Decentralized Resource Allocation in Networks with Strategic Agents: An Implementation Theory Approach

In this thesis we present an implementation theory approach to
decentralized resource allocation problems with strategic agents in communication networks. For wired networks we study the unicast and multi-rate multicast service provisioning problem. For wireless networks we study the problem of power allocation and spectrum sharing where each user's transmissions create interference to all (or subset of) network users, and each user has only partial information about the network.

We investigate these problems under the implementation theory scenario where agents/users are strategic, self utility maximizing. We present key concepts and ideas from implementation theory that are relevant to the problems.

We formulate the unicast service provisioning problem as a market allocation problem, the power allocation and spectrum sharing problem as a public goods al- location problem and, the multi-rate multicast service provisioning problem as the combination of a market and a public goods allocation problems.

For each problem, we develop a game form that (i) implements in Nash
equilibria the optimal allocations of the corresponding centralized
problem; (ii) is individually rational; and (iii) results in budget balance at all Nash equilibria and feasible off equilibria.