Task Allocation in Wireless Body Area Networks Using Modified Binary Particle Swarm Optimization

Abstract

The proliferation of computation-intensive biomedical applications has made energy efficiency a critical bottleneck in Wireless Body Area Networks (WBANs). This study addresses the under-explored problem of collaborative, intra-WBAN task allocation, in which neighbouring sensors jointly execute portions of a decomposed biomedical workload instead of processing tasks in isolation or offloading them entirely to a hub. The problem is formulated as a constrained multi-objective optimization that jointly minimizes execution time, energy consumption, and inter-sensor energy imbalance under body-centric channel and link-feasibility constraints, and is solved with a Modified Binary Particle Swarm Optimization (MBPSO) algorithm tailored to WBAN dynamics through a V-shaped transfer function, adaptive inertia, controlled mutation, and a body-aware connectivity model. The method is evaluated against deterministic, swarm-based, and evolutionary baselines over 30 independent simulation runs. MBPSO reduces the composite objective cost to approximately 0.336 a 34% improvement over deterministic baselines and 6.8% over a genetic algorithm while achieving near-balanced energy use with an inter-sensor deviation of only 2–3 mJ. To our knowledge, this is among the first frameworks to model collaborative task decomposition within a single WBAN, offering a scalable, energy-aware solution for next-generation deployments.