Fog Task Scheduling Using Quality-Source-Driven Multi-Anchor Synchronized Search Algorithm

Efficient task scheduling in heterogeneous IoT-Fog environments is challenging due to limited fog resources, diverse task demands, and conflicting QoS objectives. This paper proposes ASQS, a Quality-Source-driven Multi-Anchor Synchronized Search algorithm for IoT-Fog task scheduling. ASQS is biomimetically motivated by collective search behaviors in natural systems, where distributed exploration, collective memory, and probabilistic cooperation support an exploration-exploitation balance. Specifically, ASQS constructs quality layers from candidate schedules, extracts representative quality-source anchors, and reuses them through an ACO-inspired probabilistic synchronization mechanism, thereby improving the utilization of high-quality historical search information. FNO-based search and Lévy-flight perturbation are further incorporated to enhance directional guidance and long-range exploration. Experiments on 33 benchmark functions, ablation studies, sensitivity analysis, standard fog scheduling scenarios, and large-scale task-intensive scenarios were conducted to evaluate ASQS. The results show that ASQS achieves competitive optimization accuracy, stable convergence, and superior comprehensive scheduling performance in terms of fitness, makespan, latency, load balance, and constraint handling. In particular, the large-scale experiment with 100 fog nodes and up to 8000 IoT tasks verifies the scalability of ASQS under heavy workload pressure. Statistical tests further confirm the reliability of the observed improvements. These results demonstrate that ASQS is an effective, scalable, and biomimetically motivated optimizer for IoT-Fog task scheduling.