摘要

We consider a network resource allocation problem that has global information on the whole network such as the explicit forms of all users' utility functions and constraint sets and propose a novel decentralized optimization algorithm for solving it. The algorithm can be implemented through cooperation between individual users and their neighboring users, and it enables each user in a network to determine his or her own optimal resource allocation without using other users' private information such as their utility functions and constraint sets. The main result in this paper is a proof that the algorithm converges to the solution to the network resource allocation problem. We apply the algorithm to concrete network resource allocation problems and provide numerical examples for these problems.

  • 出版日期2013