobile Edge Computing (MEC) a critical challenge. While existing metaheuristic algorithms have been widely applied to optimize edge-cloud resource allocation, they frequently suffer from premature convergence, poor initial population diversity, and an inability to scale effectively in highly dynamic, ultra-dense network environments. To bridge these gaps, this paper proposes a novel joint task offloading framework within an edge-cloud collaborative architecture, utilizing an Improved White Shark Optimization (IWSO) algorithm. The proposed IWSO enhances the standard White Shark Optimization (WSO) by integrating three key mechanisms: CIRCLE chaotic mapping for diverse population initialization, an adaptive non-linear weight factor to dynamically balance global exploration and local exploitation, and Fitness-Distance Balance (FDB)-based Cauchy reverse learning to effectively escape local optima. Extensive experimental simulations demonstrate that our approach significantly improves convergence accuracy and speed over existing WSO variants and other standard metaheuristics. Key findings reveal that the IWSO not only achieves a substantially lower overall user-perceived cost by optimally balancing execution delay and energy consumption, but also exhibits exceptional scalability and robust optimization performance in ultra-dense network scenarios containing up to 1000 User Equipments (UEs), successfully overcoming the bottlenecks of traditional scheduling schemes.

Keywords:

Mobile edge computing

Task offloading

Edge-cloud collaboration

White shark optimization