Real-life modeling for EFficient RESource management in Heterogeneous multi-user systems

National Science Foundation funded project

CNS Division Of Computer and Network Systems


The economic and societal potential of emerging 5G and cyber-physical social systems is vastly greater than what has been realized so far, and major investments are being made worldwide to develop the corresponding science and technology. Though significant achievements have been obtained in the development of various architectures to study such interconnected systems, one barrier to further progress is the lack of appropriate science and technology to conceptualize and design for the deep inter-dependencies among users behaviors, interactions, and decisions. This project, called REFRESH, promotes a novel research agenda, where the proposed resource management framework and the real-life modeling approach go beyond the classical settings. The proposed approach allows for the creation of a user-centric framework that brings significant benefits for (a) the end-user, in terms of reduced power consumption, lower prices and reduced communication and computation overhead; and (b) for the network or service provider, in terms of energy savings and increased number of satisfied customers. The proposed research will influence not only the corresponding scientific directions but will stimulate changes in the emerging economies of scale, creating the basis for societal growth and paradigm shifts. Concrete plans to maximize the broader impacts of the project include educational activities, outreach actions, minority student recruitment, diversity, and data dissemination actions.

In REFRESH a novel approach to resource allocation in wireless networks is considered, based on the following ideas: (a) Instead of maximizing the Quality of Service (QoS), it is argued that better energy-efficiency is achieved by targeting satisfactory QoS levels only. This is treated by a game-theoretical solution concept referred to as satisfaction equilibrium; (b) The theory used so far has not managed to properly address the fact that individuals in real-life do not behave as neutral expected utility maximizers, but they tend to exhibit risk-seeking or loss-aversion behavior under uncertainty. To deal with this, Prospect Theory and the theory of the Tragedy of the Commons are exploited. The realism of the introduced novel resource allocation paradigm is complemented by integrating learning approaches to reduce the impact of communication and computing environment dynamicity and lack of information. Expected findings have the potential to lay the foundations of a solid theoretical and pragmatic resource management framework.

Recent News

  • March 2020, Our paper "Risk-aware Data Offloading in Multi-Server Multi-access Edge Computing Environment" has been accepted in IEEE/ACM Transactions on Networking.
  • March 2020, Our paper "Cognitive Data Offloading in Mobile Edge Computing for Internet of Things" has been accepted in IEEE Access.
  • February 2020, Our paper "Resource Orchestration in Interference-Limited Small Cell Networks: A Contract-Theoretic Approach" has been accepted in the 10th International Conference on NETwork Games, COntrol and OPtimization (NETGCOOP).
  • January 2020, Our papers "Artificial Intelligence Empowered UAVs Data Offloading in Mobile Edge Computing" and "Socio-aware Public Safety Framework Design: A Contract Theory based Approach" are accepted in IEEE ICC 2020.