PoGTR: A Game-Theoretic Blockchain Consensus for SLA Management in Edge-IoT Systems

Abstract

Ensuring SLA-compliant service delivery in decentralized Edge–IoT environments remains challenging due to the lack of integration between SLA enforcement and consensus mechanisms. Existing approaches rely on external monitoring or rule-based enforcement, limiting effectiveness under dynamic and adversarial conditions. We present Proof of Game-Token Redistribution (PoGTR), a blockchain consensus protocol that embeds SLA-related outcomes into consensus decisions through incentive-driven token redistribution. Instead of explicit misbehavior detection, PoGTR regulates validator behavior based on observed service performance, enabling robustness under noisy and variable conditions. We evaluate PoGTR using a discrete-event blockchain simulator with realistic IoT workloads and network variability, comparing it with Avalanche, BECP, and Paxos with priority preemption. Using SLA-aware metrics, results show that PoGTR achieves the highest overall throughput (up to $3. 7 \mathrm{K}$ tx/s under multilayer hashing) while maintaining competitive SLA throughput under strict latency constraints, and the highest SLA throughput under relaxed objectives. Under adversarial scenarios, PoGTR maintains stable performance and consistent SLA satisfaction. Moreover, the sensitivity analysis of the approval-cluster size parameter $(k)$ further shows stable SLA-aware throughput across typical configurations, with limited variability only under extreme settings. These results demonstrate that integrating SLA evaluation into consensus enables effective decentralized SLA enforcement in Edge-IoT systems. To support reproducibility, the implementation and experimental framework are available by request at: https://github.com/TasneemMuneera/PyChainSiM.git.