An online task offloading method based on improved starfish optimization and blockchain

Vehicular Edge Computing (VEC) is a key enabler of low-latency intelligent transportation applications. However, designing effective task offloading strategies in VEC faces challenges such as resource variability, dynamic network topology, and data isolation among distributed nodes. To address these issues, this paper proposes an Online Intelligent Blockchain-Enhanced Task Offloading Optimization System (OIBTO). The system employs a lightweight Proof-of-Authority (PoA) consensus within a two-tier architecture consisting of a vehicle layer and an edge layer. Edge nodes act as validators, jointly maintaining a distributed ledger to enable secure and efficient sharing of task dependencies and offloading decisions, effectively eliminating data silos. Additionally, we propose an Improved Starfish Optimization Algorithm (ISFOA) that utilizes Tent chaotic mapping and genetic mutation to optimize offloading decisions and task partitioning ratios, aiming to minimize a priority-weighted combination of latency and energy consumption. Theoretical analysis confirms the convergence of ISFOA. Simulation results show that the proposed framework improves average task latency and energy consumption by approximately 10% compared to state-of-the-art algorithms, demonstrating its effectiveness, security, and superiority in dynamic vehicular environments.