A Hybrid Autonomous Negotiation Framework for Trust-Aware Distributed Resource Allocation in Cloud–Edge–IoT Multi-Agent Systems
ID：155
View protection：Participant Only
Updated time：2026-07-27 15:21:28
Views：7
Online
Abstract
Abstract---With the advent of cloud, edge and IoT systems, resource allocation has become a continuous, adversarial and distributed decision problem, not just a one-time provisioning activity. Though there are several communities (mechanism design, trust engineering, reinforcement learning and distributed systems) trying to do autonomous negotiation in the field with no central arbiter for reaching a binding resource-sharing agreement between two or more self-interested software agents. This paper has surveyed principles, strategies and evaluation practices of autonomous negotiation of distributed resources and developed a layered conceptual architecture to classify and organize the literature and a comparative framework to score existing approaches, and a proposed hybrid design, according to different measures such as the negotiation success rate, resource utilization, communication overhead, convergence time, scalability and fairness in autonomous negotiation of distributed resources. The paper has been structured around the series of decisions a system architect has to take while implementing a negotiating agent and is not written chronological as done in past surveys, but has ended with the open research challenges and research agenda for the future.
Keywords
Autonomous Negotiation, Distributed Resource Allocation, Multi-Agent Systems, Edge Computing, Cloud Computing, Internet of Things (IoT), Reinforcement Learning, Blockchain, Trust Management, Resource Sharing, Fairness, Distributed Systems.
Post comments