Distributed intelligence for dynamic resource allocation in 5G networks

This video program is a part of the Premium packages:

Distributed intelligence for dynamic resource allocation in 5G networks


  • IEEE MemberUS $1.00
  • Society MemberUS $0.00
  • IEEE Student MemberUS $1.00
  • Non-IEEE MemberUS $2.00
Purchase

  • IEEE MemberUS $50.00
  • Society MemberUS $0.00
  • IEEE Student MemberUS $25.00
  • Non-IEEE MemberUS $100.00
Purchase

Distributed intelligence for dynamic resource allocation in 5G networks

0 views
  • Share

In 5G networks, we expect resource allocation problems to be: (a) localized, because of small cells; (b) more in number, because of denser cells and higher traffic; and (c) heterogeneous, because of the variety of applications and devices on such networks.  Centralized resource allocation is a poor fit for such scenarios.  We argue in favor of distributed solutions for such allocation problems.  Moreover, we want these solutions to be automated in order to scale up to the demands of 5G networks.  We propose market-based resource allocation solutions based on economic models of utility and show how they are distributed and lead to efficient allocation of resources while maintaining the quality of service as perceived by both users and network operators.  The automation of such market-based solutions is achieved by machine learning models to predict resource demand by each user such that the predicted demand can be satisfied by resources acquired by the user through a market transaction.  We illustrate such intelligent market-based resource allocation for the case of a user supported by a cluster of base stations under coordinated multi-point (CoMP) transmission.

Distributed intelligence for dynamic resource allocation in 5G networks

Sayandev Mukherjee and Bernardo Huberman, CableLabs

Advertisment

Advertisment