MAWA: A Morality-Aware Warning-Audit Mechanism for Free-Riding Deterrence in DTN Pub-Sub Systems

Abstract

Digital Twin Network (DTN) publish/subscribe (pub/sub) systems face a fundamental data-quality free-riding problem. While high-quality data benefits all participants, individual Digital Representatives (DRs) incur private costs for sensing, verification, and timely reporting, creating incentives to publish low-quality data to save resources. Existing audit, warning, and reputation mechanisms address these issues separately, but not their joint effect in budget-constrained DTN Pub/Sub governance. This paper proposes a Morality-Aware Warning-Audit (MAWA) mechanism to deter strategic undercontribution under noisy observations and limited audit budgets. MAWA is designed as a platform-side governance mechanism that integrates event-level warnings and audit commitments with cycle-level accountability and morality updates. We model the interaction between the platform and active publishers as a two-timescale Stackelberg audit game, derive the follower's bestresponse structure and local platform audit rules, and develop an online implementation with adaptive budget scarcity. Simulation results show that MAWA improves system-quality preservation and repeat-offender control compared with budget-constrained baselines. Ablation, model-mismatch, and scalability analyses further show that MAWA remains effective under limited audit resources, imperfect utility knowledge, nonlinear quality dynamics, and large event workloads.